a ~ /f'f JUBRARr GARDEN n THE BRITISH RUST FUNGI (UREDINALES) CAMBRIDGE UNIVERSITY PRESS ?loll&on: FETTER LANE, E.C. C. F. CLAY, Manager etJinbuvgli: 100, PRINCES STREET Hontion: WILLIAM WESLEY AND SON. 28, ESSEX STREET, STRAND Berlin: A. ASHER AND CO iLr ip?ig : P. A. BROCKHA1 - Ctxo Ijorl;: G. V. PUTNAM'S SONS Uombao anb ("Calcutta: MACMILLAN AND CO., Ltd. All rights reserved THE BRITISH RUST FUNGI (UREDINALES) THEIR BIOLOGY AND CLASSIFICATION BY W. B. GROVE, MA. LIBR&BTG NEW YOBK -rr^NICAL Cambridge : at the University Press l9!3 ©amfartDgc : PRINTED BY JOHX CLAY, M.A. AT THE UNIVERSITY PRESS GARDEN PREFACE TT is now twenty-four years since the publication of Plowright's -■- " Monograph of the Uredinere and Ustilaginese," and during that long period very great progress has been made in elucidating the biology of the former group. To Plowright will ever belong the honour of being one of the pioneers in this important work, especially in that branch which is the most fascinating and has the greatest number of secrets to unfold — viz. Heteroecism. But since his time numerous investigators have followed in his footsteps, while others have taken new ground and largely increased our knowledge, so that the picture which he presented of the biology and classification of the Rusts has now become, in certain directions, very incomplete and misleading. The descriptions in the present volume are naturally based upon those in the " Monographia Uredinearum " of the brothers Sydow, so far as that monumental work has been published. Those of all the species of which British specimens could be procured have been carefully revised, and there is hardly one of them that has not been added to or amended. Fischer's "Uredineen der Schweiz," and McAlpine's "Rusts of Australia" have also been found extremely useful. No attempt has been made to give a full synonymy, but merely so much as was required to show the origin and authority of the name used, and to include all the references to the various species contained in the works consulted, especially those of Cooke and Plowright. Dates have been added occasionally, but only for special reasons or if the nomenclature adopted differs from that in the a 3 vi PREFACE " Monographia," where a complete synonymy, with dates, lies at the disposal of those who are i 1 1 > < t< -i . •« I in thai matter. A plan has been adopted, in a few cases, of grouping some forms thai are closely allied under a common name; see Puccinia dispersa and /'. sessilis. With more knowledge (or more courage) this process might have been carried much farther, and it is hdicved that in this device will be found the first line of defence of sane systematists against, the excess ive multiplication of "species" by "biological" nomenclators. It cannot be justifiable to use the same word in the same branch of science to denote two widely diverse grades of evolution. Physiological, unaccompanied by morphological, distinctions should never be allowed to constitute a difference of species, unless it be as a temporary measure in cases which have not been investigated. In selecting the name for each species, the principle of priority has been followed, subject to two conditions — (1) names given to varieties need not be adopted (International Rules, 1905, Art. 49), and (2) names given to imperfect states are not to be preferred, but the earliest name given to the perfect (in this case, the teleutospore) stage (Brussels Congress, 1910); sometimes, however, the name Uredo may have included the perfect stage, as in some species of Uromyces (U. Scirpi, etc.). Since secidiospores are almost always " rounded-polygonal," their shape is not mentioned unless it deviates from this form. In the systematic part, all the spores are drawn to the same magnification (600 times), except where indication to the contrary is given. The drawings are all original and from British specimens, unless a different source is stated. When several similar spores are outlined, the surface-sculpture is not always indicated upon every one of them. My thanks are especially due to Professor G. S. West, by whose advice this work was undertaken and by whose assistance it has been chiefly carried through. The 'Plowright' Herbarium of Fungi, which is in the possession of the University, has been PREFACE vii of considerable use, and the Principal (Sir Oliver Lodge) most kindly obtained for me a grant towards the cost of preparing the illustrations. By the courtesy of Dr A. B. Rendle, Keeper of the Botanical Department of the British Museum, and that of the Director of Kew Gardens, the Herbaria at those insti- tutions have been consulted and much useful information lias been derived therefrom. Thanks are also due, and are hereby gratefully tendered, to Herr H. Sydow (of Berlin) and Mr J. Ramsbottom (of the British Museum) who have both given unstinted help, to Mr T. B. Roe (a very successful collector of specimens), Professor A. H. R. Buller, Mr A. D. Cotton Mr H. J. Wheldon, Mr C. Crossland, Mr J. Adams, Sir Frederick Moore, and many others in a smaller degree. W. B. GROVE. The Botanical Laboratory, University op Birmingham. July, 1913. CONTENTS PAGE Preface \ Introduction \ i GENERAL PART CHAPTER I The Life-History of Puccinia Caricis, the Nettle and Sedge Bust 1 CHAPTER II The Sexuality of the Uredinales — The nature of the Spermatia — Nuclear division in the Uredinales — The Alternation of Generations 17 CHAPTER III Spore-forms of the Uredinales — Grouping according to Spore- forms 30 CHAPTER IV Life-Histories of other Uredinales — Puccinia graminis —P. Poarum — P. Malvacearum — Otmnospora ngium CLA VAUI.K- forme — Endophyllvm Sempervivi — Cronartium RIBICOLA Melampsora pinitorqua — Calyptospora Goeppertiaka . . 41 CHAPTER V Specialisation — Immunity 62 CHAPTER VI Classification — Phylogeny .... Note on Manipulation ■ ■ CONTENTS SYSTEMATIC PART PAGE Pucciniaceee Qromyces PUCCINIA L28 Triphragmium 287 Phragmidii m 289 K DEHNEOLA . 299 \ ENODOI IM S . 31 >2 ( rYMNOSPOB Wi.ir.M 304 Cronartiaceae 1 11 i: , 3< 'M\ \A . 310 ( Iron urnr.M . 313 Coleosporiaceae COLKosroKir.M OCHROPSORA . 319 329 Zaghoi AMA . :v.',\ Endophyllaceae Exdoi'hyi.u'u 333 Melampsoraceae m elampsora . Melampsoridium . m elampsore1 1 \ pucciniastrum 'I'll 10 OPSOB \ . Calyptospora Hyalopsora . MlLESINA I'm DINOP8IS . 336 358 360 364 368 372 373 376 37!) Appendix Hemileia ( Jhrysomyxa . Uredo . iEdDIUM 381 384 385 386 I EOMA . 388 Excluded Sn .... 389 Glossary 391 Bibliography . 393 Indexes . 398 INTRODUCTION The Uredinales form a group of Fungi which is also spoken of as the Uredinese or the Rusts. An accurate acquaintance with their nature is of great importance to the gardener, the forester, or the agriculturist, on account of the enormous loss which is caused by them every year and which can, at least in part, be avoided by a fuller knowledge. All the species are parasitic, growing upon or in a living plant, which is called the host. The majority of the species of Uredinales have more than one stage of growth, distinguished by the form and arrangement of the spores which they produce ; the number of distinct kinds of spores which a single species can possess varies from one to five. If the various spore-forms are all borne upon one host, the species is called autoecious. But it is a remarkable fact that a large number of the Uredinales pass their existence alternately upon two hosts, certain of the spore-forms being always produced upon the one, and the remainder upon the other. Such species are called heteroeciotis or metcecious. Many of those which grow upon grasses or sedges are probably hetercecious, though this has not been shown in every case, and there are a few proved exceptions. In order to convey a notion of the complex nature of the Uredinales, one of the hetercecious species will be taken as the type, and its various stages will be described. ADDITIONS AXU CORRECTIONS p. 24. Fig. 225 should be Fig. 226. p. 28. Fig. 211 should be Fig. 242. p. 139. Puccinia tinctoriicoUt Magn. (Ester. Bot. Zeitschr. p. 491 (l'.i02) should be added as a synonym to P. tinctoriae. p. 150. P. Leontodontis. 1 have since found this species on L. hirtus (Thrincia hirta). p. 238. P. oblongata has also been found on Luzula silvatica. pp. 290 — 7. It is stated that Phragmidium violaceum infests most of the subspecies of Rubus fruticostis, except those belonging to the group "Corylifolii,"' while Ph. Rubi is conhned to that grouj) and the allied llubus ausiits. CHAPTER I LIFE-HISTORY OF PUCCINIA CARIGIS, THE NETTLE AM) SEDGE RUST Pnccinia Caricis has two of its stages, the spermogonial and secidial, on the Nettle ( Urtica dioica and other species), and two others, the uredo- and teleutospore-stages, on various species of Carex, especially C. paludosa. The first appearance on the nettle is in the spring, about the end of April or the beginning of May, when small swollen yellowish spots can be seen on the upper surface of the nettle-leaves. These spots are round and convex above, sunken beneath, and about 3 — 4 mm. in diameter ; soon they turn orange on the upper surface, owTing to the development thereon of the spermogones, small flagon-shaped bodies walled in by a large number of slender orange hyphge and filled with many hundreds of minute spore-like cells, the spermatid, which are orange in mass, though singly they appear colourless (Fig. 1). Fig. 1. P. Caricis. Section of leaf of Nettle, showing the hypertrophy produced by the mycelium of the aecidium-stage ; a spermogone, on the upper side of the leaf, and two secidia, one closed, on the under side. The upper face of the leaf is turned downwards. x 60. G. U. SI'EK.M()(;oXE.S The mycelium in the leaf is strictly localised, forming little knotted masses (plectenchynil A but the mycelium produced is very scanty. Their function will be discussed in a later paragraph, but it may be mentioned here that all attempts to produce infection by means of them have uniformly failed. The iEcmiUM. Before the spermogones have completed their development, similar but larger conglomerations of hyphse arise on the lower surface of the leaf, a little way below the epidermis, which they raise up into a rounded dome. These masses enclose a number of erect hyphse of two kinds — an outer series of parallel closely- joined colourless hyphae, forming the peridium, consisting of more or less hexagonal cells, which meet above and roof over the dome- shaped cavity ; and an interior series which remain shorter and give off from their upper end parallel chains of spores called aecidiospores, which fill the whole of the enclosed space. Each new spore is produced beneath the older ones, which are thereby pushed gradually up. Finally, this struc- ture, which is called an aecidium, ruptures the epidermis, forces its way between the cells, the peridium bursts at the summit, the edges become revolute, and there is formed a white cup-shaped hollow, its floor covered with erect chains of orange spores (see Fig. 1). These spores have a rather thin, colourless, finely warted cell-wall, and are filled with rich bright-orange granular and oily contents. The mycelium which forms the a?cidia is continuous with that which bears the spermogones, and its cells are uninucleate, but the spores themselves are binucleate. The origin and meaning of this change will appear afterwards. 1—2 Fig. 3. Leaf of Nettle (under side) showing the secidia of P. Curie is on the lamina and petiole, x i. 4 .KCIDI \ The oldest spores being a1 the top, bhey separate as soon .-is mature, and arc blown away by the wind; fresh spores are produced for a time from below. Since these cups are in dense clusters over the whole of the affected spot, they are known as ■•Cluster-cups." The mycelium in the leaf continues to develop, and the spot enlarges and ceases to be round. But the swelling of' the tissues within which the mycelium is living can be carried to such an extern as to distort and curl the leaf, much in the same way that /'J.mascus deformans (the Peach Leaf-curl) affects the Peach leaves. The epidermis above and around the spot, also, often becomes coloured red or purple by an anthocyan- derivative, both these effects being a response by the leaf-cells to the stimulus of the parasitic growth, and perhaps part of an effort to throw off or checkmate the invader. The mycelium, moreover, is not confined to the leaves ; it may originate in the petiole, in the stipules and in the stem. In the latter case it causes notable curvature and distortion ; an instance is recorded where the curved gall-like mass, formed on the stem, measured as much as 10 cm. in length, and similar but smaller growths may frequently be met with. Here should be noticed the close parallelism, up to a certain point, in the formation of the spermogones and the secidia — they have similar shapes, they are both enclosed by a layer of sterile hyphae, they arise in ( usually) quick succession on the Mime mycelium, they give off basipetal chains of spores from their base. On the other hand, besides the difference in the nuclei, the secidiospores differ from the spermatia in their larger size, in possessing a large store of reserve food, in their capacity for germination, and for producing infection in another plant. The .Ecidiospores. The mode in which the secidiospores are produced is as follows : — The upper cell of the sporiferous hypha usually divides into two, an upper sterile cell, and a lower fertile or basal cell, each with one nucleus. The upper cell disintegrates and perishes; the lower fertile cells conjugate with one another in pairs, the process consisting in the formation of a small hole .HCIDIOSPORES 5 in the dividing wall which afterwai'ds enlarges until at length hardly any trace of the wall is left. The two cytoplasms thus form one mass, but the nuclei arrange themselves more or less side by side without fusing or in most cases even touching. This process must be regarded as an act of fertilisation or rather as a substitute for such an act1. The double cell is called a fusion-cell. The two adjacent nuclei are said to be " paired,-' and to- gether constitute a synkaryon or dikaryon. Occasionally, in some species of Puccinia and other genera, these fusion-cells contain three nuclei, probably by a double fusion. Such cells produce trinucleate spores, but the fate of these is not known. The paired nuclei of the fusion-cell then divide side by side and simultaneously — a process called conjugate division — and a wall is formed between the two pairs. The wall is formed in such a way that the two nuclei in the same cell are not sister- nuclei (see Figs. 19, 20). The fact that in conjugate division the paired nuclei divide so that the two are usually in the same stage of mitosis at the 1 The stages connected with the sexual process in general ai-e three in number — (1) the association of two (almost always non-sister) nuclei in the same cell, (2) the fusion of the two nuclei, preparatory to (3) the reduction in the number of chromosomes, or at any rate in the amount of chromatin, to the previous ordinary vegetative condition. These three stages may or may not follow closely upon one another. The first and second, or the second and third, or the third and first, may be separated by an intervening series of cell- divisions. Which is to be regarded as the actual fertilisation ? In all proba- bility the first. The nucleus is the director, the cytoplasmic mass is the ivork- people of the cell-factory. The presence of the two directors is the essential fact. The fusion is of less importance ; it is often delayed for a long period, and in certain cases it is known that, even when fusion has taken place, the chromosomes still retain their individuality for a considerable time. There is reason to believe that on the first introduction of fertilisation these three stages followed immediately (i.e. without intervening cell-divisions) after one another, the series of vegetative divisions being intercalated between (3) and (1), as in Coleochaete (Allen, 1905) ami Spirogyra (Trondle, 1907; Karsten, 1909). In the majority of the higher plants, stages (1) and (2) occur without much or any interval, but a long series of divisions (the sporophyte) is intercalated between (2) and (3). In most of the Uredinales, the chief series of vegetative divisions is intercalated between (1) and (2), and (3) follows immediately after (2). In certain Algae (Griffithsia, Dictyota) the intercalation takes place on an extensive scale, both between (2) and (3) and between (3) and (1). 6 aJCJDIOSPORES same time, shows thai they mutually influence each other, and implies thai all the cell-processes which :_,r" on are likewise under their common control. The upper cell, cul off from the fusion-cell, is the aecidio- spore- mother-cell ; the lower grows a little longer and then divides again in the same way, and thus a vertical series of aecidiospore-mother-cells is formed, the oldest at the to]>. Each of the aecidio- spon-mol her-cells, as soon as produced, nits off, by conjugate division, a small (•"11 below, called the intercalary cell; this soon becomes disorganised and disappears, while the other portion becomes the aecidiospore. Thus the chain at first consists of alternations of a?cidiospores and abortive spores, both containing two nuclei (Fig. 4<) ; the function of the latter may be, by their disintegration, to enable the ajcidiospores to separate more easily from one another, and thus to aid dis- persion by the wind. Thr ficridiiim is composed of cells homologous with the spore-mother- cells. They arise from basal cells con- taining two nuclei in exactly the same way ; occasionally even an intercalary cell is cut off, but it does not dis- integrate. All of them thicken their walls and remain in close contact with one another; ultimately their contents disappear ( Fi-. '>). The central part of the dome-shaped peridium is composed of the terminal cells of the central spore-bearing hyphse ; they are probably less closely connected with one another, and the peridium ultimately bursts at that point. The spores, if placed in a damp atmosphere, germinate readily when mature and fresh. Only those spores which are at the top of the chains and can be shaken out easily by Fig. 4. /'. Caricis. Chain of young recidiospores, x 500. ", fusion-tissoe; //, basal (fusion) cell, with con- jugate nuclei ; c, aecidio- Bpore-mother-cell ; d. in- tercalary cell ; e, yonng a eidiospore. GERMINATION OF vECIDIOSPORES 7 tapping are mature enough to germinate, and even they, owing to their thin walls, may lose this power in a few days according to circumstances. Especially can they be killed by rapid Fig. 5. P. Caricis. a, three cells of the peridium, on Nettle ; b, a cell in optical section; c, the same in surface-view ; d, two acidiospores. x 600. drying. Instances are known, however, where some of them, kept in a cool place, retained their capacity for germination about seventy days, though most of them were dead after eighty days. Each spore has a number of germ-pores ; in Puccinia Caricis the number is about five or six ; in other species of Uredinales the number varies from two to eight. These pores, which are scarcely visible until germination begins, are thin places in the inner layers of the outer wall, the whole cell bearing a close resemblance to many kinds of pollen-grains (microspores). (Fig. 6.) It is worthy of notice that in the spores of Endophyllum, and others of the less advanced type, there are no real germ- pores ; the germ-tube merely forces its way out at the first place that gives way. From this state of things there is a gradual transition from numerous to few germ-pores ; in the highest types of all (Uromyces and Puccinia teleutospores), each cell has one and only one well-defined germ-pore. Fig. 6. vEcidiospore of P. Caricis, germinating in water, x 180. s GERMINATION OF .-KCIDlo.SPoKEs The germ-tube of the secidiospoiv of /'. ('ariri.s can grow bo a Length ten or twenty times greater than the diameter of the spores, and often executes, at Leasl in water, a number ol spiral turns (Fig. 6); it remains nearly of the same diameter t hroughout, or may bear shor( stump} branches here and there. The granular contents of the spore travel along the tube with its growth, remaining always towards the distal end. But unless the germinating spore has fallen upon its proper habitat, the Leaf of a Carex, its attempt at growth comes to an end when its reserve-material is exhausted. If, however, the spore has fallen upon a Carex, its germ- tube travels over the surface until it reaches a stoma, through the pore of which it enters the respiratory chamber, forms a swelling just inside as a kind of hold-fast, and then begins to branch and traverse the intercellular spaces, occasionally sending an haustorium into the mesophyll-cells (Fig. 7). The cause of its entry is pro- bably the search for water-vapour, since the germ-tube of a Uredine is found (De Bary, 1863; Gibson, 1904) to enter the stomata as freely on another leaf as on one of its proper host-plant, and also to pass through a hole, comparable in size to stomata, in a thin india-rubber membrane which separated it from air saturated with water- vapour (Balls, 1005). But its further growth is influenced by chemotaxis of a more complicated nature : unless the right kind of stimulus is furnished by its host, it cannot form effective haustoria, de- velopment is poor and abnormal, and death soon ensues (Gibson, 1904). The resistance of the host to the parasite, shown perhaps by the secretion of destructive enzymes, has also to be considered. Once inside the stomatal chamber, however, the fungus is largely protected from outside influences, such as desiccation : this preservative habit has no doubt contributed much to the wide-spread prevalence of the Uredinales. Fig. 7. Haustorium of /'. ( mil- is in cliloro- phvllose cell; m, my- celium ; //, hausto- rium; n, nucleus of cell, x 500. UREDOSPOKES !» The Uredospores. The germ-tube soon forms a more or less extensive m\ celium, which may penetrate the greater pari of the leaf of the Garex, but in many species of Puccinia is strictly localised to a small defined spot. All its cells are binucleate, like the spore from which it originated. The cells of the mycelium, in every stage, send haustoria into the cells of the host : when an haustorium arises from a binucleate mycelium, it is itself Fig. 8. Section of leaf of Garex. paludosa, with a sorus of uredospores of P. Caricis ; a, upper epidermis ; 6, a vascular bundle. Most of the pedicels have lost their spores. x 180. likewise binucleate. After a few days this mycelium begins to form the third kind of spore — the uredospore. A knot of hyphse is formed just beneath the epidermis ; some of the branches turn upwards and form a regular layer parallel to the surface — the spore- bed (Fig. 8). The upper rounded cell of each hypha is divided into two daughter-cells, the lower of which is developed into a stalk, the upper becomes the uredospore (Fig. 9). The spore is oval or roundish : when mature it is enclosed in a double cell- wall, the outside being cutinised and provided with spine-like projections, some- what like those of the secidiospore, only more pointed. In the inner layers of the exospore there are usually three (rarely Fig. 9. P. Caricis. De- veloping uredospores (1, 2, 3, 4, show stages of growth ; 5 is a pe- dicel from which the spore lias vanished), x 500. 10 rKKlXlSPOKKS : Fig. 10. Leaf of Carex pendula, with uredo- and teleuto-sori, slightly en- larged. four) germ-pores ; in tact, tin- uredo- resembles the aecidiospore in character, and musl b»- <-<>nsidered as homologous with it— the stalk-cell corresponding to the intercalary cell of the latter. But they differ considerably in the fad that the uredospore is always produced singly, uot in chains. (This is not true, however, of all the Uredinales.) The membrane of the uredosjxiiv is nearly colourless, but it encloses a bright orange granular and oily mass, with two nuclei. Every cluster of uredospores produced on the same spore-bed is called a sorus ; it is surrounded by the lacinise of the epi- dermis, which is more or less torn or split by the enlarging mass. In many cases, several sori become confluent and form a larger pustule (Fig. 10). In other species of Uredinales the uredospores have coloured membranes or possess a larger or smaller number of germ- pores. Moreover the distribution of these pores over the surface is characteristic for each species : they may be placed equatorially, as they are in P. Garicis (Fig. 11), or towards the poles, or scattered over the surface with regularity or without any order. A uredospore may be very easily detached from its pedicel, and conveyed (chiefly by wind, though sometimes by insects) to another leaf of Carex, on which it germinates, the germ-tube enters a stoma, produces a fresh crop of mycelium and another sorus of uredospores ; this process can be repeated indefinitely. The my- celium can also grow up and down the leaf, producing fresh sori in its course ; for this reason the sori are usually arranged in linear series, owing to the parallel venation of the Carex-leaf. The germ-tubes of the uredospores are often curled or branched like those of the ;ecidiospores, and the germination is of the same character in Fig. 11. A uredospore which hasgerrninated, showing the three germ -pore s. x 600. TELEUTOSPORES 11 both (Fig. 12). The uredospore retains its capacity for germination for a longer time, even for more than three months ; in fact, in certain foreign species, some of these spores acquire a thicker wall which enables them to act as a kind of resting-spore — these are called amphi- spores, but they are not formed by P. Caricis. It is found, generally, that if the spores of the Uredinales are dried gradually, they retain their power of germination for a longer time and in a better degree than if dried quickly or not dried at all. Most likely a slow drying enables them to mature more perfectly. Fig. 12. Two uredospores of P. Caricis, both germi- nating in distilled water. xl80. The Teleutospores. After a time, probably in response to the weather or other change of environment, the mycelium which has hitherto given rise only to uredospores begins to produce, at first in the same, afterwards in separate sori, the fourth kind of spore — the teleutospore. In the genus Puccinia this is almost always a compound body, formed of two superposed cells ; each cell is really a spore, and is capable of independent germination. In many species of the genus the teleutospores readily break apart at the septum, e.g. in Puccinia fusca and P. Pruni- spinosae, and the lower half may be, and has been, mistaken for a uredospore. Those of P. Caricis do not easily break apart until they are old and dead. The teleutospores are formed on a pedicel, much in the same way as the uredospores, except that the uppermost cell is again divided, but apart from that they differ widely in their character. They have a thick dark-brown exospore, covered with a chitinous cuticle ; in this species the exospore I: TELEUTOSPORES i- much thicker a1 the apex than olsou here ( Fig. |:J). While tin' spores are standing in a densely crowded sinus, the apex is the part most exposed to tin \\eatlnT.aml tlicivti)!'.- mosl needing protection. There is a thin endospore t" each cell: the contents aiv granular and at first oily : there is a largo and conspicuous nucleus in the centre of each. This nucleus, in its resting stage, is almost homogeneous except for its nucleolus, and was mistaken by the older observers for a " vacuole" Since the mycelium from which the teleuto- spores, as well as the uredospores, were formed contains paired nuclei, the cells of the teleuto- spore were at first in the same condition. When its wall, however, begins to thicken, i.e. when it is becoming mature, the conjugate nuclei unite, and form one large fusion- nucleus (Fig. 14). The two fusing nuclei, after the v< i\ Fig. 13. Teleuto- spore of /'. Caricis. x 600. Fig. 14. Formation of teleutospores of P. Falcariae (after Dittschlag) ; a, the spore-bearing hypha; b, the same divided into pedicel and spore-cell; c, the spore divided into two cells; d, a young teleuto- spore ; e, the same after the fusion of the nuclei, x about 800. numerous conjugate divisions during the long period of growth from the formation of the fusion-cell of the secidium, would be related, as it were, like very distant cousins, especially since the nuclear divisions during this period, though indirect, appear to show a very simplified form of mitosis, tending rather to be of the nature of amitosis. The fusion, as already intimated, is not to be considered as the act of fertilisation, but merely as a necessary preliminary to chromatin-recluction. MESOSPORES 13 It is in the germination of the teleutospore, presently to be described, that its most distinctive feature is to be found. The chief function of teleutospores is to act as resting-spores, and in the majority of cases they will not germinate until they have passed through a period of quiescence ; in the present instance this period is the winter, but it is not necessarily always so. The resting-spore is primarily a device to tide over an unfavourable period — whether of food-supply, moisture, temperature, or resistance of host — without regard to season. Some species, however, have teleutospores which can germinate immediately, as in P. Malvacearum ; those teleutospores usually have thin walls. P. Malvacearum is sometimes supposed to hibernate by a perennial mycelium, but there is reason to believe that in most cases infection each year proceeds from over-wintered teleutospores. Most of the species which have these thin-walled spores also produce some with thicker walls, which act as resting spores in the ordinary way. Besides the two-celled teleutospores, several species of Puccinia also produce similar spores with only one cell — these are called mesospores. A mesospore can occasionally be found in many Puccinias, even in P. Caricis (Fig. 15), but in others they are abundant, e.g. in P. Porri, where careful search is often required before a two-celled spore can be detected. Mesospores arise merely by the omission of the last nuclear division ; they are exactly of the same nature as the two-celled teleutospores F- ,. . and germinate in the same way. By this spore of P. Car- means they can be distinguished from the amphispores previously mentioned, but not of course from the teleutospores of Uromyces. In fact authors have described some species which produce them as Uromyces, overlooking the rarer two-celled spores that occur with them. See remarks under Puccinia Porri and Uromyces ambiguus. 14 GERMINATION OF TELEUTosi'ultKs Ckhmixation of the Teleutospoee. We ii"\\ approach the consideratioD of a process which has been in the past much discussed, and upon the right inter- pretation of which the whole question of the systemal ic p« >>ition of the l/redinales depends. Kadi cell of tin- teleutospore of P. Caricis has one germ-pore, though some genera allied to Puccinia have toleutospores with more than one germ-pore to each cell, e.g. Phragmidium, Uropyads, Calliospora. The germ- pore of the upper cell is in the thickening at the summit, that of the lower cell is lateral and just beneath the septum. Bach of these pores is a canal passing through the cell-wall, and covered only by the cuticle. Through these pores the germ-tube passes, first appearing as a roundish swelling, the protoplasm being surrounded by the thin endospore. This then elongates, the nucleus squeezes through the relatively narrow pore and enters the tube where it divides twice, and forms four superim- posed cells, separated by thin cell-walls (Fig. 16). This row of four cells was formerly known as a promycelium, but is now called a basidium. If kept in water these cells can round off and separate from each other1, and germinate by sending out a tube, like the mycelial cells and spores of many fungi. But if in a damp- atmosphere, each cell without separation produces a sterigma at the end of which a basidiospore is formed, like the basidiospores of Agaricini. These basidiospores can germinate at once, even before they are detached from the sterigma, by sending out a short tube which may produce a conidium resembling the basidiospore at its end. Fig. 16. P. Caricis. a, a mature teleutospore ; b, upper cell of same, germinating and forming a basidium ; c, basidiospores germinating, x 600. 1 This method is said to take place normally in Barclayella. - It is noted by many observers that, in a state of nature, it is a layer of dew, not of rain, that is favourable to germination. GERMINATION OF BASIDIOSPORES 15 If one of these easily detached basidiospores or conidia is conveyed to the surface of a leaf or young stem of Nettle, its germ-tube bores through the cuticle and enters the tissues (Fig. 17), where it ramifies and forms a my- celium. The teleutospore is large and heavy, and firmly attached to its spore-bed on the leaf of Carex ; the basidiospores enable its con- tents to be transferred easily to Fi§- 17\ Endophyllum Sem- J pervwi. (Terminating basi- the surface on which alone they diospores (after Hoffmann) ; i i c c j-i j.1- "D j. s, the spore; v, the germ- are capable of further growth. But v'esicle; mlder t£e cuti*le of their wall is thin and they can live the epidermis ; a,b, c, show , n i , • the passage of the spore- only for a short time ; they contain contents into the vesicle. but little food-supply and could not x about 20°- form a long germ-tube. That is the reason why their germ- tubes do not, like those of the other spores, search for a stoma, but enter by the quickest means. Nevertheless they can abnormally enter by a stoma ; De Bary records such a case in his account of P. Dianthi (see Fig. 24). The germination of the teleutospores of P. Caricis takes place about the second week in April, and on the mycelium produced by the basidiospores in the nettle there arise, in about a fortnight, first spermogones and then secidia like those with which we started. But the mycelium arising from the basidiospores does not always proceed immediately to spore- production. In some species, e.g. Endophyllum Sempervivi, it hibernates in the growing point of the shoot, or in the leaves if they are evergreen, as in Puccinia Buxi, or in the steins or branches in the case of some that live on shrubs or trees, as in Cronartium ribicola. Rather more than a twelfth of the species of Uredinales are now known to be heteroecious. This mode of life may be regarded partly as a device by which the parasite tides over the time during which one of the host-plants is not available. The leaves of the Nettle are delicate and soon perish in the autumn ; those of the Sedge persist throughout the winter. The power of heteroecism increases the ability of the fungus to L6 HETEECECISM adapl itself to new conditions and thus tends to perpetuate the race, while the change of host, which is equivalent to a change nt' diet, may very possibly tend to an increase of vigour in the hhIi\ [dual. The former statement, however, must not be taken to refer to all cases of hetercecism, since there are instances, e.g. in species of Goleosporium, which cannol be explained on this ground. The reason why P. Caricis has been taken as the typical rivdine, instead of the usual P. gra minis, is that the aecidium of the latter is now very rarely found in this country and is t herefore not available for demonstration, while that of P. Car ins is common in all suitable localities. Even if not existing in any place, it can be readily introduced if the three prerequisites are at hand: — a pond bordered by Garex paludosa and by Urtica dioica, and also a quantity of the leaves of Garex infested by the parasite. The Nettle and the Sedge are nol injured appreciably by the disease, nor would it be of much consequence if they were. To introduce the fungus into the new locality, it is only necessary to obtain a bundle of the required leaves (say about 500) from some place where the Pucdnia exists, in January or February, and lay them on the ground where a patch of nettles is known to occur. The latter will be seen in spring to be beautified by the aecidium, and in the summer the disease will spread to the surrounding sedges (see Grove, Journ. But. L913, p. 42). CHAPTER II THE SEXUALITY OF THE UREDINALES De Bary suggested in 1884 that, if there was any sexual act occurring in the life-cycle of the Uredinales, it would probably be found in connection with the aecidium. At that time nothing was known on this point, and De Bary anticipated that something might take place analogous to what happens in the formation of the asci of certain Ascomycetes. The first discovery was made by Blackman (1904), who found that, in laying the foundation of the spore-bed of the secidium of Phragmidium violaceum, cells became binucleatc by the passing into them of a nucleus from an adjoining vege- tative cell. He saw that this passage took place through a nar- row opening formed between the cells (Fig. 18). The bi nucleate cell then became the equivalent of an oospore, and formed the be- ginning of a sporophytic genera- tion. In other words, he supposed that the fertilised cell represented a female gamete and the vegeta- tive cell replaced a now vanished or functionless male gamete. The fertilisation would then be of the nature of a semi-apogamy. He considered that the spermatia were the functionless male gametes. From the cell to be fertilised he saw an upper sterile cell cut off, which soon G. u. 2 Fig. 18. Phragmidium violaceum. Development of *cidium (after Blackman) ; e, the epidermal cells ; s, sterile cells ; in the hasal cells below, a nucleus is seen migrating into a fertile cell, /'. 18 BASAL CELLS OF .lK'lMA degenerated; this he considered to represenl an abortive tri- chogyne, in accordance with l)e liaiy's anticipation. rii<- acceptance of this interpretation implies the existence of a close affinity between the Uredinales and the Red Seaweeds. In 1!)05, Christ man published the result of his researches into Phragmidium speciosum, Caeoma nitens (= Gymnoconia = Puccinia Peckiana), etc. According (<> him. the process that took place was the fusion of the contents of two equal and similar gametes, with the exception that the nuclei remained side by side unfused. A considerable portion of the wall between the two fusing cells was broken down, and the process was of the nature of a conjugation, not a fertilisation. Blackman and Fraser (1906) next examined a number of other species, and in Melampsora Rostrupii they found the same process which Christman had observed, though they still considered that other species, e.g. Puccinia Poarum, showed instances of the migration of a nucleus as in the first subject studied. Christman, in 1907, showed that a similar act of conjugation between two equal cells takes place in the formation of the primary uredospores of Phragmidium Potentillae-canadensis (= Kuehneola Tormeutillae, Arthur, q.v.), the primary uredo- spores in this species replacing the secidium which is absent. In 1908 Olive, in examining the primary uredospores of Triphragmium Ulmariae, tried to reconcile the difference be- tween these opposing views: he considered that conjugation took place between two cells, one larger and one smaller, and that either a large opening was formed so that the two proto- plasts fused, or a narrow hole was produced through which the nucleus of the smaller cell passed into the larger. He considered the upper sterile cell as a degenerating tip-cell, not an abortive trichogyne. The fusing cells might be placed in almost any position with respect to each other. In Puccinia transformans, a micro-form, possessing only teleutospores, he shows that the basal cells which produce them arise equally by the fusion of two uninucleate cells. Kurssanow, in 1910, investigating Puccinia {Gymnoconia) Peckiana, found both cases that of Blackman and that of STERILE CELLS 19 Christman, occurring side by side. All the conjugating cells had an upper sterile cell which he calls a "buffer" cell; but the passage of the nucleus only he put down, as others have done, as a pathological phenomenon, caused perhaps by the method of fixing. In November of the same year Dittschlag, investigating Puccinia Falcariae, tried to settle the question and decide definitely the function of the spermatia. This Puccinia is an -opsis form, having spermogones and secidia, followed later by teleutospores, but without uredospores. He showed that the cells of the spore-bed of the secidium unite in pairs by the disappearance of not quite all the separating wall. If a sterile cell could be seen at all, it was seen equally on both (Figs. 19, 20). But this does not militate against its being considered as a degenerate trichogyne : it is certain that the two cells which Fig. 19. Puccinia Falcariae. Con- jugation of two female cells to form the basal cell of the ascidio- spore-chain (after Dittschlag). The uppermost cell on the left in a does not belong to the others. Each fertile cell has a sterile cell above it. In b, the hist conjugate division is just com- pleted (Diagrammatic). Fig. 20. P. Falcariae. Formation of jecidiospores (after Ditt- schlag) : a, the basal cell; b, an secidiospore-mother-cell ; c, the same in the act of conjugate division (the nucleoli are seen in the middle) ; d, the inter- calary cell cut off. fuse are in most cases exactly alike, and therefore, if they represent potential female cells, each of them would naturally be provided with a trichogyne in equal degree, if at all. 2—2 20 ISASAL CELLS ()!•' .ECIDIA Again, in l!»l I, Hoffmann invest igated a lTrodine of a lower t\ |»- i li; iusi of i hose |)i-e\ iously considered, \ i/. tindophyllum Sempervivi. This genus differs from all the other Uredinales in its mode of development. It has only spermatia and secidio- spores, the latter functioning also as teleutospores in that their conjugate nuclei fuse, and then on germination they produce a basidium and basidiospores. These basidiospores reinfect the host and produce both spermogones and aecidia. On the spore- bed of the secidium two ad- jacent cells unite by the disso- lution of the intervening walls (Fig. 21); first a small hole is formed, which grows larger until at last almost no trace of the wall is left. The disap- pearing wall is often horizontal, not vertical as in most of the other cases,and the conjugating cells are not situated in any definite plane. In such cases a sterile (trichogyne) cell was not seen. Finally Fromme (1912) found that in Melampsora Lini the spermogones and aecidia are produced simultaneously and only from infection by basidiospores. The spermatiophores differ from all others described in being many-celled, each cell producing a single spermatium on a sterigma-like process; they arise from a regular layer of large rectangular cells at the base of the spermogone. The aecidia are stated to be undistinguish- able from the spermogones externally, but produce female gametes in the usual way, generally with one or two " buffer " cells which speedily disintegrate. The female gametes conju- gate, in abundance, laterally in pairs, often in threes or fours ; the fusing cells are of equal rank, but need not be in the same horizontal level. iEcidiospores were observed with several nuclei, and one secidiospore-mother-cell was seen with as many as eleven nuclei. (See also note on p. 29.) Fig. 21. Endophyllum S< mpervivi. Formation of aaeiciio-teleutospores (after Hoffmann), a, fusion-tissue; /;, basal cell ; e, intercalary cell ; d, spore; e, intercalary cell, degenera- ting. TRICHOGYNE HYPOTHESIS 21 From a comparison of all these observations it now becomes certain that, in the Uredinales, the typical mode by which the binucleate condition arises is by the conjugation of two similar cells, each provided with a large nucleus and an abundant supply of food. This fusion-cell can afterwards branch by the formation of lateral buds, usually in basipetal succession, and may thus produce several rows of spores at once ; by a similar branching bunches of uredospores and teleutospores can arise in the sori of those spore-forms (Blackman, '06, Christman, '07, Dittschlag, 10, Hoffmann, 11). See Figs. 37, 156. At first, it may be supposed, the two conjugating cells belonged to a definite basal layer or spore-bed, as in P/irag- midium, Puccinia spp., and Melampsora Rostrupii ; but after- wards they ceased to be arranged in a layer, and conjugation took place between two purely vegetative cells in the mycelium beneath. The beginnings of this change are seen in M. Rostrupii, and its final product in such wn'cro-forms as P. Adoxae where the greater part of the mycelium has synkarya. Whether the upper sterile cell which is so frequently met with is to be considered as an abortive trichogyne is not so certain. But it may be remarked that, since the sori generally arise beneath the epidermis, no fertilisation could have taken place by non-motile spermatia unless there were something of the nature of a trichogyne to protrude through a stoma. In this connection it is important to remember that in Uredinopsis, one of the lowest of the Uredinales, the sori of primary uredo- spores, i.e. secidiospores, seem always to arise beneath a stoma ; other sori can arise in many genera in the same way, and in the genus Hemileia the pedicels of the uredospores protrude into the air through the stomatal pore. Moreover we know that, while secidia of the enclosed higher types, as in P. Caricis, arise at some depth in the host-tissues, and the basal layer and its peridium are covered by a considerable thickness of dead empty cells (which are afterwards pushed to the sides), the secidia of the more primitive form, the cseoma type, are shallow and are not enclosed in a peridium, but are either quite naked or surrounded only by a few paraphyses. In the typical Uredinales, the conjugation of the two ±1 ABSENCE OF ASCOGONIUM (female) cells takes place just before the formation of the secidium or its representative ; in the reduced micro- and other forms, such as /'. Adoxae and Uromyces Scillarum, it is probable that the conjugation of the two (vegetative) cells takes place at some more or less indefinite period before the formation of teleutospores. There is a general agreement among investigators that a structure resembling an ascogonium (from which the mass of basal cells may be supposed to originate) does not exist in the Uredinales, notwithstanding the suggestions to that effect by Massee (1888) and Richards (1896). If it did exist, or had existed, it would do something towards accounting for the definiteness in form usually presented by an secidium ; if it is totally unrepresented, the secidium cannot be regarded as a morphological unit, but only as a collection of female cells. It is possible that traces of its existence are shown by the large multinucleate cells, containing 12 — 15 nuclei, which have been described by Olive (1908) in the mycelium at the base of the young secidium of Puccinia Cirsii-ianceolati and by others at the base of teleuto-sori, but this question must remain open till further investigations are made. The existence of an asco- gonium of that kind would, of course, be inconsistent with the trichogyne-interpretation of the sterile tip-cell in the secidium. /Ecidiospore-mother-cells and secidiospores with three el- even more nuclei are frequently met with : these represent the result of a fusion of three or more cells of the spore-bed. But since uredospores and teleutospores with cells containing more than two nuclei seem to be unknown, it is probable that these abnormal secidiospores undergo no further development. The Nature of the Spermatia. There are two and only two possible interpretations of the spermatia — either they are male gametes, or they are conidia, i.e. merely additional multiplicative spores like the uredo- spores. In favour of the former view the following arguments can be adduced : (1) The time of their appearance, just before the formation NATURE OF THE SPERMATID 23 of the secidium or its representative, and on the same my- celium. For instance, in those cases where there is no gecidium, but primary uredospores which are formed, like yecidiospores, from a fusion-cell, followed by secondary uredospores which are not so formed although similar in all other respects — then the primary spores alone are accompanied by the spermogones. This argument is the most decisive. (2) Their size and character. They are much smaller than the other spore-forms, with thin walls, a large and not very degenerate nucleus though often without a nucleolus, little protoplasm and no reserve-stuff (oil, etc., with which the ordinary spores of the Uredinales are so richly provided), thus reminding one of the spermatia of the Floridere. Sharp (1911) reports, in Puccinia Podophylli, spermatia three times as long as the nucleus, and therefore containing some appreciable amount of cytoplasm. But protoplasm is not reserve-stuff. (3) They will not reproduce the species. All the efforts that have been made to cause them to do so have uniformly failed. All the other reproductive cells of these Fungi can be successfully used for that purpose, if applied to the proper host. In some species, as Cronartium ribicola, the spermatia can be collected in large quantities: Klebahn made numerous trials with them, but entirely without result. Jaczewski and others have confirmed his experience. It is a commonplace observa- tion that highly specialised male cells cannot in themselves reproduce the species, while female cells can, as in the cases of parthenogenesis, both true and false. This difference in be- haviour is partly correlated with the difference in the amount of food-reserve available, with which the larger female gametes are usually well supplied. In conjugation, where the two (male and female) gametes are approximately of equal size (as in certain Mucorini), each may form a functional azygospore. (4) They will hardly germinate in water, probably because they have no reserve-food. If food is supplied by cultivating them in nutrient solutions, a little growth is obtained, but it is very insignificant and soon perishes. The same thing is true of male cells in other organisms. Conidia, under such 24 NATURE OF THE SPERMATIA circumstances, would shew quick and luxuriant growth : tln-v cannot be degenerate conidia, because the nucleus is large and well-formed though at times no nucleolus can be seen. (5) They are sometimes accompanied by a sweet fluid, which gives off a pleasant, or more rarely an unpleasant, smell, as in P. suaveolens (obtegens). Uromyces Pisi, Cronartium Quer- chs, etc. It is said that, in Japan, children lick the abundant spermogones of C. Quercus on account of the sweet juice that oozes from them. The presence of this can be readily under- stood, if the aid of insects is invoked as well as wind, in order to carry the passive spermatia to the trichogyne projecting through a stoma, but otherwise is without explanation. The larva of a fly (Diplosis) or a similar organism, is to be found crawling about the leaf and feeding on the spermatia and aecidio- spores of many Uredinales; its body is quite orange in colour through being filled with them, and the spermatia would adhere to its outer surface. Though the spermogones of P. Caricis are usually on the opposite leaf-surface to the aecidia, yet in very many species they occur intermixed, and not infrequently the secidia grow habitually in circles round little groups of spermogones; a remarkable instance is seen in Phragmidium Rubi-idaei (=P. gracilei : see Fig. 225. (6) The most likely theory of the evolution of the Uredinales is that which places the majority of the micro- (including the lepto-) forms as the most recent. It is just in these, and in no others, that the spermogones are least often to be met with (see p. 39), as wTould be expected if they are furthest in descent from the primitive forms in which a true act of fertilisation occurred. (7) If, on the other hand, we look upon the spermatia as conidial forms, i.e. as merely an additional means of vegetative multiplication, we are confronted by this difficulty (as well as those referred to above) that they appear just at that period of development at which they are least wanted, whilst they are missing in many micro-forms where additional help would be most welcome. The ascidiospores have been shown in many ways to possess an unusual amount of vigour and to be able to produce a stronger infection than the uredospores, which stand NUCLEAR DIVISION 25 next in order. Even, therefore, if the spermatid could produce an infection, their feeble aid would be wasted at such a time of rejuvenation. (8) There is also to be considered the fact that the spermogones and spermatia of the Uredinales resemble those of the Collemacea?, which have been shown by Stahl (1877) and Baur (1898) in all probability to fulfil the male function1. It may be pointed out, in this connection, that the great similarity of the spermogones to the pycnidia of the Ascomycetes has been too much ignored, and that its significance is not yet fully appreciated. In the Ascomycetes the pycnospores in most cases undoubtedly act as conidial forms, and have lost all traces of their primitive male function — in the Uredinales the sper- mogones have equally lost their function, but have not taken on the secondary role of conidia : it may be suggested that the latter are otherwise well provided for in that respect, and hence feel no necessity for additional conidia. The spermatia of Polystigma rubrum are, however, functionless either as male cells or as conidia (Blackman and Welsford, 1912). Nuclear Division in the Uredinales. This is always of a simple type, not primitive, but reduced. The number of chromosomes seems to be always somewhat uncertain, and the chromatin forms masses which vary in number from one to four. In the ordinary vegetative division, which may be regarded as approaching rather to the nature of amitosis, the nuclear membrane disappears, the nucleolus is extruded, and the chromatin masses are drawn apart on a kind (>f rudimentary spindle to form the daughter nuclei. In syn- karya, the two paired nuclei in a cell are almost always in the same stage of division at the same moment, and the four resulting daughter nuclei move apart in such a way that the two nuclei in each daughter cell are never sister-nuclei (Hoff- mann, 1911). 1 A remarkable instance in Collema, though outwardly not at all resembling the case of the Uredinales, is described by Bachmann (1912). 26 NUCLEAB DIVISION In the beleutospore (i.e. tetraspore-mother-cell), the first division is of a slightly higher type. The fusion-nucleus is large, round and (when unstained) perfectly clear and homo- geneous, but for its nucleolus, so that it looks like a vacuole : it occupies almost invariably the middle of a cell. The dense chromatin mass is loosened out into a kind of spireme which becomes shorter and thicker; the nuclear membrane then disappears, and the spireme thread splits longitudinally, though the splitting is often indistinct. It then divides transversely into segments which become arranged or strung out on a spindle (sometimes, but more rarely, in an equatorial plate): then the daughter nuclei are formed at the poles, and the next division, which is homotypic, follows immediately (Harper and Holden, 1903; Blackman, 1904). Hoffmann considered that in Endophyllum Sempervivi he could count eight chromosomes just before the reducing division. Since each of these nuclear divisions of the teleutospore- contents is usually followed at once by the formation of a cell- wall, there are obtained four cells which are generally super- posed in a row. But Weir (1912) records a case in Coleosporium Pulsatillae where they were arranged in a "tetrad," by which word he means presumably (for he gives no figures) in a square or tetrahedral manner. According to Dittschlag (1910) the nucleus of the spermo- gonial hyphae is oblong and shows a slight chromatin network, but usually without a nucleolus. After abstriction (for which see Blackman, 1904), each nucleus enters again upon a resting stage, and the chromatin network becomes looser. Each sper- matium has a rather large nucleus, occupying about two-thirds of the cell, showing a decided chromatin network, but almost never a nucleolus. It has been frequently noticed that many spermatia soon become binucleate, but the nuclei are sisters, and this condition is merely a beginning of vegetative growth which, however, usually aborts. In the secidia, the " fertile " (female) cells have a medium- sized nucleus, with a fine chromatin network and a deeply colouring nucleolus, as well as abundant finely granular proto- plasm. When the conjugate nuclei have arisen, they lose their ALTERNATION OF GENERATIONS 27 nuclear membrane, the chromatin in each falls together in little masses, the nucleolus lying by the side, of the mass ; these thou separate, a spindle is formed with an equatorial plate and the chromatin masses pass at different rates towards the poles. The nucleoli then disappear, and four chromatin balls are formed which are the daughter nuclei. Only after this is com- pleted, do they separate into two pairs, and the formation of a cell-wall begins (Dittschlag). The Alternation of Generations. Assuming, as we are now justified in doing, the truth of the foregoing ideas, we may represent the alternation of genera- tions in a typical Uredine by the following diagram : teleutospore basidiospores uredospore. SPOKOPHYTE (2n generation) uredospore^ ascidiospore" intercalary cell' GAMETOPHYTE (n generation) spermatium $ gamete ? gametes fusion-cell The n generation is that in which the nucleus has the haploid number of chromosomes or, if there are no distinct chromosomes visible, the single amount of chromatin ; in the 2/; generation each cell has the double (diploid) number of chromosomes or the double amount of chromatin, at first sur- rounded by two nuclear membranes, afterwards by one. 28 TK'I'i: VSPORE-MOTHER-CELL The beleuto3pore-cell is a sporo-mother-cell, exactly com- parable with a tetraspore-mol her-cell. The" basidium" is merely tlif same cell, removed outside the old outer cell-wall for the convenience of the ensuing processes; in Goleosporium this removal does ool lake place, bul the "basidium" is formed internally, the mucilaginous nature of the cell-walls of the teleutospores of thai genus allowing the sterigmata to protrude through them (see Fig. 241 >, which could qo1 be done through the hard cutinised cell-walls of the teleutospores of Puccini" ( \i ricis. The Hist division of the nucleus of the beleutospore is heterotypic, and really initiates the gametophyte; but, since this is not a sufficiently definite point in the cycle, it is usual to consider the gametophyte to begin after the next division, i.e. by the formation of the four basidium cells, which constitute a tetrad. These four cells are the true tetraspores. In water, deprived of air, they can each send out a germ-tube and, it is presumable, could cause infection. The formation of basidio- spores on sterigmata is a secondary process, viz. the product ion of conidia suitable for wind-dispersion: this is shown by the fact that, if the basidiospores are not on the proper host plant, they can themselves germinate with the production of fresh conidia of an exactly similar character. The mycelium (thallus) of the gametophyte, issuing from the basidiospores, bears male and female organs, the spermo- gones and aecidia. The spermatia disintegrate without any result: the "fertile" cells of the secidium (usually, perhaps nearly always, after cutting off a sterile cell, as is the habit of female gametes) are stimulated to further growth by con- jugation with one another, the delay in the fusion of the nuclei being of little or no importance from this point of view. This so-formed fusion-cell is a zygote and is the beginning of the sporophyte. The recidiospores and uredospores which are borne by this are conidial forms, devoted to increased multiplication, and may continue indefinitely till the time arises for beginning the cycle again. This is indicated by the dotted lines on the left of the diagram. It is known that a fusion of two nuclei, comparable with that COMPARISON AV1TH BASIDIOMYCETES 29 which takes place in the maturing teleutospore, occurs also in the basidium of the Agaricini and Polyporei, followed by a division of the fusion-nucleus into four nuclei of which one passes into each basidiospore, although it is not yet ascertained how or where the cells of the hyphse of those Fungi become binucleate. From this point of view it is evident that the chief difference between the basidiospore-formation in the Basidiomycetes and in the Uredinales lies in the fact that, in the former, the four tetraspore-nuclei are not surrounded by cell-walls previously to the production of conidia, as they are in the latter. The basidium of some of the Hemibasidiomycetes, e.g. of the Auriculariese, is divided into a row of four super-imposed cells of an exactly similar character to that of Puccinia, each cell also giving rise to a basidiospore on a sterigma in the same way. The similarity of this basidium to that of Coleo- sporium is not diminished by the fact that it also is surrounded by a gelatinous mass through which the sterigmata protrude. Had not this primitive mode of forming the conidia been modified into that typical of the Agaricini, there would have been no opportunity for those wonderful and intricate contri- vances for facilitating spore-dispersal which Buller has pointed out (1909) and which find their highest and latest development in Coprinus. The Uredinales must be considered a highly organised group of comparatively recent evolution, as is evident also from their exceedingly complex parasitism. They are not a stage in the evolution of the ordinary Basidiomycetes, but the end-group of a different branch. Note. Werth und Ludwigs (1912) showed that the teleutospores of Puccinia Malvacearum arise by the conjugation of two basal cells like those of an secidium, but usually of unequal size. The nucleus of the smaller cell passes into the larger; the fertilised cell then forms, by conjugate division, a short chain of binucleate cells, of which the two upper become the teleutospore. CHAPTEB III SPORE FORMS OF THE UREDINALES /ECIDIUM. . Kcidia are usually of a cup-like shape, partly embedded in the host, and with the free protruding edge more or less recurved. This is the typical and presumably the most highly evolved form. In it the spores are at first completely enclosed li\ a firm structure, the peridium, the cells of which have the membrane thickened on the inside wall or the outside or both, ami arc arranged in very definite rows like the spores. Bu1 there are a number of variations on this type, most, if not all, of which belong to a lower stage of evolution. Sometimes the peridium has the cells less definitely arranged in rows, and i herefore opening more irregularly ; at other I hues the peridium is thin-walled and delicate, and in that case it usually opens by a rounded pore and the edges do not roll back. In Hyalopsora and its allies such a peridium is formed round the uredo-sori, and there are reasons for believing that in these cases the pore arises just beneath a stoma. A still simpler stage is represented by those cases where there is no definite peridium at all, but merely .-i surrounding circle of paraphyses which in a few cases are almost or even totally non-existent. This is called a Carina and indicates a more primitive form ; it is found in Phragmidium and Melampsora. In the non-British Gymnoconia Peckiana (= Puccinia Peckiana, on Rubus) peridium and paraphyses are both entirely absent. Again, if there is a definite peridium, it need not have the .shape of a cup. It may be elongated-cylindrical, straight or curved like a horn : this is called a Roestelia and is confined to FORMS OF ^ECIDIA 31 the genus Gymnosporangium, on the Pomaceas. Or it may be oblong or globular, and more or less inflated : this is called a Peridermium, and appears to be confined to the Coniferae (leaves and stems) as hosts and to belong only to the genera allied to Coleosporiam and Gronartium. None of the Peridermia have been found in Australia. It is a remarkable fact that aecidia are never found, upon Juncaceas and Cyperaceaa, nor upon Graminese with only two exceptions — the sscidium of Uromyces JJantlioniae, on Danthonia in Australia, and of Puccinia grami- nella, on Stipa in North and South America. In all other Uredines parasitic on these families, if aecidia enter into the life-cycle of the fungus at all, they are formed upon some broad- and thin-leaved Monocotyledon, or upon a Dicotyledon, usually though not invariably belonging to one of the more specialised orders and above all to the Composite. The essential characteristic of the aecidium is that its spores are produced in chains from a fusion-cell, as described in Chapter I. The spores themselves are always unicellular, mostly with orange contents, and separated by intercalarv cells. Their customary polygonal shape arises entirely from crowding and their verruculose sculpture presents a remark- able similarity through all the group. The cells of the peridium of the higher types are homologous with the aecidiospore- mother-cells, and represent a division of labour for the sake of protection : the paraphyses and the lower forms of peridium are not of the same character, and may have had a somewhat different origin. All gecidiospores, except those of Endopliyllum, germinate conidially as in P. Garicis and the germ-tubes enter the host through a stoma : the germ-pores are numerous and almost always indistinct. When there are secondary gecidiospores, i.e. such as arise from the germination of a previous gecidiospore, they always take the place of uredospores. In such cases, only the primary gecidiospores arise from fusion-cells, and are accompanied by spermogones. There is said to be one case where gecidiospores are uninucleate, and thus comparable with azygospores, but further evidence is required before this statement can bo accepted (Moreau, 1911). 32 spermogones Spermogones. While aecidia are always subepidermal, and in the higher forms sunk rather deeply in the tissues of the host, the spermogones are in certain genera subcuticular and in others subepidermal, bul always shallow. This doubtless corresponds to the primitive form of tin- .e-idium, when the trichogynes protruded through a stoma in order to catch the spermatia. Wheu fertilisation was dropped, it became possible for the secidia to be surrounded tor protective purposes by a continuous and firm peridiura and to be more deeply sunk within the host: the spermogones. being outside the range of selective evolution on account of their uselessness, have retained more or less of their original character. The description of those of P. Garids applies essentially to nearly all, except that in certain genera they are flat, not Mask-shaped, and open by a wide pore. not by an ostiole, and in these and others there are no protruding ti laments. Spermogones never appear alone; they are always accom- panied or closely followed by some other spore-form, either acidio-, undo-, or teleutospores. In comparatively few cas - as in P. Malvacearum, have the spermogones disappeared en- tirely. Uredospores. Uredospores are usually distinguished from a-cidiospores by being produced singly at the apex of a short pedicel from which they easily fall off: this pedicel is the homologue of the intercalary cell of the aecidiospores, as was shown by Christman. In certain genera, however, e.g. Goleosporium and Chrysomyxa, the uredospores are produced in short chains. Uredospores can reproduce uredospores for an indefinite number of generations. ( )ccasionally there are two kinds, primary and secondary uredo- spores: in such cases the primary ones arise from fusion-cells and take the place of aecidiospores. So far as is known true uredospores never arise from a uninucleate mycelium, though they may spring from a mycelium which was at first uninu- cleate, but became binucleate at some point in its development. UREDOSPORES 33 This would be the case in a true Hemipuccinia, but none of these have so far been cytologically investigated. Uredospores are always unicellular, except in the case of a few monstrosities. Primarily they must be considered a device to aid in rapid propagation and hence may be called stniinier-sipores: for this reason they usually germinate with great readiness when mature, always forming a tube which enters a stoma of the host. The number of germ-pores varies from 1 to 10 (usually 2 to -i) : only one case is known in Uromyces where they possess a single pore ( U. uniporulus) and one in Puccinia (P. monopora). They present a wonderful sameness in shape throughout the whole group, and in colour vary from yellow and orange to brown. It may be taken as a general rule that if the wall is colourless, the contents are yellow or orange from an abundance of that yellow oily substance which occurs in ascidiospores ; if the wall is distinctly brown, the contents are often colourless when mature, though at first they frequently contain the usual yellow oil. It is only in a few instances, in the lowest genera, that uredospores are quite without colour. The uredospores of Puccinia dispersa and in a smaller degree of P. graminis are noticeable for a curiously dull appearance which is very characteristic, because they combine orange contents with a brownish membrane. The outer wall of uredospores is almost always covered with spines (echinulate), needles (aculeate) or warts (verruculose) ; it is very rarely smooth ; these projections enable them to cling more readily to the surface of the host. They are often inter- mixed with paraphyses, which are usually clavate or capitate in shape ; these are found in a few species of Puccinia, but more especially in Melampsora and Phragmidium. These paraphyses are homologous with the spores, being binucleate at first, but the nuclei soon disintegrate. There is very frequently found in the uredo-sori of many species a parasite belonging to the Deuteromycetes, called Darluca Filum. It consists of a black pycnidium, enclosing numerous uniseptate pycnospores which are faintly apiculate at each end. This has been sometimes mistaken for another spore form of the Uredine. G. u. 3 :;| AMPHISPORES Amphispores. In countries which are arid or Mini-arid, there is found in certain species a form <>f uredospore which has led to several mistakes owing to its misleading appearance. The spores are provided with a thick cell-wall or have the summit thickened conically, after the style of many a teleutospore of Uromyces, and are supported on a persistent pedicel, so that one would not take them for uredospores; nevertheless they will be found to have more than one germ-pore and to germinate by a germ- tube, although only after a period of rest. These were named by Carleton amphispores : they were first discovered in Pnccinia renins. They are evidently ;i provision to enable the spore to pass through an unfavourable period unharmed, and reinfect a hosi of the same species when occasion arises. The amphi- spores of Pnccinia atrofusca Holway, though echinulate and possessing two equatorial germ-pores, were first described by the discoverers. Dudley and Thompson, as the teleutospores of a Uromyces, and the same thing has happened in other cases, e.g. Puccinia convolvuli, P. tosta, and P. cryptandri. The nearest approach to amphispores found in British species is in P. Pruni-spinosae (Fig. 22), in which the}' have been mistaken for paraphyses and were so figured in a well- known text-book. There is another kind of spore, presently to be described, called a ■Fig. 22. Amplnspore raesuspore, which bears a superficial resem- of Puccinia Pruiii- * A . . spinosae. x600. blance to an amplnspore. It is a mistake to call amphispores a transition- form between uredo- and teleutospores, since they are of later evolution than the two latter. Teleutospores. The meaning of the word teleutospore is end-spore ; it was considered to represent the stage when growth was ceasing for the season. This is not the case, however, in all species, and the word must now be used with another connotation, viz. a teleutospore is one which germinates by the production of a basidium and basidiospores. TELEUTOSPORES 35 The teleutospores are generally produced in sori like those of the uredospores ; they frequently arise on the same mycelium, and very often on the same spore-bed, mingled with the uredo- spores. If both are found in any species, the teleutospores are always formed at least not earlier, and usually later than the uredospores. Their primary function now is to tide over an unfavourable period; for this reason they are sometimes called, in England, winter-spores. They may be one-celled as in Fig. 23. Figures of various Teleutospores of Pucciniaceae (after Sydow). From left to right, they are (in the top row) Puccinia roesteliiformis, P. conspersa, P. globosipes, P. Megatherium, Uromyces gtobosux, Hapalophragmium . Derridis ; (in the second row) P. appendiculata, P. Euphorbiae var. intit- mescens, P. deformata, U. achrous, U. giganteus, U. Ipomoeae, Triphragmium Cedrelae. All x 480. (Non-British.) Uromyces, two-celled as in Puccinia, radiately three-celled as in Triphragmium, cruciately four-celled as in Pucciniastrum, linearly many-celled as in Pliragmidium and especially in Xenodochus (see also Fig. 23). But it is misleading to speak of a many-celled spore : each cell, taken separately, is a spore and can germinate by itself without reference to the others. 3 — 2 36 TELEUTOSPOBES In the formation of teleutospores in the higher [Jredinales, the spore-mother-cell first divides into an upper fertile cell and a lower sterile cell, which elongates more or less to form the pedicel. The upper cell may remain undivided, or may divide again : the lower of these two may then continue to divide and so on, to form a many-celled chain. When the chain is long, as in Xenodochus, it is seen very clearly that the spores are formed like a^cidiospores to this extent that the uppermosl is always the mosl mature. This may be taken as a sign that they are modifications of secidiospores to form resting-spores. In Endo- pJiyllum the a-cidiospores previously mentioned germinate as soon as mature with a basidium, and are therefore teleutospores also: this is the primitive state of things from which t he- present wide division of labour into rejuvenating (aecidio-), multiplying (uredo-), and resting (teleuto-) spores has been evolved. In some of the lower Uredinales, the teleutospores are termed beneath the cuticle or in the epidermal cells, bu1 the usual position is directly beneath the epidermis. Throughout the whole group the colour of teleutospores is almost uniformly brown, varying in shade from a pale yellowish-brown up to nearly black. Their contents are, like those of uredospoiv>. at first often oily and yellow, afterwards colourless. In the lo\ - genera, those found on Ferns, the teleutospores are quite hyaline. Their surface is most often smooth externally, but occasion- ally marked with superficial unevennesses, such as waits, tubercles, lines, stria?, reticulations, and pits; a few have spiny, papillose, or finger-like processes, either at the summit or all round. The majority of them have one pore to each cell, as in Puccinia and Uromyces, covered at times by a distinct, often hyaline, pore-cap; this is the highest type, being furthest removed from the many-pored aecidiospores. Other genera have 2 to 4 germ-pores to each cell, as in Phragmidium and Gymnosporangium. In some cases, as in Uromycladium and Ravenelia, the teleutospores are borne in bunches at the top of a common stalk, either with or without accompanying hyaline cysts, i.e. abortive spores. An approach to this is found in the BASIDIOSPORES 37 British Puccinia Pruni-spinosae and other species, where the short pedicels are all closely bound together in bunches at the base. Paraphyses are naturally not so common in teleuto-sori as with uredospores, since the former do not need such protection, but they are found in P. Sonchi, P. dispersa, P. persistans, etc., although in these cases the so-called paraphyses are not at all of the same character as those found in uredo-sori, e.g. of Puc- cinia Baryi and others. When the teleutospore of a normally two-celled species becomes one-celled (by the omission of the last cell-division), it is called a mesospore : the mesospores of Puccinia are practically identical with the teleutospores of Uromyces and germinate like them. BASIDIOSPORES. All normal teleutospores develop under natural conditions in the same way ; the cell-contents divide themselves into four parts, by a heterotype followed immediately by a homotype mitosis. This formation of what are really (and might with ad- vantage be called) tetraspores can take place in two ways : — the "basidium" can arise within the teleutospore-cell or outside it. The first method is the more primitive, the second is an adaptation to the tough cutinised or chitinous exospore of the more advanced types. In the Coleosporiaceas the teleutospore, i.e. the tetraspore-mother-cell, divides into four superposed cells (like the tetraspores of Gorallina) while still in the sorus, during the autumn ; each cell (spore) germinates, in late autumn, by protruding a sterigma through the thin gelatinous wall of the teleutospore and forming a basidiospore (conidium) at its apex. Zaghouania shows an intermediate form of germination. But in all the other families the cell-contents of the teleutospore, clothed only in a thin endospore, pass out through a germ -pore in the form of a longer or shorter tube (" basidium ") ; the contents pass to the distal end of this, and are there divided into four oblong cells. The median septum is sometimes formed first, and the two lateral ones after. In water the " basidium " is usually long, in air it is short. In the absence of sufficient 38 BASIDIOSPOEES air, moreover, i he four cells may qoI be formed ; the " basidium may rcsoinblc, more or less, an ordinary germ-tube and possibly may function as such. Even if the torn- cells are formed, bhey may germinate by the protrusion of a germ-tube, which pre- sumably can cause infection by penetrating the cuticle (see Coons, 1912. ]». 225). But, with access of air, each cell forms a sterigma and a basidiospore as previously described. The-. conidia are obviously adapted for wind-dispersion. If they alight on a suitable surface, they send forth a short tube which invariably bores straight through the cuticle into the under- lying epidermal cell of the plant, and there begins to form a mycelium. The only instance in Puccinia known to the con- trary1, out of the many observations that have been made and figured of this process, is De Bary's record of the case {Ann. Sri. Nat. Bot. 4. xx, 1863, pp. 88-9) where the germ-tubes of the basidiospores of Puccinia Dianthi (q.v.) penetrated through the stom- atal openings of Diantlius barbatua (Fig. 24). In grasses and sedges, it is easy to see that the siliceous cuticle would present a great ob- Pig. 24. Basidiospores lb) of . , „ , , Puccinia Dianthi, germina- stacle to the entry ol such a tube, ting on leaf of Dianthus, wnile not impeding germ-tubes showing the germ-tubes mak- , r ° ° ing straight for the stomata which enter through a stoma, and (afterDeBary'Bfignre). x390. ^ ig probably the reason why secidia are so rare on the order Glumiflorse. In the hetercecious species no one has yet brought forward indisputable evidence to show that basidiospores can infect the host which bore the teleutospores, although statements to that effect are made. The shapes of basidiospores are not irregular ; they are more or less constant in each group. In Puccinia and Uromyces they are ovate, somewhat flattened on one side, or kidney- shaped. In the Phragmidiea? they are almost spherical ; in the Melampsoraceae small and roundish. In Endophyllum they are ovate; in Goleosporium they are large and ovate and a 1 Statements have been made of other instances, but most of them on insufficient authority, and some have been specifically disproved. The genus Coleosporium seems, however, to be an exception. GROUPS OF SPORE-FORMS 39 little flattened on one side, while in Ochropsora they are spindle-shaped. This is one of the reasons for suspecting that it is probably incorrect to classify these two latter genera in the same family, on the basis of the " internal " basidium merely. The genus Chrysopsora, which has the same kind of basidium, belongs to the Pucciniacea?. Grouping according to Spore- forms. For certain purposes it is convenient to have names for the groups into which the Uredinales may be divided according to the number of spore-forms possessed by each, though it must never be forgotten that such a grouping does not in any way indicate affinity. The method usually employed is that put forward by Schroter, which (with a little modification according to present ideas) may be presented in the following scheme. Denoting spermogones by O, oecidia by I, uredospores by II, and teleutospores (with the ensuing basidiospores) by III, we call a fungus possessing fAuteu-form, if with all four on one plant. Hetereu-form, if O, I on one species, and II, III on another, an -opsis-form. a Brachy-form. a Hemi-form (in many cases perhaps the half of a Hetereu-form). a Micro-form (spermogones sometimes absent). Thus Melampsora Rostrupii is a Eumelampsora, Gymno- sporangium Sabinae is a Gymnosporangiopsis, Uromyces Anthyl- lidis is a Hemiuromyces, and Puccinia Ca7npanulae is a Micro- puccinia. A Leptoform is one, of whatever kind, in which the teleutospores germinate as soon as mature, without any resting period ; thus the spores of P. Malvacearum belong chiefly to the lepto-form, those of Endophyllum Sempervivi entirely so. Maire (Progress. Rei Botan. 1911, iv, 115) has proposed a much more complex arrangement on the same lines, which is quite needless and, it is to be hoped, will be quietly ignored. O I II III a Eu-form 0 1 III 0 iiiii mil [0] in K) (iROll'S OF SPORE-FORMS The relations of the various groups to one another are represented in the following diagram. The circle represents the Eu-form8; the substitution of any one chord in the place of tin' arc which it subtends shows how the life-history is short, ned in the other eases. Only the abnormal Endophyllum cannol be Included in such a scheme; it> spoiv-^ruuping could only be represented by the symbol O TTTf • basidium teleutospore uredospore ft o uredospore basidiospore mycelium mycelium aecidiospore fusion-cell CHAPTER IV LIFE-HISTORIES OF OTHER UREDINALES Puccinia graminis. The Black Rust or " Mildew " of Corn. Another common Puccinia, whose life-history is of greater economic importance than that of P. Caricis, is the well known P. graminis, the Rust or Mildew of Corn. This species has its spermogones and gecidia on Barberry (Berberis vulgaris and Mahonia Aquifolium) and its uredo- and teleutospores on many species of grasses, especially on cultivated wheat. It must not, however, be assumed that any rust found on wheat is P. gra- minis ; there are at least two other species common on the same host which, unless carefully examined, may be confounded with it, viz. P. triticina and P. glumarum, not to mention a form of P. coronata which sometimes also occurs on cereals. For this reason these four species are now distinguished as the Black Rust, Brown or Orange Rust, Yellow Rust, and Crown Rust of corn, respectively. The uredo-stage of P. graminis is known as Red Rust. If one merely substitutes Barberry for Nettle and Wheat for Carex, what has been said about P. Caricis is true in all essentials of P. graminis. The differences are not in the life- history, but in certain minor details of occurrence : e.g. the spots caused on the leaves of Barberry are small, round and red, while uredospores of P. graminis are most common on the leaves, and the teleutospores form long black striae on the 42 PUCCINI A GRAM I MS Fi^'. 2~>. Puccinia graminis. a, fficidia on Berberis ; b, uredo- spore ; c, teleutospores. culms; for these differences the systematic part can be con- sulted (Fig. 25). There is, however, one point of difference connected with P. graminis which possesses greal biological interest — its virtual in- dependence of the aicidial stage. For a long time it had been known that Barberry bushes in the hedges caused " mildew :' on the corn in the neighbouring fields, and when, in 1864-5, De Bary proved the hetercecism by experimental cultures, it was too hast ilv assumed that the a-cidiuni on the Berberis was just as es- sential to the rust on the corn as that on the Nettle is to the rust on the Sedge. Many facts new tend to show that this is not the case. In Australia and the plains of India the Barberry is un- known except as an introduced plant, yet the Puccinia occurs everywhere and does enormous damage. McAlpine records, in his Rusts of Australia that he made numerous attempts to infect imported species of Berberis with the rust of Australian wheat which is morphologically undistinguishable from the P. graminis of Europe, but all his efforts were in vain. The inevitable inference is that P. graminis, as it occurs in those countries, is a " biological " race which maintains itself by other than the primitive means. A similar thing is true, according to Lagerheim, in Ecuador, where also rust flourishes and does great harm. The facts now known concerning the specialisation of the Black Rust are treated of in a separate chapter, but there is one point which must be mentioned here. This concerns the mode by Avhich fresh epidemics are produced each year. Even if the Barberry is present, it is by no means certain that it plays any important part in these annual attacks. Apart from that, there are several possibilities: (1) the fungus may winter PUCCINIA GRAMINIS 43 by its uredospores, (2) by a perennial mycelium, (3) by Eriks- son's mycoplasm. The first possibility is entirely a matter of climate : it may take place in one country and not in another, or in the same country it may take place in one season and not in others. McAlpine and Cobb find viable uredospores all the year round in Australia, and Lagerheim says the same for Ecuador. But in northern climates it has been shown that the uredospores of P. graminis frequently lose all capacity for germination during the winter ; this is proved true of Sweden, North Germany, North Dakota, etc., but in the United States, south of Ohio, Bolley found germinable uredospores all through the year. Similarly in Bohemia, uredospores of P. dispersa, P. glumarum and P. Lolii can survive mild winters or in sheltered places (Baudys). Even though uredospores capable of germination may sometimes be found on wild grasses during the winter, it does not follow that those could start an epidemic next spring, owing to the specialisation which has been proved to exist, by which a form of P. graminis on one host is often incapable of infecting another host. In regard to the second possibility, we find again two opposing views. De Bary and others have searched in vain for mycelium in the growing wheat plants, before infection becomes visible, but Pritchard (1911) found mycelium resembling that of P. graminis both in the pericarp of wTheat grains and in various parts of wheat seedlings. He showed that large num- bers of wheat grains contained pustules of teleutospores, even visible in the neighbourhood of the hilum, but also hidden within the pericarp. He proved that the mycelium from the pericarp penetrates through the intercellular spaces, as well as through the cells, and " soon passes into the spaces between the leaf-sheaths where it grows rapidly and attacks the tissues at various points." W. G. Smith figures teleutospores within the seed of Oat (Gard. Chron. 1885, xxiv, 245, f. 53) and recidia in the pericarp and seed of Barberry (ibid. 1886, xxv, 309, f. 58). It is evident that, if this state of things prevailed on a large scale, nothing more would be required to explain the 44 PUCCINIA GRAMINIS origin of outbreaks <>f rust. It is uol inconsistent with this thai Ward was able to prove that the mycelium of a uredo- sorus extends only a little way round the margin of the sorus; thai may be and is true in certain cases, especially with regard to secondary uredo-sori, but in P.Caricis the mycelium extends up and down the leaf between the parallel vascular bundles, producing uredo-sori all along its course. The practical bearing of Pritchard's discovery is to show that seed from an infected crop should uever be used for planting. About the third possibility, it is difficult to come to any definite conclusion. Eriksson's hypothesis is that the proto- plasm of the fungus is present in the grain, mixed with the protoplasm of the host, in such a way that the two are indistinguishable. As the plant grows up, he supposes that the two grow together until, at a certain time, the protoplasm of the fungus separates itself from that of its host in the form of " Nucleoli," passes into the intercellular spaces through " invisible pores," then or earlier surrounds itself with a cell- wall, forms a mycelium, and begins its ordinary life by producing uredo-pustules. An intermediate stage, where the fungus- protoplasm has surrounded itself by a cell-wall but is still enclosed within the cells of its host, he named "special cor- puscles." The difficulty in dealing with this theory lies in its in- definiteness ; its author changes it from time to time to meet objections, and supports it by hazy microscopical observations, many of which are demonstrably the result of incorrect vision. His " special corpuscles" have been shown by Ward and Klebahn to be ordinary haustoria, Eriksson having completely overlooked the intercellular hyphse to which those haustoria were attached. It is incredible that the protoplasm of so highly evolved a fungus could live outside its cell- walls, as he supposes. Such a state of things is, of course, common in the lower fungi, Chytridina- and allied groups. In Synchytrium Solum' the fungus-protoplasm and the host-protoplasm may be seen in the same cell, before the latter has been comjjletely devoured by the former, and in that state they are even distinguishable by their microscopic- appearance. But it will need a great deal more "proof" before PUCCINIA POARUM 45 Eriksson's startling hypothesis can be accepted in regard to such a fungi as the Uredinales. The futility of Eriksson's mode of argument is seen in his suggestion (1908) that "other diseases such as the American Gooseberry Mildew can live within the infected shoots in a form scarcely visible to oar eyes." But direct evidence against the Mycoplasm Theory is accumulating. Jaczewski (1910) grew seeds obtained from many much-rusted plants, but he found that, when they were sown under glass and protected with adequate care from all outside infection, they all produced rust-free plants. Bolley, Linhart, Zukal and Klebahn had similar experiences1. Zach (1910) on investigating leaves and culms of Rye, infected with P. graminis and P. glumarum, found on the outskirts of the infection-patches all the states described by Eriksson, but he proved that in all of them fungal hyphse were present. In fact, Eriksson himself saw and repre- sented these hyphse, but calls them " radialen Strange " of his supposititious " Nucleoli," the said " Nucleoli " being merely the deformed remains of the nucleus of the attacked cell. As Marshall Ward (1905) remarked, Eriksson merely inverts all the stages of a fungus attack on a cell, and supposes the last state to be the first. This error and a misinterpretation of the microscopic appearances account for the whole wearisome persistence in an inherently improbable hypothesis. Puccinia Poarum. The Coltsfoot and Meadow Grass Rust. This species is economically of no importance ; its spermo- gones and secidia occur on the common Coltsfoot (Tussilago Farfara), and its uredo- and teleutospore stages on species of Poa, to which however they do little harm. Here, again, the 1 This does not accord with Eriksson's experience ; but then on some of his " protected " plants aphides also made their appearance, yet this does not seem to have suggested to him that the zooplasm of the aphides must also have been latent in the seed ! If the aphides got in, so would fungus spores, since it has been proved (Butler, 1905) that uredospores are carried by them and other insects. 4l> PUCCINI A POARUM life-history is in all essentials identical with that of P. Garicis, but it differs in one striking particular — there are two genera- tions of each stage during the year. The spermogones and secidia first appear on the leaves of Coltsfoot in .May and June, and are followed by the Puccinia on neighbouring leaves of Poa in July and August. Then a second crop of secidia, also accompanied by spermogones, appears from end of July to September, followed again by uredo- and teleutospores in September to November. The latter rest • lining the winter and infect the young Coltsfoot leaves again in the following spring. In countries that have a climate favourable for the growth of Poa, the uredospores may be found the whole year round, and the fungus can maintain itself by them alone. This is certainly the case in Australia, according to Mc Alpine, where the Coltsfoot does not exist, and the uredo- stage is most common in the winter months, i.e. June to September. In this country, the teleuto-sori are rather incon- spicuous, but can be found by searching carefully on the lower leaves of species of Poa round the spot where the Coltsfoot has been found affected by the secidium, especially in July and August. The aecidium of this species has been examined cytologically by Blackman and Fraser, and according to them the binucleate condition of the fertile cells is produced by the migration of the nucleus from one fertile cell to an adjoining one in the hymenial layer, and also occasionally by a migration from one vegetative cell to another at a point where the conjugating cells were below the level of the hymenium, although only a little lower. Cells with three or four nuclei were met with by them, and true conjugate division was observed in such cases. In sections which I have examined of this species, I have seen evidence which seemed to indicate (although not with perfect certainty) that conjugation also took place by the removal of a large part of the intervening cell-wall, and a con- sequent fusion of the cells as described by Christman and Dittschlag (Fig. 26). As will be seen from the figure, the cells of the peridium of this species differ slightly from those of P. Garicis in their shape, PUCCINIA POARUM 47 as shown in radial section. The outer edge of each cell is more prolonged downwards so as to overlap a portion of the cell below in an imbricated manner. The aecidium, in both, as is the Fig. 27. P. Poarum. An recidio- spore germinating on leaf of Poa annua. x 180. Fig. 26. P. Poarum. Vertical section through edge of aecidium ; showing a, the crushed cells of the upper myce- lium, pushed on one side ; b, the chain of peridium-cells ; c, the spore bed, giving rise to d, the chains of a?cidio- spores. x 600. The black dots are the nucleoli ; one aecidiospore-inother- cell has three nuclei. case in most of the higher forms of the Uredinales, arises deep Fig. 28. p. Poarum. o.ansecidio- spore germinating in water, m the leaf, and the densely x 250 ; b, the same, showing packed knot of hyphse which the germ-pores, x500. forms the starting-point of the hymenium, where the cells are full of protoplasm, is covered over by a number of nearly empty cells which are ultimately squeezed to the side by the develop- ing aecidium and are shown in the figure at a. It is easy to produce the teleutospores in a garden on a tuft of (say) Poa annua or P. pratensis, by planting quite near to it 48 PUCCINI A MALVACEARUM and overhanging it some Coltsfoot affected by the secidium : another tuft abuut ten yards <>tV can be used as a control. The uredo- and teleutospores will appear mi the former tuft in about 14 to 21 days. If this is done late in the year (September) I have found that only teleutospores are formed on the leaves of the I'mi. For the germinating a^cidiospores see Figs. 27, 28. Puccinia Malvacearum. The Hollyhock Rust. This Rust differs from all the others that will be mentioned in the simplicity of its life-history, and also in the fact that it is not confined (as almost all the others are) to one kind of plant or even to a few, but ap- pears, so far as is known at present, to range over the greater part of a sub-family. It has been found on over 50 species belonging to nearly all the genera of the Malvese, and it seems to be identically the same in every case. The mycelium develops in spring in the intercellular spaces of the young leaves and stems and produces little knots under the epidermis, on which a thick, round, hard, pale-reddish cushion of teleutospores is formed. These spores have short or very long pedicels according to their posi- tion ; they are mostly typical and two-celled, but mesospores with only one cell are not uncommon, and occasionally a few may be met with having three or even Ger. four cells. Most of them germi- minating spore ; a, a basidium nate at once, in the SOHIS, pro- breaking up into separate cells ; . . , , b, a basidiospore, x 600. ducing basidiospores in the usual PUCCIN1A MALVACEARUM 41) way (Fig. 29). The accumulated basidia and spores give a greyish tinge to the red-brown sorus. These spores can cause fresh infection and so the disease spreads rapidly. It is most active about the end of summer, and has often been the cause of a serious epidemic on the more susceptible kinds of Holly- hock. The sori are found on every green part of the plant, stems, leaves, petioles, bracts, sepals, carpels and fruits, and are even reported on the petals. There are no uredospores. The chief biological interest of this fungus concerns the way in which it passes the winter, a point about which there has been much dispute. There are two possibilities, (1) by perennial mycelium, (2) by over-wintering teleutospores. The first has been strongly advocated, and it is very likely (though one can hardly say it has been proved) that the mycelium does winter in the young leaf-rudiments that are formed on shoots at the base of last year's stems. Freshly formed sori have also been found on the cotyledons of seedlings which grow up in late autumn round the parent plants and which in certain cases can survive the dead season. But there seems to be absolutely no justification for the claim that the mycelium winters in the embryo of the seed. The disease can undoubt- edly be carried with the seed, in sori either on the bracts (portions of which are often mixed with the "seeds," i.e. fruits) or on the outside of the carpels themselves. Eriksson has lately {Ueber den Malvenrost, 1911) published a theory, similar to his well-known theory about the Rust of Corn, and standing or falling with it : he says that P. Malvace- arum perennates in the form of " mycoplasm " in the cells of the autumn buds at the base of the shoots, as well as in the embryos of the seeds of the infected plants. With these he says it grows up in an imperceptible form, mingled with the protoplasm of the host, permeating the newly-formed leaves and at last suddenly breaking out in the form of pustules of primary teleutospores, which afterwards spread in the acknow- ledged way. He explains the presence of this mycoplasm by stating that certain teleutospores of the previous autumn germinated by sending out " germ-tubes " which cut off " end- conidia." (This mode of germination of the late-formed spores o. u. 4 50 PUCCINIA MALVACEA.RUM of I'. Malvacearum is well known, though not usually interpreted in that way, see Fig. 29.) These " end-conidia " do not form a short tube, to penetrate t he cuticle of the host, but " pour forth their protoplasm, as it seems, without the formation of an opening, through the plasina-eonneetious of the outer wall of the epidermis of the host into an epidermal cell," and so into the tissues where it vegetates till required. It also exists in the same state in the seeds of the infected plants. The fungus, he says, "passes from the plasmatic into the filamentous state just before the outbreak of the primary pustules." It is clear, however, that the figures he gives do not prove what he asserts. Putting aside this purely supposititious and intangible method, the chief means of perennation probably lies in the fact that certain teleutospores produced at the end of the growing season have the power of lasting through the winter and germinating in the spring. Plowright, Massee and Tau- benhaus all agree in this: the latter (1911) kept infected leaves, gathered at Cornell University in the United States from the living plant on November 26th, both indoors at a low temperature and outdoors, and by testing spores taken from them at intervals from December to April found that they still remained germinable, though more and more slowly as time went on. Dandeno, however (9th Report Mich. Acad. Sci. 1907, p. 68), states that the fungus does not winter in the seeds ; he tried seeds of diseased plants, carefully excluding infection from outside, and found that they all produced healthy plants. His experience also was that no teleutospores remained viable till the next spring, but that the fungus maintained itself the whole winter through on mallow plants in sheltered spots. These differences may be partly a matter of climate, and as regards the " seeds," unless there were sori on them, they could hardly be supposed to carry the infection, even if they came from infected plants, except by the presence of " mycoplasm " or mycelium, neither of which has been proved. For this reason the chief means of preventing the disease (apart from using " seed " from uninfected plants) must be to gather and burn all dead leaves from the infested bed. When GYMNOSPORANGIUM CLAVARLEFORME 51 the disease does appear, spraying with Bordeaux mixture is the best remedy against its spreading. It is the very young shoots and the upper side of the leaves that most require spraying ; although the pustules appear chiefly on the lower surface of the leaves, there can be little doubt that infection by the basidiospores takes place mainly through the upper surface. Gymnosporangium clavariaeforme. The Hawthorn and Juniper Rust. This fungus produces its secidia on the branches, leaves and fruit of the Hawthorn and on the leaves of the Pear, and its teleutospores (there are no uredospores) on branches of the common Juniper, where it causes fusiform swellings. There are three other allied British species, of which G. Sabinae also attacks the Pear, but its teleutospores are formed only on Juniperus Sabina (the Savin Juniper). On the swollen branches of the Juniper, the parasite pro- duces in April and May numerous orange masses, which ooze out and sometimes reach more than 1 cm. in height. They vary in shape, but are usually more or less conical or tongue-shaped (Fig. 30). They consist of a mucilaginous mass in which large numbers of teleutospores are embedded. The mucilage is produced by the gela- tinisation of the cell-walls of the fungus, especially of the very long- pedicels ; it naturally swells and becomes more conspicuous in wet weather. The teleutospores ger- minate at once, while still in the mass ; the basidiospores are pro- duced in the usual way ; it has been shown lately in another species (Coons, 1912) that they do not fall off, but are jerked off the 4—2 Fig. 30. Gymnosporangium cla- variaeforme. Masses of teleuto- spores on branch of Juniperus (■■mi munis (slightly reduced) ; two teleutospores, x GOO. 52 HYMXnsl'nKAXCir.M CLAYAKI KI-OUME sterigmata, much in the same way in which the basidiospores are thrown off in the Agaricini1. Tiny accumulate in large numbers on the outside of the mucilaginous mass, and present the appearance of a golden-yellow powder. The mycelium of the fungus is perennial in the Juniper, spreading from branch to branch and producing a fresh crop of teleutospores each spring. If one of the basidiospores is blown by the wind or carried by insects to a moist leaf or young fruit or stem of Hawthorn, it germinates and holes through the cuticle in the ordinary way, and tonus there a limited patch of mycelium. It is said that the infection has been known to be conveyed for half Fig. 31. Gymnosporangium clavariaeforme. -Ecidia on leaf, fruit, and branch of Hawthorn (reduced) ; a, peridium, x 16. The fruit and gall on branch are shown as they appear when the peridia are old, and the mass looks somewhat like a honeycomb. a mile. This mycelium produces the aecidium-stage, which is not usually seen until the end of June, in about 18 — 20 days. The secidia are not of the ordinary cup-like shape, but of the form called Rcestelia : they are cylindrical, brown at the base and ochreous above where the peridium is torn into numerous slender filaments (Fig. 31). On the leaves they appear in roundish patches a few mm. in diameter, but on the twigs they form large spongy masses and the fruits are often so covered with them as to look like a cluster of little spikes. 1 I am indebted to Professor Buller for calling my attention to this fact. ENDOPHYLLUM SEMPERVIVI 53 Within these secidia the aecidiospores are produced ; these will only infect the Juniper, on which they begin the cycle again. For all these species of Gymnosporangium the only remedy is to remove and burn the diseased Juniper, if it can be found ; if it may not be destroyed, at least the affected branches should be cut off, and the wounds dressed with Stockholm tar. It is of no use to spray or otherwise treat the Hawthorn or Pear. In them the disease is purely local ; it comes to an end when the summer ends, and will not recur next year unless fresh infection is conveyed from the Juniper. The harm done to them is confined to the loss of the foliage which naturally weakens the tree to some extent. Endophyllum Sempervivi. The Houseleek Rust. This parasite attacks the common Houseleek and numerous other species of Sempervivum. It differs from nearly all the other Uredinales in having only spermatia and aecidiospores, the latter functioning also as teleutospores and producing basi- diospores. This fungus has been thoroughly investigated by Hoffmann (1911) from whom the following account has been derived. The aecidiospores mature on the leaves in April and May ; they have no visible germ-pores. They germinate at once, while still in the secidium ; the germ-tube forces its way out at some point of the circumference of the spore and elongates to form the four-celled basidium. Each basidium produces four basidiospores on long sterigmata; occasionally more are pro- duced— Hoffmann observed as many as eight on one basidium. The basidiospores may be blown on to the leaf of a Houseleek where they germinate at once and bore through the cuticle ; they form a holdfast (somewhat as a uredospore does) below the outer wall and penetrate into the epidermal cell (see Fig. 17). The mycelium then branches, passes through the intercellular 54 ENDOPHYLLUM SEMPBRVIV] spaces (sending haustoria into the cells) until it reaches the base of the leaf: thence it penetrates into the axis and so up to the growing point, where it hibernates till the following year. In the spring it grows on into the freshly formed leaves which become yellow and longer and more erect: on these, on b«»th sides, spermogones appear in March and April, followed by secidia (Fig. 32) which repeat the cycle. The affected plants are easily recognised by the different attitude of the leaves, which imparts an unusual irregularity to the rosette (Fig. 33). The most interesting point about this species is that established by Hoff- mann, that the secidiospore-chain arises in the way already described for Puccinia Garicis from a cell produced by the fusion 'of two adjacent cells of the spore- bed, after the manner described by Christman except that the conjugating cells were not situated in any definite plane. The binucleate secidiospores then became uninucleate by the fusion Fig. 32. iEcidia of En- dophyllum on leaf of Si mpervivum monta- num (reduced). Fig. 33. Two plants of Sempervivum, one (left) affected by Endophyllum Sempervivi, the other not. of the conjugate nuclei. On germination, when the fusion- nucleus divides into four, the first division shows slight differences from the others so as to make it certain that it is the reducing division. CRONARTIUM RIBICOL A 55 The life-cycle may thus be represented as follows : i Basicliospore Mycelium .Ecidium Spermogone Female cells Spermatia Fusion-cell Gametophyte J (n generation) Sporophyte (2/> generation) Spore-mother-cell JScidio-teleuto- spore Intercalary cell o o o o o o Basidiospores This life-history is especially worthy of consideration because it probably represents that which obtained at the first evolution of the higher Uredinales. The various types of development, seen in the genera Puccinia, Uromyces, etc., and described under the names Eupuccinia, Micropuccinia and so on, may all be derived from this original form. See Grove, New Phytologist, 1913, p. 89. Cronartium ribicola. The Rust of Currants and Five-leaved Pines. This disease, called the Currant Rust in one stage, and the Weymouth Pine Blister Rust in the alternate stage, can do enormous damage in the second phase ; it threatens in places to put a stop entirely to the cultivation of the Weymouth Pine. It has been imported into England and the United States with young trees of the latter from the continent of Europe. The remedy is to inspect Weymouth Pines, in nurseries and plan- tations, annually, cutting down and burning all those that show infection, and to remove all currant bushes from their neighbourhood. A currant-free belt, 300 — 500 ft. wide, is considered sufficient for security. The teleutospores are produced on leaves of various kinds .-,(! «'1:oxai:tiim ribk OLA of Ribes (Fig. 34); they have been recorded on 2(j < »ut of about 50 known species. The speroiogones and secidia are formed on stems and branches of the five-leaved species of Pinus: they have been found on five out of the eighteen Pines of that group, bul do not attack species having 2 or 3 leaves in a fascicle The following account is founded on that of Spaulding (1911). The h.oidiospores are formed about the beginning of August . and if they are blown by the wind, and adhere to moist young branches of the Pine, the germ-tube enters and products ;i mycelium which lives in the branch for several years, ultimately causing it to become considerably swollen in a fusiform or irregular manner. In about half the cases it is the main trunk Fig. 34. Cronartium ribicola. a, a spore of Peridermiwm Strobi, x 60o ; b, the teleutospore-columns on leaf of Bed Currant (reduced) ; c, a uredo- spore; d, top of a column of teleutospores, x600. that is infested. On this swollen portion spermogones appear at almost any period of the year, followed in spring by the secidia, which break through fissures in the bark ; these may be even as much as 1 cm. high, yellowish-white in colour, with orange spores. When the peridium bursts open, in an irregular manner, the spores may be carried by the wind to an}' plants of Ribes that may be near and can at once infect them. The distance to which the aecidiospores can be effectively carried is estimated to be less than 500 ft. These spores cannot infect the Pine ; but if they fall upon a moist leaf of Ribes, the uredo- pustules usually appear, on the underside, in from 10 to 20 days. These uredospores can, as usual, reproduce themselves on MELAMPSORA PINITORQUA 57 any leaves of Ribes to which they may be carried, thereby forming a means of rapidly spreading an infection which is once started. After a few weeks, brownish thread-like growths appear in the uredo-sori : these are the filaments on which the teleutospores are borne, and the latter may be found from July until the fall of the leaf, and even upon the fallen leaves. The teleutospores may germinate at once, perhaps also after a con- siderable time, but the basidiospores which they produce can only infect young branches of Pine as described above. They are distributed by the wind, but probably cannot be carried to any great distance. Owing to the long incubation period of the mycelium which will produce the secidiospores, the fungus cannot be seen on the Pine until it is at least three years old, although infection may have taken place in the seedling : by that time the leaves have naturally fallen off the part which received the infection. Though this disease does little harm to the currant, it is necessary to destroy the infested bushes, since they form a focus of infection for the Pines. Weymouth Pines that are more than 20 — 25 years old are rarely liable to attack. Melampsora pinitorqua. The Pine and Aspen Rust. The secidial stage of this Melampsora lives on young shoots of Scots Pine (Pinus silvestris) and its uredo- and teleuto-stages on leaves of Aspen (Poptdus tremula). The teleutospores germinate after a winter's rest, and the basidiospores infect the young pine-buds, just beginning to elongate in May and June. The mycelium produced penetrates the cortex, and reaches also the bast and medullary rays. On these shoots the spermogones appear about the end of May, and are followed by the secidia. The cortex of the affected part becomes orange and dead, while the unaffected part still con- tinues to grow. Thus thin shoots may be killed altogether, but in the thicker ones curvature takes place owing to the one-sided growth : the negative geotropism of the growing 58 MEL LMPSORA I'IMTi > K < t> I " A point, combined with this lateral curvature, causes S-shaped
  • turti<.ns which have given rise to the name "Pine Branch Twist" for the disease (Fig. 35). It is suggested by Massee that the secidiospores can repro- duce the secidia ;inn the Pine u ii limit reference to the alternate hosl . bul do proof is given of i his statement. The mycelium is almost certainly perennial in the affected branch, and thus fresh outbreaks arise year by year more or less, according to the weather in the spring. If the a-cidiospores are blown on to a leaf of Aspen, they germinate there and the mycelium produces uredospores during Fig. 35. Melampsora pinitorqua (from a German specimen, ex herb. Sydow). a, a young shoot of Pine, in June, with newly-formed leaves, showing three ca?omata (jixi,ft sta^i- <»f evolution was the separation of this spore-form into two, one (the secidiospore) germinating conidially, the other (the teleutospore) following it and germi- nating basidially : types approximating to this stage are seen in the section Pucciniopsis. It is quite certain that uredospores are only modified aecidiospores, formed as a mere multiplying device without the intervention of another fusion-cell. The peridium which is found in these later stages of evolution round the aecidium was at first represented (doubtless even in the primitive Endophyllum) by a mere circle of paraphyses or not at all. From a cytological point of view, the fusion of the two nuclei in the teleutospore may be taken as paralleled by the similar fusion in the basidium of the Basidiomycetes; the division into four basidiospores follows in both cases, although the mechanism is different. If the view propounded in a previous chapter is adopted, that the four cells of the "basidium" of the Uredinales are the real tetraspores1 and the basidiospores are merely conidia whose function is to facilitate dispersion by wind, it will be seen that the difference in the Basidiomycetes consists in the fact that cell-walls are not formed round the tetraspores previously to the production of conidia. This may recall the fact that in the Red Algae the four spores in a tetra- sporangium are also not surrounded by cell-walls before their discharge into the water. Of course, in the subaerial Uredinales and Basidiomycetes such naked masses of protoplasm would be comparatively ineffective for propagation, and are here replaced by methods more suitable to a land environment. The throw- ing off of the basidiospores with a jerk appears to be the same in both these groups. A similar comparison with the Ascomycetes cannot be made with equal advantage, until the students of that group of Fungi have come to some semblance of agreement as to the actual course of its cytological history. But it is impossible to over- look the remarkable parallelism between the cytology of the 1 In the Himalayan Barclayella, which is placed among the Melampsor- aceae ("?), these tetraspores are said to round themselves off and separate, apparently as the normal mode, without forming basidiospores. PHYLOGENY 81 Uredinales, as now known, and that attributed by Claussen (1912) to Pyronema confluens. In his paper, which is a con- clusive reiteration and confirmation of his earlier work, he shows that the numerous male nuclei of the antheridium enter the ascogonium, and in it pair with the numerous female nuclei, but without fusing with them. These synkarya then pass out into the ascogenous hyphae, and there multiply by numerous conjugate divisions. Finally a pair of descendants of these nuclei are seen in the young ascogenic-cell, one being male and the other female : here they divide conjugately into two pairs, one pair being the ascus-nuclei, and the other pair reserve-nuclei which may repeat the process in several ways. The two non-sister ascus-nuclei fuse ; then the fusion-nucleus divides, the first division being heterotypic (meiotic, reducing, possessing synapsis and diakinesis stages) and the two following ones, by which eight spores are formed, being homotypic. There is thus in the life-cycle a single fusion, followed by a single reduction. The ascus is a spore-mother-cell, comparable to the teleutospore of the Uredinales, but forming an octad, not a tetrad of spores. The two "reserve-nuclei," left after the formation of the ascus, answer to the two nuclei left in the " basal " cell of the ajcidium. Compare in this respect especially the process as it takes place in Endophyllum. The sporophyte generation consists then in Pyronema only of the ascogenous hyphce, whose cells contain the diploid number of chromosomes though arranged within two nuclear membranes. In certain species of Laboulbenia (Faull, 1912) there is a similar cytological history. The ascogenic hyphse contain two nuclei which divide homotypically by conjugate division, and two non-sister nuclei pass into each ascus where they fuse ; the two left in the ascogenic cell may repeat the process. The fusion-nucleus of the ascus divides to form eight nuclei of which four soon degenerate : the first division is meiotic and the others homotypic. There is no double fusion in this group and the same statement may justifiably be inferred to be true of other Ascomycetes. On the other hand, Harper (1900), Blackman, Welsford, Eraser, Brooks, Carruthers (1911) and others, maintain that in o. u. 6 B2 PHYLOGENY Pyronema confluens and in many other members of the Asco- inycetes there is a double fusion (one in the ascogonium and one in the young ascus), followed by a double reduction in the ascus during t ho format ion of the eight spores, the first division being meiotic, I he sec ! homotypic, anil the third brachymeiotic. In view of the established relationship between the Ascomycetes on the one hand and the Uredinales and Basidiomycetes on the other, this idea seems to be very unlikely. If correct, the double process is a special development, peculiar to some only of the Ascomycetes. The matter can only be decided by fresh investigations, but it seems in all probability that the hype- thesis of a second fusion and subsequent brachymeiosis is the result merely of a misinterpretation of the observed phenomena. According to Lutman (1910). in the Ustilaginales most of the cells of the mycelium are binucleate,but the perfect resting spores are always uninucleate, as are the cells of the basidia. Rawitscher (1912) says the same, and adds that the con- jugate condition arises (according to the species) by the anastomosis in pairs either of the basidium-cells, or of the basidiospores, or of the cells of the mycelium produced by them, and the passage of the nucleus of the one cell into the other to form a synkaryon. Finally, it may be pointed out that the ideas embodied in the foregoing discussion are in harmony with the now generally accepted doctrine of the polyphyletic origin of the Fungi, by which it is assumed that their various groups are not derived from one or two ancestors, but originated separately from distinct sub-divisions of the Alga?, much in the same way in which (on a smaller scale) the non-chlorophyllose Phanerogams have arisen from various orders or families of Flowering Plants. From this point of view, according to which the vast majority of the Fungi originated from the Red Alga-, it is not without significance that already some of that group are known which (though still rightly classed as Algse) have assumed a true holo- parasitic habit — a statement which cannot be made to the same extent, if at all, about other algal groups. Examples are found in the well-known Harveyella mirabilis (Sturch, 1899) and in Clioreocolax Polysiphoniae (Richards, 1891). PHYLOGENY 83 From these considerations the probable phylogeny of the Uredinales may be represented as in the following schemes. Ascomvcetes Basidiomvcetes Uredinales Ustilaginales Rhodophyceae Pucciniacese Endophyllum Coleosporiaceae Cronartiaceae (Endophyllum-like ) Melampsoraceae Primitive Uredine 6—2 84 manipulation Note on Manipulation. I )ri«(l specimens of Uredinales keep most of their characters unchanged for an unlimited time, but the colours fade except those of the teleutospores. The only two difficulties found in examining them arc in regard bo the markings on the outer surface of the spores, and the number of germ-pores. For the first, different methods succeed in different cases, but the finer markings can usually be seen by examining the spores under a one-sixth inch in air. or in water after squeezing ou1 their granular contents by tapping or pressing hard upon the cover- glass. For the second, boiling for about a minute in a drop of lactic acid, on a glass slide over a spirit-lamp, is the best course, although expulsion of the contents under pressure frequently brings the germ-pores into view: in fact so plain do they often become that they can be photo-micrographed with ease. Boiling in lactic acid also restores old collapsed spores to their former size and plumpness. The preparation from which Fig. 37 was drawn was obtained in the following way: lav the side of the leaf opposite to the sori in a thin layer of 5 , KOH solution for an hour or so, then reverse and brush or scrape away the softened tissue as far as possible ; on mounting the remaining surface, epidermis upwards, in glycerine and water, the arrangement of the parts can be clearly seen. If it is wished to observe the germination of the spores in a hanging drop, almost any uredo will serve; for teleutospores Puccinia Malvacearum and for secidiospores sEcidium Ficariue are usually the most handy. One of the best double stains to use is Diamant Fuchsin and Light Green; the former stains the nuclei red and the latter the cell-walls green. Stain heavily with the former and wash out with alcohol till the desired tint is arrived at; then use the Light Green dissolved in clove oil. UREDINALES A group of Fungi which are obligate parasites on ferns and the higher plants. Mycelium filamentous, branched, septate; developed within the tissues of the host, producing teleuto- spores (resting spores, chlamydospores) which on germination give rise to a generally four-celled "basidium," each cell of which may in turn produce, on a sterigma, a single basidio- spore (conidium). In addition, there are often produced sperrnatia (in spermogones), recidiospores (in gecidia), and uredo- spores (in sori). PUCCINIACEiE Puccinie/E. Teleutospores of one or two cells, ( scarcely gelatinous (except the pedicels in some ■< " . o N \Puccmva. foreign species). \ PHRAGMiDiEiE. Teleutospores of more than two cells, the walls of the pedicels subgelatinous. f Tripfiragmium. Teleutospores more or less verrucose. \ n, ... 1 { L hragmiaium. _ , , , f Kuehneola. leleutospores nearlv or quite smooth. A ^ 7 , Gymnosporajstgie^e. Walls of pedicels of teleuto- [ spores becoming highly gelatinous. {Gymnosporangium. UROMYCES Link. Autoacious or heteroecious. Spermogones deeply embedded in the tissues of the host, flask-shaped with conical mouth and ostiolar filaments. ^Ecidia with an evident, usually cup-shaped peridium ; a?cidiospores with indistinct germ-pores. Uredospores formed singly on their pedicels, with several usually rather distinct germ-pores which are often surrounded by a thickened border, rarely accompanied by paraphyses. Teleutospores one-celled, on dis- tinct pedicels, almost always with an apical germ-pore. Basi- diospores flattened on one side or kidney-shaped. 86 UROMYCES The species are arranged according to the families to which the hosts belong: sec I'nrr'uiia. This genus is often considered the must highly (a1 least the latesl I evolved of the Uredinales ; ltut rather it forms a heterogeneous group, the species of which have arisen at different times from various species of Puccinia. 1. Uromyces Valerianae Fckl. Uredo Valerianae Schum. PL Sail, ii. 233. /Ecidium Valerianearum Duby, Bot. Gall. ii. 908. Cooke, Handb. p. 540; Micr. Fung. p. 196. h-i-t/tlii'ii Yuli -ri(inii<: Berk. ; ('coke, Handb. p, 532 : .Micr. Fung. p. 222. Uromyces Valerianae Fckl. Symb. Myc. p. 63. Plowr. Fired, p. 12*. Sacc. Syll. vii. 536. Sydow, Monogr. ii. 19. Fischer, Ured. Schweiz, p. 54, f. 41. Spermogones. Epiphyllous, in small clusters, honey- coloured, turning black. Mddiospores. iEcidia hypophyllous, and often on the nerves, petioles and stalks, seated on pale thickened spots, densely aggregated or circulate, cup-shaped, whitish-yellow: margin revolute and torn; spores covered with minute crowded warts, yellow, 18 — 25 x 16 — 20 (jl. Uredospores. Sori amphigenous, usually on indefinite yellow spots, scattered or aggregated here and there, minute, punctiform, pulverulent, brown; spores globose to broadly ellipsoid, verrucose-echinulate, yellowish-brown or brown, 21 — 28 //. ; epispore 1\ — 3 ^ thick, with two or three germ-pores. Teleutospores. Sori similar, but longer covered by the epidermis, dark-brown; spores ellip- soid or ovate, with a flat subhyaline papilla at the summit, smooth, pale clear-brown 20 — 30 x 13 — 21 /j,; epi- spore thin, scarcely thickened above ; pedicels short, thin, hyaline, rather „. .)0 „ 17 , . rp , deciduous. Fig. 88. U. Valerianae. Teleu- tospores and uredospore (the . . ,_ ~ . latter viewed dry) on V. offici- On Valeriana dwica, V. qffi-ci- ""hs- nalis. iEcidia in May and June ; uredospores from June, teleutospores from July to October. Common. (Fig. 38.) UROMYCES 87 The uredospores seem to be variable in their markings ; some are distinctly verrucose with pointed warts ; others are as distinctly echinulate. Distribution : Europe and South Africa. 2. Uromyces Scrophulariae Fckl. JEtidium Scrophulariae DC. ; Cooke, Handb. p. 544 ; Micr. Fung. p. 199. Uromyces Scrophulariae Fckl. Syrnb. Myc. p. 63. Plowr. Ured. p. 139. Sacc. Syll. vii. 559. Sydow, Monogr. ii. 27. Fischer, Ured. Schweiz, p. 75, f. 56. V. concomitans B. et Br. ; Cooke, Micr. Fung. p. 213. Spermogones. Few, singly or in little groups, simul- taneously with the secidia. JUcidiospores. zEcidia hypophyllous or on the stems, on yellowish spots, in rounded clusters or in more or less elongated patches on the nerves and stems, cup-shaped, yellowish ; margin involute, entire ; spores verruculose, smooth below, yellowish, 18—21 x 14— 18 /*. Teleutospores. Sori small and roundish, arranged like the secidia except that they form more elongated groups (as much as 10 cm. long) on the stems, long- covered by the lead-coloured epider- mis, at length naked and pulverulent. dark -brown ; spores very irregular, obovate, fusiform, or ellipsoid, angu- lar, rarely sub-globose, apex rounded, truncate or slightly pointed, some- what thickened (up to (5 /j,), with tenuated below, smooth, brown, 18 — 35x11 — 18^; pedicels persistent, hyaline or yellowish, nearly as long as the spore. On leaves, petioles and stems of Scroph ularia aquatica, S. nodosa. July — -September. Not common. (Fig. 39.) The spots on the leaves are pallid, edged with violet-brown. The teleutospores especially cause considerable distortion of the leaves and stems. The two kinds of spores may be produced on the same mycelium, and the secidia and teleuto-sori can occur simultaneously and intermixed, or the latter surrounding the former (Grevillea, iii. 181, pi. 36). For this Fig. 39. U. Scrophulariae. Teleutospores on S. aquatica. a dark-coloured cap, at- 88 UROMYCES is one of the species in which it is stated Dietel, Flora, L895, Ixxxi. 396), tli.it the secidiospores can reproduce the secidia. Spermogones are found sparingly only with the first generation of the secidia and at the same time. The secondary secidiospores, in fact, take the place of uredospores. Distribution: Europe generally. 3. Uromyces Limonii LeY JEcidmm Statices Desm. ; Cooke, Micr. Fung. p. 197 ; Grevillea, i. 7. Uromyces Limonii Lev. Diet. Hist. Art. Uml. p. 1!). Cooke, Bandb. p. 518; Micr. Fung. p. 212. Plowr. Ored. p. 122, p.p. Sacc. Syll. vii. 532 p.p. Sydow, Monogr. ii. 41. sEcidiospores. iEcidia araphigenous, often on red or brownish spots, in roundish clusters or elongated along the nerves, usually shortly cylindrical, whitish, with a torn margin : spores densely and minutely verruculose, yellowish, 21 — 32 x 18 — 26 fi. Uredospores. Sori amphigenous, scattered, generally round- ish or, on the stem, oblong, long covered by the epidermis, at length naked, pulverulent, cinnamon; spores varying from globose to oblong, densely verruculose with minute- papillae, yellowish-brown, 22— 32 x 20— 28 fi; epispore 1£— 2£/t thick, with two or three germ-pores. Teleutospores. Sori amphigenous or caulicolous, scattered or circinate, roundish or oblong, Long covered by the epidermis, pulvinate, black ; spores subglobose or more frequently oblong or clavate, sometimes rounded, sometimes at- tenuated at the apex, where the wall is up to 10 fi thick, attenuated below, smooth, brown, 24 — 50 x 14— 25 /jl: pedicels as much as 80 /x long, thick, pale-brownish, persistent, On leaves and stems of Statice Limonium. Not common. /Ecidia in June and July; uredo- and teleutospores from July to October. (Fig. 40. ) Fig. 40. U. Limonii. Teleuto- spores and uredospore. ON PLUMBAGINACE^E 89 This species was formerly united with C. Armeriae (q.v.), but the teleutospores are distinctly different. Distribution : Europe, North Africa, Siberia and North America. 4. Uromyces Armeriae Lev. Caeoma Armeriae Schlechtd. Fl. Berol. ii. 126. Uromyces Armeriae Lev. Ann. Sri. Nat. ser. 3, viii. 375. Sydow, Monogr. ii. 40. Fischer, Ured. Schweiz, p. 52, f. 39. U. Limonii Plowr. Ured. p. 122 p.p. Sacc. Syll. vii. 532 p.p. Spermogones. Scattered among the a?cidia, honey-coloured. sEcidiospores. yEcidia amphigenous, scattered or in small clusters, at first hemispherical, then cup-shaped, with a whitish incised margin ; spores densely. and minutely verruculose, yellow, 17— 28x 16—22^. Uredospores. Sori amphigenous, sometimes on purplish spots, rounded or elongated, surrounded or half-covered by the cleft epidermis, pulverulent, cinnamon ; spores globose to oval, very densely and minutely verruculose, yellowish-brown, 24 — 32 x 21 — 28 /a ; epispore 2| — 3 /x thick, with two or three germ-pores. Teleutospores. Sori similar, dark-brown ; spores globose to ovate, rounded and thickened (7 /jl) at the apex, with a broad flat cap, usually rounded below, smooth, brown, 24 — 36 x 21 — 32 /x : pedicels hyaline, nearly as long as the spore, seldom persistent. On leaves and peduncles of Armeria maritima. Not uncommon. iEcidia in May and June ; uredospores from June onwards ; a few teleutospores begin to appear in the uredo-sori towards the end of July. (Fig. 41.) This species was united by Plowright with U. Limonii, but is distin- guished by the more readily pulverulent sori, the shorter and broader teleutospores, and the shorter hyaline pedicel which is easily detached. The distinctness of the two species does not seem, however, to have been tested by experimental cultures. Though the uredo- and teleutospores Fi<^. 41. U. Armeriae. Teleutospore and uredospore. 90 I'koMYCES e occurred for many years consecutively on Thrift in my garden, I have never noticed the secidia ; the uredospores lasl through the winter mi the evergreen leaves, and reproduce the Pungus about June; teleuto- spores are rather scarce. Distribution: Central and North-Western Europe. 5. Uromyces Trifolii LeV. Puccinia Trifolii Hedw. )'. in I"'. Flor. fr. ii. 225. Uromyces Trifolii Lev. Ann. Sci. Nat. ser. 3, viii. 371. Plowr. [Jred. p. 12 1 p.p. Sacc. Syll. vii. 534 p.p. Sydow, Monogr. ii. 132. Fischer, Ured. Schweiz, p. 23 p.p. Trichobasis fallens Cooke, Micr. Fung. p. 226 p.p. /',/.////,/ fallens Cooke, Handb. p. 508 p.p. Nigredo fallens Arthur, X. Amer. Fl. vii. 2">4. Uredospores. Sori hypophyllous and on the petioles, and like those of U. Trifolii-repentis ; spores differing in having 5 — 7 germ- pores. Teleutospores. Indistinguishable from those of U. Trifolii-repentis. On Trifolium hybridum, T. incarna- f a in, T. medium, T. pratense. Not com- mon. (Fig. 42.) Lire proved by culture-experiments that the Uromyces on Trifolium repens could not be transferred to T. hybridum or T. pratense. Since this difference is accompanied by the absence of the secidium in the latter species and by a difference in the number of the germ-pores, they are considered distinct by Sydow. But on Trifolium i>rntr, ise I have found uredospores with not more than four germ-pores, each covered with a low- hat hyaline cap. In that case only the absence of the secidium would separate the two forms, though the average number of germ-pores is no doubt different in the two cases. An secidium has been found elsewhere on T. pratense, but this has been experimentally proved (Dietel, Flora, 189"), lxxxi. 398) to belong to another European and North American species, U. minor Schrot. = U. ob- longus Vize (Grevillea, v. 110), which has no uredospores, but only a'cidio- and teleutospores. Cooke's species, U. apictdatus (Grevillea, vii. 136), is indefinite ; the form on clover may belong here, that on Lathy rus prate asis to U. Pisi. 1 Mstribution : Europe, Asia Minor, Persia, North America, etc. Fig. 42. U. Trifolii. Uredospores on T. pratense. ON LEGUMIXOS/E 91 (!. Uromyces Trifolii-repentis Liro. U. Trifolii-repentis Liro, Act. Soc. Faun. Flor. Fenn. xxix. 15. Sydow, Monogr. ii. 131. Fischer, Ured. Schweiz, p. 23, f. 19. U. Trifolii Plowr. Ured. p. 124 p.p. Sacc. Syll. vii. 534 p.p. McAlpine, Rusts of Australia, p. 97, f. 142, & pi. G, f. 32. Trichobasis fallens Cooke, Micr. Fung. p. 226 p.p. Puccinia fallens Cooke, Handb. p. 508 p.p. Nigredo Trifolii Arthur, N. Amer. Fl. vii. 255. Spermogones. Epiphyllous, honey-coloured, forming minute clusters. JEcidiospores. iEcidia hypophyllous, in clusters, roundish on the leaves and as much as 5 mm. long on the nerves and petioles, shortly cylindrical, whitish-yellow; margin white, torn, hardly revolute; spores minutely verruculose, yellowish, 17 — 21 x 14— 18 fi. Uredospores. Sori hypophyllous and on the petioles, scattered over the leaves or gregarious, small or rarely confluent and larger, soon naked, pulverulent, pale-brown ; spores globose, ovate or ellipsoid, echinulate, yellow-brown, 19 — 26 x 17 — 24 /m ; epispore about Hyu, thick, with two to four (generally two) equatorial germ-pores. Teleutospores. Sori surrounded by the cleft epidermis, similar, but elongated on the petioles, and darker brown; spores globose to ovate, rounded at the apex, with a very small hyaline papilla, smooth or at times bearing a few minute warts arranged more or less in lines, brown, 18—30 x 16—25/*; epispore about 2fi thick ; pedi- cels short, thin, hyaline, decidu- ous. Fit;. 43. U. Trifolii-repentis. a, secidia on young leaf ; b, urerto- aml teleuto-sori, on later-formed leaf. On leaves and petioles of Trifolium repens. iEcidia (rare) from April and uredospores from May onwards. (Fig. 43.) This species is distinguished from U. Trifolii Lev. by the smaller number of germ-pores of the uredo, and also by the presence of the secidia, which cause long crooked swellings on the petioles and nerves, but not on 92 IKOMVCKS the leaves. The mycelium of the secidial stage is said to be perennial in tin' host; Dietel says that in some Localities tin- secidiospores can re- produce themselves, and that then the uredo is suppressed. Both this species and the preceding are distinguished from U. flectens in the fad that tlu' sori are smaller, distributed more uniformly over the leaf, and do not ( ause distortions. The a M- id i i mi is ran- in I'.ritain (I have seen specimens only from Perth : most of our records of I'romyces on T. repens belong to the following common epecies, U. flectens. Pseudopeziza Trifolii (a Discomycete) is common on leaves of white clover and is not infrequently mistaken for the uredo-stage of U. Trifolii-repentis, bui is distinguishable by its being confined to the upper surface of the leaves. No practical means of prevention are known for either the Clover Rust {Uromyees) or the Clover Leaf-spot {Pseudopeziza). DISTRIBUTION: Europe, Asia Minor, Persia, North and South America, Australia. 7. Uromyees flectens Lagerh. Uromyees flectens Lagerh. Svensk Bot. Tidskrift, iii. ."50. Sydow, Monogr. ii. 360. Grove, Journ. Bot. 1911, p. 366. Puccinia neurophila De Toni, Sacc. Syll. vii. 698. Teieutospores. Sori hypophyllous, or more often on the nerves and petioles where they cause swell- ings and distortion, scattered, rather large, ^ — 2 mm. long or even continent and larger, long covered b}* the epidermis, then pul- verulent, dark-brown ; spores as in U. Trifolii-repentis. Fig. 44. U. flectens. Teieutospores ou t. On Trifolium repens. May— October. Common. (Fig. 44.) repens, It has been frequently noticed that the Uromyees on Trifolium repens behaves differently in different localities ; sometimes forming teieutospores only, from .May to October; at others forming both secidia and uredospores during the same time. Plowright records an interesting experiment which he performed (Ured. p. 125) ; in October he brought a, plant of T. repens, with the Uromyees upon it, indoors and kept it there till the following summer. During all this time it produced only teieutospores. Lagerheim, in 1909, noticing that the form which produced only teieutospores had sori which were larger, more predominant upon nerves and petioles, and remained longer covered by the epidermis, described this as a distinct ON LEGUMINOSvE 93 species, to which evidently Plowright's specimen may be ascribed. Cooke's figure of his U. apiculosa, on Trifolium repens (Micr. Fung. pi. vii. f. 154), is probably the same species. Distribution : Middle Europe and Persia, 8. Uromyces striatus Schrot. Credo apiculata var. Trifolii Strauss, Ann. Wett. ii. 97 p.p. Uromyces striatus Schrot. Abhandl. Schles. Ges. 1872, p. II. Sacc. Syll. vii. 542 p.p. Sydow, Monogr. ii. 115. Fischer, Ured. Schweiz, p. 31, f. 24. Cf. Plowr. Ured. p. 134. As in Uromyces Pisi.] [Spermogones uEcidiospoi^es Uredospores. Sori amphigenous, scarcely ever on the nerves, without spots, scattered, occasionally aggregated and confluent, minute, pulverulent, cinnamon ; spores globose to ellipsoid, faintly and sparsely echinulate, yellowish-brown, 15 — 22 /x ; epispore lh, — 2 /x thick, with 4 — 6 or even more germ-pores, each with a small hyaline cap. Fi^\ 45. U. striatus. a, two teleutospores on T. minus (ex herb. Broome); b, a teleutospore on T. arvense (foreign, ex herb. De Thiimen); c, uredo- spoi'e from the same leaf as a. Teleutospores. Sori similar, but darker ; spores globose to ovate, with a minute and narrow papilla, striated from apex to base by longer or shorter lines of warts, brown, 18 — 24 x 15 — 20 /z; epispore 1^ — 2/i thick; pedicels short, hyaline, deciduous. [iEcidia on Euphorbia Cyparissias, not known in Britain ;] uredo- and teleutospores on leaves and stems of Trifolium minus. Bath (Herb. Broome) ; King's Norton (Worcestershire). Very uncommon. July — August. (Fig. 45.) See remarks made about the tecidial stage under U. Pisi. Schroter proved the connection of an secidium on Euphorbia Cyparissias with an exactly similar Uromyces on Trifolium agrarium. U. striatus is found elsewhere on Trifolium procumbent and also on many species of Medicago, 94 I ROM! I ES including all the British Bpecies, bul I have seen no specimens on these from this country. The teleutospores on T. minus which I have observed are more distinctly verrucose and less striated than in the figures given by Fischer, and may possibly not belong to the same species. DlSTRim ti<».\": Europe. North .-iml South America, Easl I in lii-s. !). Uromyces Loti Blytt. Uromyces Loti Blytt, Christ. Vidensk.-Se].>kal>.s !■"< >rh null. |s90, p. ?,~. Sydow, Monogr. ii. 110. Grove, Journ. Bot. 1911, p. 367. U. Eupkorbiae-Comiculati Jordi, Centralbl. f. Bakt. 1904. ± xi. 791. Fischer. Cred. Schweiz, p. 34, f. 26. [Spenitof/oiies. Hypophyllous, numerous, scattered amongst the lecidia. .Kcidiospores. iEcidia distributed uniformly over the lower surface of the leaf, cup-shaped, with a torn white revolute Fig. 46. U. Loti. Four teleutospores, all on L. eorniculatus : a shows how the spores look when wet, the others are viewed dry. margin; spores densely and minutely verruculose, orange, 18— 23 /*.] Uredospores. Sori amphigenous, but mostly hypophyllous, scattered, minute, round, sometimes confluent, surrounded by the cleft epidermis, soon naked, pulverulent, cinnamon ; spores globose to ellipsoid, with short blunt and rather distant spines, brownish, 17 — 25x16 — 23 yu, ; epispore '21 — 3 fi thick, with 2 — 5 germ-pores. Teleutospores. Sori similar, but darker in colour; spores globose to obovate, often with a low flat pore-cap at the apex, which is not thickened, beset with minute warts and ridges which are often arranged in undulating longitudinal lines, brown, 17 — 25 x 14 — 21 fi; pedicels short, hyaline, deciduous. ON LEGUMINOS/E 95 [iEcidia on Euphorbia Oyparissias ;] uredo- and teleuto- spores on Lotas angustiasimus, L. corniculatus ; July, August. (Fig. 46.) Plowright refers to this species (but not as British) in a note on p. 13-t. The markings on the teleutospore are very delicate and can scarcely be seen except when the material is fresh and the spores are viewed dry. I found that the longitudinal lines of warts were more strongly marked and anastomosed more frequently on spores from L. cor niculatus than from L. angustissimus (Xewquay, Cornwall), on which they were fainter and more irregular, but this difference may have been partly due to the fact that the Litter had been gathered (by Dr Vigurs) many years before they were examined. Jordi proved that tecidiospores from E. Cyparissias would freely infect L. comiculatus. But see U. Pisi. Distribution : Western, Central and Southern Europe, and Japan. 10. Uromyces Anthyllidis Schrot. Uredo Anthyllidis Grev. in Sm. Eng. Fl. v. 383. Uromyces Anthyllidis Schrot. Hedwig. xiv. 162. Plowr. Ured. p. 135. Sacc. Syll. vii. 551. Sydow, Monogr. ii. 64. Fischer, Ured. Schweiz, p. 36, 543, f. 28. JJredospores. Sori amphigenous, widely and irregularly scattered, or sometimes with a circle of small ones round a larger one, minute, roundish, black and shining, soon naked, then pulverulent, cinnamon ; spores globose or subglobose, sparsely and finely echinulate, yellowish-brown, 18 — 25/x ; epispore 3 — 3i jx thick, with 4 — 6 germ-pores (4 — 5, Bubak ; Anthyllidis.' 5 — 8, Fischer). Uredospore. Teleutospores. Sori similar, but darker in colour. Spores globose to ovate, with a minute papilla at the rounded apex, verrucose, brown, 16 — 22x15 — 20^,; epispore rather thick; pedicels short, hyaline, deciduous. On leaves of Anthyllis Vulneraria. Not common. June — October. (Fig. 47.) It is probable that this species occurs only on A. Vulneraria (and on the continent, A. maritima), but it has many close allies on other Leguminosse. Teleutospores are rarely formed ; in specimens gathered in mid-September I have found only one or two, in the midst of abundant 96 UROMY< ES uredospores. The warts on the teleutospures are not numerous and are rather easy to see. Distribution: North-western and Middle Europe. 11. Uromyces Ervi Westendorp. /Ecidium Ervi Wallr. Fl. Crypt. Germ. ii. 2J7. Uromyces ErviWesid. Bull. Acad. Roy. Sci. Belg. x.xi. pt. 2, p. 246, f. '■',. Plowr. Qred. p. 140. Sydow, Monogr. ii. !)<;. Fischer, Qred. Schweiz, p. 69, f. 53. /Eddiospores. zEcidia amphigenous, or on the petioles, solitary or 2 — 8 together in little scattered groups, cup-shaped, whitish: margin faintly revolute, scarcely torn ; spores densely and minutely verruculose, pale-yellowish, 16 — 25 x 14 — lunounded by the ruptured epidermis, cinnamon ; spores ovate or ellipsoid, dis- tantly echinulate, brownish-yellow, 20 — 30x18 — 22^, with two (rarely three) germ-pores. Teleutospores. Sori amphigenous, or more frequently on the petioles and stems, scattered, minute, oblong, surrounded by the ruptured epidermis, blackish-brown ; spores subglobose to obovate, usually darker and rounded above (where the wall is up to 8 /j, or more thick), rounded or attenuated at the base, smooth, brown. 20 — 28x14 — 20 /la; pedicels brownish, persistent, as long or twice as long as the spore. On leaves, petioles, and stems of Ervum hirsatum ( Vicia hirsuta). /Ecidia, May — October; teleutospores from July onwards, lasting through the winter on the dead stems. (Fig. 48. ) It has been proved by many culture experiments that Plowright was correct in his belief that this species is strictly confined to the one host. The secidiospores ait- capable of reproducing the secidium and are found throughout the season ; the uredospores are, perhaps in consequence, not abundant, only a few being occasionally found and usually intermixed with teleutospores. Distribution: Europe, Japan. Fig. 48. U. Ervi. Leaf of E. hirsutum, with fficidia, slightly en- larged ; two teleuto- spores. ON LEGUMINOS/E 97 12. Uromyces Fabae De Bary. Uredo Fabae Pers. in Rom. Men. Magazin, i. 93. Uromyces Fabae De Bary, Ann. Sci. Nat. sor. 4, xx. 72. Plowr. Ured. p. 119. Sacc. Syll. vii. 531 p.p. Sydow, Monogr. ii. 103. Fischer, Ured. Schweiz, p. 65, f. 49 — 51. McAlpine, Rusts of Australia, p. 93, f. 307. Trichobasis Fabae Cooke, Handb. p. 508 ; Micr. Fung. p. 225. Uromyces appendiculatus Lev. ; Cooke, Micr. Fung. p. 212, pi. vii. f. 149—150 p.p. Puccinia Fabae Link, referred by Cooke to this species, has no existence in nature (Handb. p. 508; Micr. Fung. p. 211). Spermogones. Hypophyllous, growing among the secidia. jEcidiospores. iEcidia hypophyllous, seated on pale-yellow spots, solitary or in small round or elongated clusters, shortly cup-shaped, with a whitish, torn, revolute margin ; spores densely and minutely verruculose, yellow, 14 — 22 p. Uredospores. Sori amphigenous, scattered or circinate, girt by the ruptured epidermis, minute, pulverulent, pale-brown ; spores globose to ovate, distantly echinulate, at length pale- brown, 20 — 30 x 18 — 26 fx\ epispore 1^ — 2iyu, thick, with three or four germ-pores. Teleutospores. Sori similar, but per- sistent and darker or blackish-brown ; spores subglobose to obovate, rounded or truncate and thickened above, where the wall is dark and 7 — 11^ thick, sometimes with a colourless papilla, smooth, brown, 25 — 38x18 — 27 /* ; pedicels brownish, persistent, thick and as much as 40 — 70 /x long. ( )n leaves and stems of Faba vulgaris, Lathyrus pratensis (?), Pisum sativum, Vicia Cracca, V. sativa, V. sepium. iEcidia in April, May; uredospores from May, teleutospores from July onwards, lasting through the winter on the dead stems. (Figs. 49—52.) One of the most widely spread of the Uredinales, occurring in every o. u. 7 Fig. 49. U. Fabae. Teleutospores on stem of Broad Bean. 98 DROMYCES pari oi the world ; reported on raanj Leguminosae, but doubtless some of these are distinct Bpecies. Jordi has distinguished under U. Fabae three Pig. 50. U. Fabae. 'IVleutospores and uredospore on Vicia Cracca. Fig. 51. U. Fabae. Telento- spores and uredospore on Vicia sepivm. "biological" races— (1) on Faba vulgaris and Pi&um sativum, -1 on Lathyrus vernus and probably also on Pisum sativum, (3) on Vicia Cracca, Pimm sativum, and possibly also Vicia kirsuta. The secidial generation is frequent on some hosts, such as Vicia sepium and species of Lathyrus. On Faba vulgaris and Pisum sativum it is, on the contrary, very rare, being recorded by Sydow, on the Pea, only from Norway, East Indies and Japan 'Once from cadi . It has been seen on the Bean in the East Indies, 1 >ut seems to be not uncommon in artificial cultures, in which Plowright produced it both on Pea and Bean from the same teleutospores (Plowr. Ured. p. 121). On the leaves of the common field Bean only the uredospores are generally to be found, even as late as mid-October, but on the stems the teleuto- spores form large black sori. On Vicia sepium the uredo-sori are often darker and covered by the epidermis for a shorter time than on the Bean, while the teleuto-sori occur in great abundance on the leaves and even on the tendrils. If all the infected haulm, etc., were burnt instead of being put on the manure heap or left to rot on the ground, the disease would become less prevalent, especially if Jordi's idea is true, that the Bust on the wild Vetches is a distinct biological race. In Ecuador, at Quito, which has a very equable climate of "perpetual spring/' U. Fabae has, according to Lagerheirn, become almost an isolated uredo. The same thing is true of it in other tropical climes. Distribution : world-wide. Fig. 52. U. Fabae. iEcidia on Tea, from Plowright's culture, in which he produced them ou both Pea and Bean from the same teleutosport s. ON LEGUMIXOS.i; 99 18. Uromyces Orobi Lev. JScidiwm Orobi Pers. in Rom. Neu. Magazin, i. 92. Cooke, Handb. p. 542 ; Micr. Fung. p. 1!>7. Uromyces Orobi Lev. Ann. Sci. Nat. ser. 3,'viiL 371, 376. Plowr. Ured. p. 121. Sydow, Monogr. ii. 106. Fischer, Ured. Schweiz, p. 69, f. .)!' Spermogones. Hypophyllous, mixed with the secidia. jEcidiospores. /Ecidia hypophyllous, on yellowish spots, in dense clusters 1 — 5 mm. long, rarely soli- tary, shortly cup-shaped, with whitish revo- lute margin ; spores densely and minutely verruculose, yellowish, 14 — -21 /x. Uredospores. Sori amphigenous, scat- tered, minute, punctiform, pulverulent, brown ; spores globose to ovate, distantly echinulate, pale-brown then darker, 20 — 28 x 18 — 25 /x: epispore 3 — 4/u, thick, with three or four germ-pores. Teleutospores. Sori similar, but darker; spores subglobose to ovate, rounded or sub- conical above, and much thickened (7 — 11 /x) and darker, smooth, brown, 25—35 x 18 — 28 ft; pedicels persistent, yellowish, thick, as much as 100 n long. On Orobus tuberosus (Lathyrus montanus = L. macrorrhizus). iEcidia in May and June ; uredo- and teleutospores from June onwards. (Fig. 53.) Cooke says that the secidia also occur on the stems. The thicker membrane of the uredospores, which is also less strongly echinulate, distinguishes them from those of U. Fabae. Jordi made attempts to infect various other species of Lathyrus, and also species of Vicia and Pisum, from U. Orobi, but in every case without success. Distribution : North-western and Central Europe. Fig. 53. U. Orobi. Teleutospores on 0. tuberosus. 14. Uromyces Pisi Wint. Uredo appendiculata var. Pisi Pers. Obs. Myc. i. 1' JEcidium Cyparissiae DC. Flor. fr. ii. 240. Uromyces Lathyri Fckl. Symb. Myc. p. 62. 7—2 00 UROMY< ES U. Pm Winter, Krypt Flor. i. L63. Cooke, Grevillea, vii. L35. Plowr. [Jred. p. 133. Sacc. S\]]. vii. 542 p.p. Sydow, Monogr. ii. L24. Fischer, [Jred. Schweiz, p. 28, f. 22. Spermogones. Hypophyllous, numerous, scattered amongsl the a-cidia. /Ecidiospores. ^Ecidia distributed uniformly over the lower surface of the leaf, cup-shaped, with ;i while, torn, broadly revolute margin; spores densely and minutely verruculose, orange, 18 — 23 /x. irrcf/')sj)nre.s. Sori generally hypophyllous, scattered, minute, soon naked, pulverulent, cinnamon; spores globose or subglobose, minutely verruculose, yellow-brown, 21 — 25 fi diam. ; epispore 1^ — 2h /j. thick, with 3 — 5 germ-pores. Teleutospores. Sori similar, but sometimes confluent and larger, dark-brown; spores sub- globose to ovate, with a small hyaline papilla (as much as 3 p. high), everywhere minutely and rather densely verruculose, brown, 20 — 2^1x14 — 22 yu,: epispore 1£ p, thick ; pedicels hyaline, short, deciduous. .Ecidia on Euphorbia Cyparissias, May, June; uredo- and teleutospores on Pisum sativum and Lathyrus pratensis, July — September. Rare. (Fig. 54.) Although both will equally infect E. Cypurissias, it is probable that the Uromyces on Pisum is biologically distinct from that on Lathyrus. It is not certain that the latter has been found in this country, but the former is recorded from various places. It must be remembered that U. Fabae occurs also on the same two genera, though all the spore-forms of the two can be easily distinguished. An secidium on /:'. Cyparissias, and attributed to U. Pisi, was found at Dover, May, 1909 (Rev. T. Taylor) ; the specimen is in the British Museum, but there is no proof that this belonged to U. Pisi, because it has been shown that U. Loti, U. striatus (both of which are British), as well as two other (non-British) species, equally produce on E. Cyparissias secidia which are morphologically indistinguishable. This secidium pos- sesses a perennial mycelium, which permeates the whole host and deforms and bleaches it. The connection of one form of it with Uromyces Pisi has been experimentally demonstrated by Schroter, Rostrup, Fischer and others. The Uromyces on Vicia Oracca which was formerly considered to FiK. 54. U. Pisi. a, a cidia and s, spermogones on leaf of Euphor- bia ( 'yparissias (from the Dover specimen). ON LEGUMINOS.K 101 belong to U. Pisi has been proved by Jordi to be confined to that species and not to be transmissible to Pisum sativum or Lathyrus. It has been named by Magnus U. Fischeri-Edv&rdi, but is not known as British. Distribution : Europe generally ; North America less commonly. 15. Uromyces Phaseolorum De Bary. sEcidium Phaseolorum Wall. Fl. Crypt. Germ. ii. 256. Credo appendiculata var. Phaseoli Pers. Syn. p. 222. Uromyces Phaseolorum De Bary, Ann. Sci. Nat. ser. 4, xx. 80 (1863). Cooke, Grevillea, vii. 135. U. Phaseoli Wint. Pilze, p. 157 (1884). Plowr. tired, p. 122. York- shire Fung. Fl. p. 186. U. appendiculatus Link, Obs. ii. 28. Sacc. Syll. vii. 535. Sydow, Monogr. ii. 120. Fischer, Ured. Schweiz, p. 19, f. 16. McAlpine, Rusts of Australia, p. 92, f. 306 (all pro parte). Nigredo appendiculata Arthur, N. Amer. Fl. vii. 257 p.p. [Spermogones. In little clusters, whitish, then yellowish. ^■Ecidiospores. iEcidia hypophyllous, clustered in little roundish groups 2 — 3 mm. wide on yellowish or brownish spots, cup-shaped, whitish, with a torn revolute margin ; spores poly- gonal or oblong, densely and minutely verruculose, colourless, 18—36 x 16—24 /x.] Uredospores. Sori generally hypophyllous, on indistinct spots, scattered or in little clusters here and there, minute, soon naked, surrounded by the cleft epidermis, cinnamon ; spores subglobose to ovate, distantly but sharply echinulate, brownish-yellow, 18 — 28 x IN — 22 fx ; epispore brownish-yellow, about 1 \ fi thick, with two germ-pores ; contents Fig gg v PJuueolorum. colourless. Teleutospore and ure- Teleutospores. Sori similar, but con- fluent, larger, amphigenous and blackish-brown ; spores sub- globose to ovate, rounded above, with a wide germ-pore and a hemispherical hyaline papilla, smooth or rarely provided, espe- cially near the apex, with a few hyaline warts, chestnut-brown, 24 — 35 x 18 — 25 /jl; epispore up to 3^ fi thick; pedicels hyaline, rather thin, about as long as the spore. K)2 UROMYi ES On leaves of Phaseolus vulgaris. .May. .Inly — October: even earlier on forced plants, Uncommon. (Fig. 55.) De Bary I.e. proved the genetic connection of the aecidia with the uredo- and teleutospores. The aecidia are rarely met with; they may occur either before or in company with the other spore-forms. 1 have seen no proof that they have been found in this country. The description i- founded upon that of Sydow. Fischer says that this species is rery common in Switzerland on Phaseolus; it may become a dangerous pan on forced Beans. All affected plants (leaves and stems should be burnt. /'. appendiculatus of Sydow, which oc.-m-s nany Leguminosie, is probably a collective species, though no experiments bearing on this point are available. Distribution: as a collective species (U. appendiculatus) world-wide. 16. Uromyces tuberculatus Fckl. /Ecidiwm Eupkorbiae Gmel. in Linn. Syst. Nat. ii. 1473 p.p. Purton, Midi. Flor. iii. 293. Cooke, Handb. p. 537 ; Micr. Fung. p. 195 p.p. Plowr. Ured. p. 270. Uromyces excavatus DC; Cooke, Grevillea. ii. 161 ; Micr. Fung. p. 213. U. tuberculatus Fckl. Symb. Myc. p. 64. Sydow, Monogr. ii. 165. Fischer, Ured. Schweiz, p. 43, f. 33. U. proeminens Lev. ; Sacc. Syll. vii. 553 p.p. Hypophyllous, spread uniformly over the Spermogones I uEcidiospores whole leaf; secidia immersed, cup-shaped, with a short denticu- late margin; spores orange, densely verruculose, 17 — 25 x 14 — 20 fi. Uredospores. Sori hypophyllous. scattered, at length naked, cinnamon ; spores more or less globose, yellowish -brown, aculeo- late,20 — 25/x: epispore 1^ — 2^/la thick, with 5 — 7 swollen germ- pores (4 — 5, Fischer). . Teleutospores. Sori amphi- genous and on the stems, round, scattered or sometimes arranged in little groups, pulverulent, blackish-brown or black: spores globose to ellipsoid, occasionally Fig. 56. U. tuberculatus. Two ma- ture teleutospores ; a, a teleuto- spore before the tubercles are developed. ON GERANIACE/E 103 with a flat, broad, hyaline papilla, at first smooth, then covered with more or less distant, broadly conical, obtuse, subhyaline warts, chestnut-brown, 20 — 30 x 18 — 24 /j, ; epispore 2 — 1\ fi thick ; pedicels hyaline, deciduous. On Eupltorbia exigmt. Rare; Hampshire, Mr Hill (Plowr. I.e.) ; King's Cliffe, Norths. (Grevillea, I.e.). Midlands (Purton, I.e.). (Fig. 56.) For a long time this species was considered to have only uredo- and teleutospores, but the connection of these with the secidium occurring on the same species of Euphorbia was established by Tranzschel. Berkeley, at King's Cliffe, found them all together. The description given above is partly founded upon those of Tranzschel, Sydow, and Fischer. The mycelium of the secidial stage infests the whole plant, that of the teleuto- spores is more or less localised. Distribution : France, Germany, Switzerland. 17. Uromyces G-eranii Otth et Wart. jEcidium Geranii DC; Cooke, Handb. p. 543; Micr. Fung. p. 199 p.p. Trichobasis Geranii Cooke, Handb. p. 530. Uromyces Geranii Otth et Wartm. Schweiz. Krypt. no. 401. Cooke, Micr. Fung. p. 213. Plowr. Ured. p. 12fc!. Sacc. Syll. vii. 535. Sydow, Monogr. ii. 190. Fischer, Ured. Schweiz, p. 16, f. 14. Spermogones. Mixed with the secidia, orange. sEcidiospores. /Ecidia hypophyllous or on the petioles, on the leaves chiefly in the vicinity of the nerves and there forming large dense clusters on thickened spots, on the petioles forming elongated clusters and often causing great distortion, at first hemispherical and closed, then opening by a round pore, at length with a very slightly revolute incised margin, orange ; spores somewhat ovate, densely and minutely verruculose, yellow, 22 — 28 x 18 — 24 p,; epispore rather thick. Uredospores. Sori hypophyllous, generally on brownish or reddish-yellow spots, scattered or gregarious, minute, rounded, pulverulent, cinnamon, surrounded by the cleft epidermis; spores globose to obovate, sparsely echinulate, brown, 20 — 30 x 18 — 24 fi ; epispore about 2 /j, thick, with one (rarely two) germ-pores. 104 i ROMYCES Te/t'itttispures. Sori similar, but less pulverulent, and blackish-brown : spores subglobose to ovate, not thickened above, but with a hyaline papilla as much as (i /x high, smooth, brown, 22 — 35x18 — 25/a; pedicels short, hyaline, deciduous. On Geranium dissectum, G. 'mode, G. Teleutospores on G. pratense, G. pyrenaicum, G. silvaticum. silvaticum. XT . „-, . ,. ■... , T JNot common, /hcidia, Marcn to June: beleutospores, June — October. (Fig. 57.) Lire proved that the secidium of this parasite from G. silvaticum produced uredo- and teleutospores on the sunn' plant, and Bock showed that the uredospores from the same specie- reproduced themselves on ether hosts of the same genus. But there is another secidium occurring on G. pratense and G. silvaticum, which belongs to a quite different life- cycle. This is JEcidium sHvtjuiitolentiim Lindr., and is the secidial stage of the hetercecious Puccinia Polygon! umphibii Pers. (q.v.). It differs from the secidium of C. Geranii in being seated on conspicuous blood-red or deep-purplish spots which are not distinctly thickened ; moreover the shape of the spores is that usual in secidiospores, viz. rounded-polygon, d, while those of C. Geranii are always more or less ovate, and have a thicker wall. Again, there is an ajcidium on G. pusillum which, according to Sydow, is probably also found on G. molle and G. rotundifolivm, and which belongs to Puccinia /'<>/ 'ygoni-Convolvuli (q.v.) — a form of P. Polygoni-amphibii which is often separated as a distinct species. The uredo- and teleuto- sori would, of course, not follow the secidium on the same plant in either of these two cases. On G. pyrenaicum there is another Uromyces (U. Kabatianus) which differs in the arrangement of its sori ; sec below. Distribution : Europe, except in the extreme South. 18. Uromyces Kabatianus Bubak. Uromyces Kabatianus Bubak, Sitz. kon. holme Gesell. Wissen. 1902, p. 1, f. 1—5. Sacc. Syll. xvii. 249. Sydow, Monogr. ii. 194. Fischer, Ured. Schweiz, p. 18, f. !.">. Spermogones. Amphigenous, few, large, honey-coloured, then darker, on the same spots as the a^cidia. yEcidiosporeft. ./Ecidia hypophyllous, on round yellowish spots, in little clusters 2 — 4 mm. wide, hemispherical, opening • on geraniacea: 105 by a pore ; spores roundish-polygonal to oblong-ovate, densely verruculose, yellow, 24 — 33 x 18 — 2(i jx. Uredospures. Sori hypophyllous, on yellow spots, in circi- nate groups, seldom scattered, rather large, pulverulent, choco- late-brown ; spores roundish, brown, distantly echinulate, 22— 26 fx ; epispore about 2 /x thick. Teleutospores. Sori hypophyllous, on yellowish or reddish Fig. 58. U. Kabatianus. Teleutospores and uredospore, on G. pyrenaicum. spots, rather large, covered with the thin silvery-shining epi- dermis, generally in circinate groups, soon confluent, pulverulent, brown ; spores ovate, often oblong, scarcely thickened above, but with the germ-pore provided with a prominent subhyaline papilla (up to 7 /x high), smooth, pale-brown, 22 — 42 x 13 — 20 /x\ pedicels short, hyaline, deciduous. On Geranium pyrenaicum, and possibly on G. /nolle and G. pusillnm. (Fig. 58.) This has been separated by Bubak from U. Geranii on the ground of the circinate arrangement of the uredo- and telento-sori, and the more oblong and longer teleutospores : the description of the ascidia and teleu- tospores is taken from that of Bubak. According to Sydow, Lind has succeeded in transferring this species from G. pyrenaicum to G. molle and G. pusillum. Bubak thought that only this form occurred on G. 'pyrenai- cum, but Bock (Centralbl. f. Bakt. 2, xx. 584) showed that typical Uromyces Geranii could also be produced on that host. The two Uromyces are very closely allied, but several distinctions are alleged : the uredospores themselves are identical, but not the sori ; according to Sydow, the teleutospores of U. Kabatianus do not appear till towards the end of October, are paler, longer and more oblong in shape with a higher papilla, while their sori are generally circinate, paler and less compact. The teleutospores of U. Geranii appear at the beginning of summer, the sori are nearly black, rather compact, and more scattered. I have specimens collected at Cambridge in August, on G. "pyrenaicum, having the uredo-sori in circinate groups on conspicuous yellow spots, and L06 ritoMYCES containing no teleutosporea ; it appears probable that these belong to U. Kabatianus, which will no doubt be found in many places, if lookedfor. Distribution: a few places in Europe and Asia Minor. in. Uromyces Alchemillae Lev. Uredo Alchemillae Pers. Obs. Myc. i. !,s. Uromyces Alchemillae LeV. Ann. Sci. Nat. :'., viii. ;>T 1 Is 17 . Plowr. Ored. p. L37. Sacc. Syll. vii. 553. Sydow, Monogr. ii. 196. Fischer, (Jred. Schweiz, p. 44, f. 34. U. intrusa Cooke, Handb. p. ">l!t: Micr. Fung. p. 213. Trachyspora Alchemillae Fckl. Bot. Zeit. xix. 2.">u. Arthur, N. Amcr. Fl. vii. 178. Urerfuxpores. Sori hypophyllous, radially arranged, occupy- ing nearly the whole leaf-surface, rounded or elongated, often confluent and com red by large fragments of the torn epidermis, then pulverulent, orange, yellowish or even whitish ; spores ellipsoid to oblong, faintly echinulate, orange or yellowish, 16—25 x 14—21 fi. Fig. 59. 17. Alchemillae. Teleutospores on A. vulgaru. Teleutospores. Sori hypophyllous, scattered, rarely con- tinent, minute, round, pulverulent, brown; spores globose to obovoid or oblong, not thickened above, coarsely warted, brown, 26 — 40x20 — 30 ft: epispore 2 — 2 k fi thick; pedicels hyaline, very deciduous, short or rather long; teleutospores are also formed in the uredo-sori. On Alchemilla vulgaris. Common. Uredospores, April — June; teleutospores, July — October. (Fig. 59.) The mycelium perennates in the rhizome and grows up with the young leaves, causing them to stand more erect, making them paler and con- spicuous, but smaller and often deformed. The separate teleuto-sori are ON RANUNCULUS 107 formed on other leaves on a localised mycelium, cause no deformation and are not conspicuous ; in them are a few secondary uredospores. The teleutospores have unusually coarse warts, mostly towards the apex, or are sometimes nearly or partially smooth. Bubak records (Centralbl. f. Bakter. 2. xvi. 158) that in many trials in three years he could never get the teleutospores to germinate, and could not artificially produce infection in Alchemilla, though Klebahn (Zeitschr. f. Pfianzenkr. 1907) did so readily with the uredospores. This species can be gathered at considerable altitudes in Wales and Scotland (and as high as 7200 ft. in Switzerland). Distribution: Europe, Asia Minor, Greenland. 20. Uromyces FicariaB Lev. Credo Ficariae Schum. PI. Sail. ii. 232. Uromyces Ficariae Lev. Ann. Sci. Nat. 3. viii. 390. Cooke, Handb. p. 518 ; Micr. Fung. p. 212, pi. 7, f. 156—7. Plowr. Ured. p. 140. Sacc. Syll. vii. 568. Sydow, Monogr. ii. 208. Fischer, Ured. Schweiz, p. 13, f. 12. Teleutospores. Sori amphigenous or on the petioles, about ^ mm. diam., rounded, frequently col- lected into dense orbicular or elongated clusters, on pale-yellow spots, especially on the petioles where they cause not- able distortion, soon naked, pulverulent, chocolate-brown ; spores more or less obovate, often irregular, not thickened above, but with a conical hyaline papilla, Fig- 60. U. Ficariae. smooth, pale-brown, 22 — 38 x 18 — 26 fx ; Ficaria^0™* pedicels hyaline, deciduous ; a few sub- globose, pale-brownish, faintly echinulate uredospores, each with three germ-pores, are occasionally found intermixed, but arc usually abortive. On Ranunculus Ficaria. March to early June. Very common. (Fig. 60; see also Fig. 79.) The eecidium on the same host belongs to the life-cycle of Uromyces Poae, and is considered to have no connection with the Uromyces on R. Ficaria, though it may be found on the same leaf. Klebahn proved that the teleutospores reproduce themselves. But there is a curious conclusion arrived at by Tranzschel, as the result of his experiments ION CTROMYCES see Bot. Zeit. Ixiii. 75), thai an sscidium which he finds on R. Ficaria is connected with U. Ruvnicis (q.v. . Tlie spores of U. Ficariae and U. Rumicis are very similar. I >istribution : Europe generally, except the extreme South. 21. Uromyces caryophyllinus Wint. Lycoperdon caryophyllinum Schrank, Baier. Flor. ii. fin's. Uromyces caryophyllinus Winter, Filzc p. L49. Sacc. Syll. vii. .">4.~>. Sydow, Monogr. ii. 210, 362. Fischer, Ured. Schweiz, p. 11, f. 10. Trans. Brit. Mye. Soc. iii. 122. McAlpine, Rusts of Australia, p. 102, f. 152—4 and pi. G, f. .3(1—1. Nigredo caryophyllina Arthur. X. Amer. Fl. vii. 246. Uredospores. Sori amphigenous or on the stems, sometimes on pallid spots, scattered, minute, round or oblong, soon linked, pulverulent, cinnamon: spores globose to ellipsoid, sparsely echinulate, yellowish-brown, 20 — 35x18 — 25 /a: epi- spore 2^ — 3/a thick, with three to five germ-pon •-. Fig. 61. U. caryophyllinus. Teleutospores and uredospore from the saim' sorus, on carnation ; leaf of carnation with two groups of sori. Teleutospores. Sori confluent and large, mostly oblong, surrounded and often covered by the cleft epidermis, sub- pulverulent, brownish-black : spores globose to ellipsoid, with ON CARYOPHYLLACE/K 109 a flat hyaline papilla, densely and minutely punctate, chestnut- brown, 20 — 31 x 18 — 24 //,; epispore 2 — :]p, thick, not thickened at the summit ; pedicels short, hyaline, deciduous. On Dianthus barbatus, D. Caryophyllus, D. chinensis. ( )n cultivated carnations practically all the year round. (Fig. 61.) The "Carnation Rust" was introduced into England on imported plants about the year 1890 ; it sometimes occurs as an epidemic, causing much injury. The teleutospore-containing sori are often clustered on the leaves and stems in circinate or elongated swollen patches ; uredospores are mixed with them. The punctation of the teleutospores is perceptible only when they are viewed dry, and at the best is very indistinct. It is stated by Tranzschel and Fischer that this species is hetercecious, and has its secidium on Euphorbia Gerardiana, but as this Euphorbia does not occur in Britain, the parasite probably maintains itself here without hcteroecism. It is remarked by Sydow that the same is true in Switzerland, at least in certain cases; but see Fischer (p. 530) who produced the fungus, from an lecidium on E. Gerardiana, on Saponaria ocymoides, but not on Dianthus. The fungus has now spread round the world in greenhouses, but only in the sporophytic stage ; the aecidium has not been recognised anywhere except in Europe. The best means of prevention are (1) the selection of resistant varieties, (2) good and careful cultivation, especially sufficient ventilation. If spraying is resorted to, potassium sulphide solution (h oz. to 1 gallon) is perhaps the best, but dilute Bordeaux mixture or copper sulphate solution (1 lb. to 50 gallons), or sponging with a rose-red solution of permanganate of potash have also heen tried. The latter can be used even when the plants are in active growth. Besides spraying, every infected leaf should be plucked off and burnt as soon as discovered. This disease must not be confounded with the outwardly similar "Fairy Ring of Carnations," caused by Heterosporium, but the same remedies apply to both. Distribution : Europe, Western Asia, Japan, South Africa, North America, Australia. 22. Uromyces Behenis Unger. JEcidium Behenis DC. Encycl. viii. 239. Cooke, Handb. p. 541 ; Micr. Fung. p. 197. Uromyces Behenis Unger, Einfluss d. Bod. p. 216. Cooke, Micr. Fung, p. 213. Plowr. Ured. p. 138. Sacc. Syll. vii. 559. Sydow, Monogr. ii. 218. Fischer, Ured. Schweiz, p. 64, f. 48. sEcidiospores. yficidia usually hypophyllous, seated on spots that vary both in size and colour (yellow or purple) and are generally very conspicuous, solitary or collected into I 10 UROMYCES clusters, cup-shaped, whitish-yellow, with ;i torn revolute margin; spores densely and minutely verruculusr, yellowish I 5 -21 /j. diani. Fig. 62. U. Behenis. a, secidia on early leaf. 1>. secidia on later leaf, of S. in fid tn ; two teleutospores. Teleutospores. Sori hypophyllous and on the stems, often surrounding the secondary secidia, irregularly scattered, gre- garious or circinate, rounded or oblong, covered for a considerable time by the lead-coloured epidermis, rather small and compact, brownish-black or black; spores subglobose or obovate, rounded above and thickened (as much as 11 /x), smooth, pale brown, 25 — 35 x 20 — 27 jx ; pedicels persistent, faintly yellow, thick, as much as 75 /x long. On Silene inflata (latifolia), S. maritima. Not common. . hViuia and teleutospores, July — October. (Fig. 62.) The .spots occupied by the secidia vary in colour, but the primary ones are often tinged or margined with purple. This is one of the species whose aecidiospores are capable of reproducing the secidia, as Dietel has shown (Flora, lxxxi. 395, 1895). The primary secidia, on the earlier leaves, are in roundish groups or concentric circles, only a few being scattered. The secondary secidia, on the younger leaves, stand more often singly and are spread over a larger area ; the teleuto-sori spring from the same secondary mycelium or are formed separately. The secondary secidia are not confined to the beginning of the season, but continue to be produced till the end of autumn, being in fact the representatives of the uredo-sori. On this account this species is very interesting biologically. The primary secidia arise from infection by comparatively few basidiospores ; the secondary arise from the more widely dispersed secidiospores of the ON CARYOPHYLLACE.E 111 first generation, and their mycelium can produce either aecidiospores or teleutospores or both. No spermogones seem to be known. The secidium requires to be carefully distinguished from that of Puccinia Behenis (P. Silenes) (q.v.), which is much rarer and does not extend throughout the season. Distribution : Europe generally. 23. Uromyces sparsus Lev. Uromyces sparsus Lev. Ann. Sci. Nat. ser. 3, viii. 369. Cooke, Handb. 1>. 519 ; Micr. Fung. p. 214. Plowr. Ured. p. 136. Sacc. Syll. vii. 545. Sydow, Monogr. ii. 221. Uredospores. Sori amphigenous and on the stems, on pallid spots, scattered, roundish, ^ — 1 mm. diam., convex, covered for a considerable time by the epidermis, which at length splits and surrounds them, then pulveru- lent, pallid-cinnamon ; spores globose to oblong, faintly echinulate, brownish, 18—28 x 15—22^. Teleutospores. Sori similar, but darker; spores subglobose to oblong, rounded above where they are slightly thickened (up to 4//,) and darker, gene- rally tapering downwards, smooth, brown, 22 — 32x14— 21 /u,; pedicels persistent, thick, as much as 60 \x long, brownish at the apex. On Spergularia rubra (Alsine rubra). rare ; I have seen no British specimens. (Fig. 63.) Distribution : Mid- western Europe. 24. Uromyces Chenopodii Schrot. Uredo Chenopodii Duby, Bot. Gall. ii. 899. jEcidium Suaedae Thiim. Fung. iEgypt. iii. no. 53. JEcidium Chenopodii, in Gard. Chron. (1895), xviii. 135. Uromyces Chenopodii Schrot. in Kunz. Fung. Sel. no. 214. Plbwright in Trans. Brit. Myc. Soc. i. 56. Sacc. Syll. vii. 548. Sydow, Monogr. ii. 233. Fig. 63. U. sparsus. Teleutospores on Spergularia salina (foreign, J. Kunze, Exsicc, no. 216) May — July. Very I 12 UROMYCES /Eddiospores. .Kcidia amphigenous, clustered in circles 5 — l<» mm. diam., cylindrical, whitish, margin deeply torn; spores delicately verruculose, yellow, L8 — 22 /x diam. Uredospores. Sori amphigenous, scattered or gregarious, round or more frequently elongate, small, surrounded by the conspicuous torn epi- dermis, cinnamon; spores globose to oval or oblong, delicately and sparingly echinu- late, yellowish-brown, L8 — _•"> x 1(5 — 21 /x: epispore about 1 \ fx thick. Teleutospores. Sori amphigenous but mostly cauline, on the leaves rounded and 1 — 3 mm. diam.. on the stems lanceolate and even as much as 3 cm. long (Sydow ), thick, compact, dark-brown; spores vei \ variable, oval to subpyriform, rounded or subcorneal at the apex, thickened or not, smooth, brown, 24 — 35x18 — 20//,; pedicels pale-brown, persistent, up to 80 /x long or more. Fig. 64. U. Chenopodii. Teleutospores on Suaeda fruticosa (foreign, ex herb. De Thiimenl. On stems, branches and leaves of Suaeda maritime/,. Rare. Terrington Marsh (Mr H. G. Ward); North Wootton Marsh (C. B. P.); August. The gecidia are recorded from Worthing (Miss A. L. Smith, Journ. of Bot. May, 1898), as well as by Plowright from North Wootton. (Fig. 04.) Teleutospores very variable, short and broad or long and narrow in the same sorus ; thickening of apex also varying from 3 to 7 /* ; pedicels often very long and flexuous. The name of this species is misleading ; it has been found on S. fruticosa, but not on the present-day Chenopodium. Plowright mentions, as showing the distinctness of this species from U. Salicomiae, that at North Wootton Marsh it did not spread to Salicomia herbacea, which was Distribution : Germany, and most of the countries in South Europe and North Africa. ON CHENOPODIACE.E 113 25. Uromyces Betae Lev. Uredo Betae Pers. Syn. p. 220. Trichobasis Betae Cooke, Handb. p. 530 ; Micr. Fung. p. 225. Uromyces Betae Lev. Ann. Sci. Nat. ser. 3, viii. 375. Cooke, Micr. Fung. p. 213. Plowr. Ured. p. 127. Sacc. Syll. vii. 536. Sydow, Monogr. ii. 224. Fischer, Ured. Schweiz, p. 10, f. 9. McAlpine, Rusts of Australia, p. 100, f. 148—9, 316, and pi. H. Nigredo Betae Arthur, N. Amer. Fl. vii. 245. Spermogones. In little clusters, honey-coloured. JEcidiospores. yEcidia amphigenous, often on rounded or irregular yellowish spots, collected into rather large clusters which are round or sometimes irregular and confluent, cup- shaped, yellowish, with a reflexed incised margin ; spores delicately verruculose, pale-yellowish, 16 — 24 x 16 — 20 fi. Uredospores. Sori amphigenous, scattered, sometimes con- centrically arranged, thick, pulvinate, circular, up to 2 mm. diam., covered by the epidermis which at length splits, then pulverulent, cinnamon ; spores globose to obovate-oblong, sparsely and minutely echinulate, yellowish, 21 — 32 x Fig. 65. U. Betae. Te- 16— 26/x; epispore 24— S a thick, with ie^p°*eRand Td°" r" ' r r i r~ spore, on B. man tuna. two equatorial germ-pores. Teleutospores. Sori similar, but somewhat compact, dark- brown ; spores globose to obovate, rounded and slightly thick- ened above, with a minute hyaline hemispherical papilla, smooth, pale-brown, 22 — 34 x 18 — 25 /x ; pedicels short, hyaline. On leaves of Beta maritima, B. vulgaris ; also doing great harm to cultivated mangels. ^Ecidia rather rare, April — June ; uredo- and teleutospores, rather common, May — October. (Fig. 65.) In May all four spore-forms can sometimes be seen on the same leaf. Kiihn says that the mycelium of the secidia is perennial, and that its spores can reproduce the eecidia. If possible, the first leaves seen bearing the eecidia should be collected and burnt ; this will check the disease at the outset. If this is not possible, the plants may be sprayed with dilute Bordeaux mixture or with potassium sulphide solution. Since in mangels the disease would chiefly be reproduced by teleutospores from old leaves. G. u. 8 I 14 IKOMVCES of the preceding crop, all affected mangel "tops" .should be burnt: rotation of crops is of course a sure preventive, as in all such cases. I Mm i;ii:i rmx : Europe, California, South Africa, Australia, New Zealand. 26. Uromyces Salicorniae Do Bary. JEcidium Salicorniae DC. Flor. fr. vi. 92. Uromyces Salicorniae De Bary, in Rab. Fung. Eur. nos. 1385— G. Cooke, Grevillea, vii. 137. Plowr. Ured. p. 129. Sacc. SylL vii. 538. Sydow, Monogr. ii. 230. Jl-irfiospores. /Ecidin on the cotyledons chiefly, scattered or in small clusters, at first hemispherical, then cup-shaped, with erect, torn, white margin ; spores finely verruculose, orange-yellow, 17 — 35 p. Uredospores. Sori scattered or aggregated, minute, rounded, long covered by the epidermis, pulverulent, cinnamon ; spores ovate to pyriform, very finely echinulate, yellow-brown, 24 — 35 x 18 — 25 /u, ; epispore about 1| fx thick. Teleutospores. Sori similar, but larger and rather compact, dark-brown ; spores subglobose to obovate, rounded above and often thickened (up to 4 im), and surmounted by a thin, broad, dark cap, rounded below, smooth, brown, 25 — 35 x 18 — 28 \i ; pedicels hyaline, thick, per- sistent, as much as 80 p, long. On leaves and stems of Salicornia europaea {herbacea). October — November. Rare. (Fig. 66.) The teleuto-sori are chiefly on the stems, as much as 3 mm. long, and very pulvinate. Distribution : France, Germany. 27. Uromyces Rumicis Wint. Uredo Rumicis Schum. PI. Sail. ii. 231. Trichobasis Bumicum DC. ; Cooke, Micr. Fung. p. 225 p.p. Uromyces apiculosa Lev.; Cooke, Handb. p. 518; Micr. Fung. p. 212 p.p. (excl. figs.). Fig. 66. U. Sali- corniae. Teleu- tospores on ,S'. herbacea. ON POLYGONACE^E 115 U. Rumicis Wint. Krypt. Fl. i. 145. Plowr. Ured. p. 135 p.p. Sacc. Syll. vii. 544. Sydow, Monogr. ii. 238. Fischer, Ured. Schweiz, p. 9, f. 8. Uredospores. Sori amphigenous, on coloured spots, round, minute, scattered, soon naked, pulverulent, cinnamon ; spores subglobose to ellipsoid, sparsely echinulate, pale-brown, 20 — 28 x 18 — 24 fi, with two (more often three) germ-pores. Fig. 67_ Um Rumicis, Teleuto- Teleutospores. Sori similar. sPores and uredospore, on R. obtuslfolius. but darker ; spores subglobose to pvriform, with a hemispherical hyaline papilla, often narrowed below, smooth or nearly so, brown, 24 — 35 x 18 — 24 yu. ; epispore rather thick ; pedicels thin, hyaline, deciduous. On Rumex conglomeratus, R. crispus, R. Hydrolapathum, R,. nemorosus, R. obtuslfolius, and perhaps others. May — September. Common. (Fig. 67.) The spots on the leaves are small, round, and of various colours ; often the chlorenchyma in the immediate neighbourhood retains its green colour long after the rest of the leaf has become faded and yellow. It will be noticed that the spores of U. Rtimicis are exactly like those of U. Ficariae, and for this .reason Tranzschel was led to suspect some connection between the two, such as he demonstrated to exist between P. fusca and P. Pruni-spinosae, whose teleutospores are equally alike. In 1905 he reported that he had produced an secidium on Ranunculus Fiearia from the spores of U. Rumicis ; still later, he repeated this statement (1909), and added that he had infected Rumex obtusifolhis with secidio- spores from R. Fiearia. Other experimenters (Bubak, Krieg) have been unable to repeat the former of these infections ; they could only produce the secidium on R. Fiearia with the spores of Uromyces Poae. It has been suggested that there are two eecidia on R. Fiearia, one belonging to U. Poae and the other to U. Rumicis ; I have tried to infect R. obtusifolius with secidiospores from R. Fiearia, brought from a place where the secidium on it and the Uromyces on It. obtusifolius were both very abundant, but the attempt failed. Krieg (Centralbl. f. Bakt. 1906) obtained uredo- spores on R. Acetosa with iccidiospores from R. Fiearia, but the same material infected species of Poa (especially P. trivialis), and the possibility 8—2 1 hi DTROMYCES of contamination by foreign spores was not entirely excluded in his experiments. Judgmenl on this point must be suspended DISTRIBUTION: Europe, Algeria, Asia Minor, Africa, Cali- fornia, Chili. 28. Uromyces Acetosae Schrot. Uredo bifrons I >l '. ; Cooke, Eandb. p. 528 ; Micr. Fung. p. 217, pi. vii. Bgs. 137—9. Uromyces Rwmicis Wint. ; Plowr. Ured. p. 135 p.p. Fischer, Ured. Schweiz, p. 9 (not f. 8) p.p. ('. Acetosae Schrot. in Rab. Fung. Europ. no. 2080 (1876). Sacc. Syll. vii. ">37. Sydow, Monogr. ii. 241. [Spermogones. Honey-coloured, clustered. J-'.riil iospores. iEcidia amphigenous or on the petioles, in dense clusters (up to 1 cm. broad), cup- shaped, whitish-yellow, with a cut and re volute margin ; spores nearly smooth or very minutely punctate, clear-yellowish, 18—21 x 12— 18/i.l Fig. 68. U. Acetosae. r J Teleutospores, on Uredospores. Sori amphigenous, often seated on red or purple spots, scattered or circinate, minute, pulverulent, cinnamon ; spores subglobose to ellipsoid, finely and densely verruculose, yellowish or pale brownish-yellow, 18 — 25 x 17- — 22 /x ; epispore about 2|/i, thick, with three germ-pores. Teleutospores. Sori similar, but dark-brown ; spores sub- globose to ellipsoid, not thickened above, or rarely with a paler and very minute papilla, rounded below, beset with very minute warts arranged in lines, brown, 21 — 26x20 — 24 /j,; epispore rather thick ; pedicels thin, hyaline, deciduous. On leaves and petioles of Rumex Acetosa, R. Acetosella. May — September. Not uncommon. (Fig. 68.) Allied to U. Rwmicis, but U. Acetosae has shorter spores (of both kinds) and the hyaline papilla of the teleutospores is almost always wanting. The secidium has not been found in Britain, but the other stages are rather common : the uredo- and teleutospores are unusually alike, but can be distinguished by the germ-pores and the fewer warts of the latter. ON POLYGONACE.E 117 On the same host-plants is a Puccinia, which (in the absence of the teleutospores) can be distinguished only by the fact that the uredospores have two (rarely three) germ-pores and are adorned with few and distant spines. There is little doubt that many of the specimens recorded as U. Acetosae are really the uredospores of Puccinia Acetosae. Distribution : Germany, France, Sweden, Norway, Finland. 29. Uromyces Polygoni Fckl. JEcidium aviculariae Kze. ; Cooke, Handb. p. 545 ; Micr. Fung. p. 1 99. Puccinia vaginalium Link ; Cooke, Handb. p. 495 ; Micr. Fung. p. 204. Trichobasis Polygonorum Berk. ; Cooke, Micr. Fung. p. 226 p.p. Uromyces Polygoni Fckl. Symb. Myc. p. 64. Cooke, Handb. p. 519 ; Micr. Fung. p. 213. Plowr. Ured. p. 123. Sacc. Syll. vii. 533. Sydow, Monogr. ii. 236. Fischer, Ured. Schweiz, p. 61, f. 46. McAlpine, Rusts of Australia, p. 99, f. 150 — 1. ftpermogones. Honey-coloured, conical, only a few together. sEcidiospores. Mostly hypophyllous, on yellow or violet spots, irregularly aggregated or in circular groups, cup-shaped, whitish, with a cleft and revolute margin; spores verruculose, yellowish, 15—21 x 14— 18/*. Uredospores. Sori amphigenous or on the stems, scattered or in small clusters, small, round, soon naked, pulverulent, cinnamon ; spores glo- bose to ellipsoid, densely and minutely verruculose, pale-brown, 18 — 26 x 17 — 24 fi; epispore 1| — 2| /x thick, with three or four germ-pores. Teleutospores. Sori like the uredo- sori, but larger and more confluent upon pjff 69 jj Polygoni. the stems, compact, dark-brown : spores Teleutospores, on P. tit i aviculare. globose or obovate, rounded above and thickened (up to 6 //.), smooth, chestnut-brown, 22 — 38 x 14 — 22 /j, ; pedicels coloured, persistent, thick, as much as 90 /u, long. On Polygonum aviculare. /Ecidia, rare, May — June, Man- chester (T. Brittain), 1875 ; uredo- and teleutospores, very common, July — November. (Fig. 69.) 118 UROMYCES The connection of the aacidium with tlie later stages seem- not yet to have been experimentally demonstrated : but McAlpine found th( on young plants <4 /'. aviculare along with the other spore-forms. In Europe, etc. it is recorded on many other species of Polygonum. This species is said to occur also on Rumex Aeetosella on the continent, and should he looked for here on that host. If so, there would be three species i,n /,'. Aeetosella to he considered : U. Polygoni which has verrucu- lo8e uredospores and smooth teleutospores ; U. Acetosae which is distin- guished by having teleutospores beset with a few minute warts arranged in lines ; while Pttccinia Acetosae has aculeolate uredospores and teleutospores with many delicate warts. Distribution : World-wide. 30. Uromyces Lilii Fckl. Caeoma Lilii Link, Sp. PI., ii. 8. Uromyces Lilii Fckl. Symh. Myc. Nachtr. iii. 16. Sydow, Monogr. ii. 277. Fischer, Ured. Schweiz, p. 6, f. 5. Grove, Journ. Bot. 1911, p. 368. U. Erythronii DC. ; Sacc. Syll. vii. 564 p.p. Wild Fauna and Flora of Kew, p. 163 (non U. Erythronii Pass.?). Xigredo Lilii Arthur, X. Amer. Fl. vii. 242. Spermogones. Dispersed among the secidia, brownish- yellow. jEcidiospores. ^-Ecidia amphigenous, usually hypophvllons, or even on the petioles and stems, seated on lanceolate yellowish spots, in rounded or elongated groups of very different sizes, and often extending widely, more or less crowded, for a long time hemispherical and closed, at length opening by a central pore, but scarcely cup-shaped, margin not revolute, about \ mm. diam. yellow ; spores densely and minutely verruculose, yellowish, 22—35 x 18 — 26 /a ; epispore 3 /a thick or less. Teleutospores. Sori amphigenous, usually hypophyllous, on large yellowish spots, scattered or collected in groups, rounded or oblong, elongated on the petioles, at first covered by the epidermis, which at length splits, large, pulverulent, dark- brown ; spores globose to oblong or pyriform, rounded above, with a prominent hyaline papilla, covered when mature with short ridges wThich are often disposed in lines more or less ON LILIACE.E 119 interrupted and anastomosing, brown, 28 — 44 x 22 — 30 jjl ; epi- spore 2 — 3| /i thick ; pedicels hyaline, slender, deciduous. Fig. 70. U. Lilii. a, leaf of Lilium candidum, with secidia ; b, another, with teleuto-sori, nat. size ; c, cells of peridium, in section and inner face-view ; d, teleutospores ; e, teleutospore seen from above ; /, secidiospore, all x 600. On Lilium candidum. Kew Gardens; also at Birmingham, 1911-3 (C. W. Lowe). /Ecidia in April, May; mature teleuto- spores from June. (Fig. 70.) The part of the leaf occupied by the secidia is somewhat thickened, and the eecidia scarcely protrude above the epidermis. The peridia are slow in opening. The streaked teleutospores are very distinctive. The lilies on which the parasite appeared at Birmingham had been in the garden for some years, but it was not noticed till 1911. Fischer records it on Fritillaria Meleagris, and Sydow and Arthur on other species of the two genera. The true U. Erythronii differs from this species in possessing a truly cup-shaped secidium with a distinctly revolute margin. But the teleutospores of the Birmingham specimens were more like those figured by Fischer under U. Erythronii than those which he figures on Fritillaria Meleagris (f. 5), though devoid of the " Queranastomosen." Possibly the species on Lilium is quite distinct from that on Fritillaria. The longitu- dinal strise are so plainly marked as to be visible under a comparatively low power. As Fischer remarks, the cells of the lower part of the peridium are much thinner-walled than those in the upper part. Distribution : Central Europe, North America. I I'M I HoMVCES Fig. 71. U. Gageae. Teleutospores. 31. Uromyces Gageae Beck. Uromyces Gageae Beck, Verh. k.-k. zooL-bot. Gesell. W'icn, xxx. 26. Sacc. SylL vii. 568. Sydow, Monogr. ii. 27:3. Fischer, Ured. Schweiz, \>. 1, I'. 3. r Omitlwgali Plowr. Ured. p. 1 12. Teleutospores. Sori amphigenous, scattered, roundish or elliptical, 1 — 3 mm. long, covered by the lead-coloured epidermis which at length splits longitudinally, then naked, pul- verulent, dark-brown; spores subglobose to obovoid, not or scarcely thickened above, but usually with a hyaline apiculus, smooth, brown, 20 — 40x18 — 28 /a; epi- spore 2 /* thick ; pedicels hyaline, shorter than the spores. On leaves of Gagea lutea. Rare. April, May. (Fig. 71.) The teleutospores mature in spring, according to Fischer. Plowright says that the mycelium causes variously shaped pale spots on the affected leaves ; hut I find no spots and in Sydow it is said that there are none. Distribution : Western and Central Europe. 32. Uromyces Scillarum Wint. Uredo Scillarum Grev. in Smith, Engl. Fl. v. 376. Uromyces concentrieus~Le\\ ; Cooke, Handh. p. 519; (4revillea, vii. 138 ; Micr. Fung. p. 213. U. Scillarum Winter, Pilze Deutschl. p. 142; Plowr. Ured. p. 141. Sacc. Syll. vii. 567. Sydow. Monogr. ii. 278. Fischer, Ured. 2, f. 1. Sori amphigenous, usually seated on pallid or yellowish spots, small, round or oblong, up to -| mm. diam., collected into round or oblong clusters, often concentrically arranged, sometimes confluent, long covered by the epi- dermis which at length splits and surrounds them, pulverulent, dark- brown; spores subglobose to oblong, usually rounded and not thickened above, smooth, occasionally marked with a few very faint lines Schweiz, p. Teleutospores. Fig. 72. U. Scillarum. Teleu- tospores, on .S'. nutans; a, teleutospore, on S. campanu- lata. ON LILIACE.E 121 running from apex to base, evenly coloured, brown, 18 — 32 x 14 — '22 /u,; epispore uniformly thin, about I^/jl thick; pedicels hyaline, often deciduous, as long as or longer than the spore. On leaves of Endyinion non-scriptum (Scilla nutans), and also of Scilla bifolia, S. campanulata. Common. April — June. (Fig. 72.) The yellow spots and the concentric arrangement of the sori are often very marked. The mycelium is purely local. A few finely echinulate uredospores, 27 x 20 p, were found by Juel in the young sori on Scilla obtusifolia (Bull. Soc. Myc. Fr. xvii. 259). Distribution : Central and Southern Europe and Morocco. 33. Uromyces ambiguus Lev. Uredo ambigua DC. Flor. fr. vi. 64. U. Alliorum Cooke, Handb. p. 528 ; Micr. Fung. p. 217 p.p. Puccinia mixta Fckl., forma simplicior Korn. Uromyces Alliorum DC. ; Cooke, Handb. p. 518 ; Micr. Fung. p. 212. Plowr. Ured. p. 137 p.p. U. ambiguus Lev. Ann. Sci. Nat. 3, viii. 375. Sacc. Syll. vii. 543. Sydow, Monogr. ii. 262. Grove, Journ. Bot. 1911, p. 367. Uredospores. Sori amphigenous, without spots, scattered, roundish or oblong, long covered by the epidermis, yellowish; spores globose to obovate, delicately verruculose, pale-yellowish, 20—28x17— 22 fji\ epispore 3 — 3^/x thick, with six or seven germ -pores. Teleutospores. Sori am- phigenous and on the stems, on the leaves scattered, small and roundish, on the stems confluent and larger, up to 15 mm. long, always covered by the blue-grey epidermis ; spores subglobose to pyriform, rounded above, without a papilla and scarcely thickened, rounded or narrowed below, smooth, brown, 20 — 35 x 17 — 24 /*; pedicels thin, hyaline, fragile, as much as 30 fi long. Fig. 73. U. ambiguus. Teleutospores and uredospore on Allium Scorodo- prasum. I 22 UROMYCES ( )ii Allium Schoenoprasum, A. Scorodoprasum, and (ac- cording to Sydow) A . .$])haerocephalum. Not common. ( Fig. 73.) It cannot be considered as certain thai this is a species distinct from Puccinia Porri Wint., but as it presents slight differences, it is better, in the total absence of culture experiments, to keep it separate for the time. The difficulty lies in the fact that the teleutospores of this Uromyces agree I lerfectly with the mesospores of the Puccinia (except perhaps in the greater variability of the latter) ; but the teleuto-sori of the Uromyces are generally larger and more persistently covered by the epidermis, and one searches in vain in them for two-celled spores, such as are found freely in the sori of the Puccinia. Fischer says that the uredospores of P. Porri have only three germ-pores. According to Sydow, the Uromyces occurs only on the three species of Allium mentioned above and A. rotunda m, while the Puccinia is found on them as well as on many other species of the genus. Nevertheless the two forms are closely allied and from the evolutionary point of view the Uromyces must be regarded as a specialised state or mutation of the Puccinia. Neither of them has an secidial stage. Distribution : Central and North-Western Europe. 34. Uromyces Colchici Massee. U. Colchici Mass. Grevillea, xxi. 6, pi. 182, f. 16 — 18 ; Diseases of Cult. Plants, p. 292, f. 85 ; Text-book of Plant Diseases, p. 227, f. 56. Sydow, Monogr. ii. 268. Teleutospores. Sori amphigenous, scattered, rather large, elliptical, sometimes circinating, up to 2 mm. long, covered for some time by the epidermis which at length splits, then sub-pulverulent, brown ; spores subglobose to ovate, rounded above, with a broad flat hyaline papilla, smooth, pale-brown, 28 — 40 x Fig. 74. u. 20 — 28 /x ; epispore 3 — 3^ /j, thick ; pedicels hya- Colchici. ime rather long, but very deciduous. Teleutospore, ' S' J . . from the ori- On leaves oi Lolcf ncum speciosum. Kew speci- Gardens ; unknown in the world elsewhere. mem J II 11*11 . (Fig. 74.) The teleutospores remain on the dead leaves and germinate in the following spring, so that if Colchicum is again planted in the same ground or allowed to remain there, it is liable to contract the disease year after year. As in all similar cases, the best preventive against future attacks ON JUNCACEyE 123 is to remove carefully and burn all diseased leaves before they mature their spores. The fungus is stated to have attacked the foliage of the host for three successive seasons, completely destroying it, and although for the first two seasons it did not attack other species of Colchicum growing near, during the third season it spread to C. autumnale and C. bavaricum. 35. Uromyces Junci Tul. JEcidium zonale Duby, Bot. Gall. ii. 906. Cooke, Grevillea, xiv. 39. Uromyces Junci Tul. Ann. Sci. Nat. ser. 4, ii. 146. Cooke, Grevillea, vii. 139 ; Micr. Fung. p. 213. Plowr. Ured. p. 132 ; Grevillea, xi. 52. Sacc. Syll. vii. 541. Sydow, Monogr. ii. 287. Fischer, Ured. Schweiz, p. 57, f. 43. Nigredo Junci Arthur, N. Amer. Fl. vii. 238. Spermogones. Usually epiphyllous. sEcidiospores. iEcidia hypophyllous, seated on spots which are zoned with yellow and purple, in dense circinate clusters 2 — 5 mm. wide, cup-shaped, yellowish -white, with a torn revolute margin ; spores densely and minutely verruculose, transparent-yellowish, 17 — 21 /ut. Uredospores. Sori scattered, roundish or oblong, up to 1 mm. long, surrounded by the cleft epi- dermis, pulverulent, brown ; spores globose to ellipsoid, faintly echinu- late, yellowish-brown, 20 — 28 x 16 — 22 /a, with two equatorial germ-pores. Teleido spores. Sori amphigenous or on the culms, scattered or occasionally aggregated, similar to the uredo-sori, but darker ; spores oblong-ovate to clavate, rounded or conical above and much thickened (up to 14 fi), attenuated below, smooth, dark- brown, 24 — 42 x 12 — 18 fj,; pedicels thick, persistent, brownish, as much as 60 p, long. iEcidia on Pidicaria dyse?iterica, May — July ; uredo- and teleutospores on Juncus obtusiflorus, from July onwards, lasting through the winter on the dead culms. Not common. (Fig. 75.) Fig. 75. U. Junci. Teleuto- spores, on J. obtusiflorus. 124 I-ROMY* ES The coiiiR'ction of the two tonus, lii si proA ed bj Fuckel and Plowright, lias been confirmed by Fischer. < >n the continent and elsewhere, there are forms of Uromyces on other species of Juncus, some of which (and possibly mosl of them have their aecidia on other hosts than /'. dysenterica. Distribution: Central and Western Europe, Algeria, North and South America. 36. Uromyces Scirpi Burr. jEcidium Glaucis D. et M. ; Cooke, Grevillea, xv. 29. Plowr. I 'red. p. 268 ; Gard. ( 'hron. ser. 3, vii. 682, 7 16. Uredo Scirpi Cast. Catal. PI. Marseille, p. 214. Uromyces Scirpi Burr. Parasit. Fung. Illinois, p. L68. Sacc. Syll. vii. 558. Sydow, Monogr. ii. 302. U. lineolatus Desni. ; Plowr. Grevillea, xxi. 111. U. maritimae Plowr. Gard. Chron. ser. 3, vii. 682, 746 ; Jour. Roy. Hort. Soc. (1890), p. cix. Klehahn, Wirtswechs. Rostpilze, p. 328. Nigredo Scirpi Arthur, X. Amer. Fl. vii. 233. Spermogones. Usually epiphyllous. JEcidiospores. ^Ecidia hypophyllous or on the petioles, in rather small clusters, cup-shaped, with an incised re volute margin ; spores densely and minutely verruculose, transparent- yellowish, 16—24 x 14— 20/u. Uredospores. Sori hypophyllous, scattered or arranged in lines, rounded or oblong, up to 1 mm. long, surrounded by the cleft epidermis, pulverulent, cinnamon ; spores globose to ovate, distantly and minutely echinulate, yellowish -brown, 22 — 35 x 16 — 25 /x ; epispore 1^ — 2 /x thick, with three equatorial germ-pores. Teleutospores. Sori amphigenous, on in- definite discoloured spots, scattered or confluent in lines, long covered by the epidermis, brownish- black; spores oblong to clavate, tapering usually and thickened (up to 12 /*) above, attenuated below, smooth, pale-brown, 26 — 45 x 15 — 24 fx ; pedicels brownish, persistent, as long as or longer than the spore. Fig. 76. I '.Scirpi Teleutospores, on S. duviatilis. Illinois, U.S.Ai ON GRAMINE^: 125 yEcidia on leaves and petioles of Olaux maritima, May; uredo- and teleutospores on Scirpus maritimus, June to August. Banks of the Humber, Hull. (Fig. 76.) The researches by which Plowright proved the connection of these two forms are given in Grevillea, xxi. Ill, and in Jonrn. R. Hort. Soc. xii. p. ex. ; other observers have found a similar Uromyces on Scirpus mari- timus and therefrom have produced secidia on other host plants such as Pastinaca sativa (Rostrup), Berula angustifolia, Daucus Carota (Bubak), (Enanthe aquatica (Klebahn), Hippuris vulgaris and Sium latifolium (Dietel), etc. In North America, a morphologically indistinguishable Uromyces on Scirpus Jluviatilis, etc. has produced an secidium on Cicuta maculata (Arthur), and similar aecidia on allied UmbelliferEe are suspected to belong to the same life-cycle. It is evident that U. Scirpi, like Puccinia Isiacae, is in its eecidial stage a plurivorous species, though possibly some of these forms may be separated in the future as " biological " races. In any case, they are not so sharply distinguished as in other instances, but Klebahn isolates our British species as U. maritimae Plowr. See the full account in Sydow, Monogr. ii. pp. 304 — 7. Distribution : Europe and North America. 37. Uromyces Dactylidis Otth. JEcidium Ranunculi-acris Pers. Obs. Myc. ii. 22. JE. Ranunculacearum DC. Fl. fr. vi. 97 p.p. Cooke, Handb. p. 539 ; Micr. Fung. p. 196 p.p. Uromyces Dactylidis Otth, Mittheil. Nat. Gesell. Bern, 1861, p. 85. Plowr. Ured. p. 130. Sacc. Syll. vii. 540 p.p. Sydow, Monogr. ii. 309. Fischer, Ured. Schweiz, p. 71, f. 54. U. graminum Cooke, Handb. p. 520 ; Micr. Fung. p. 214. Spermogones. Epiphyllous, honey-coloured, but also a few scattered among the secidia on the lower surface. jEcidiospores. iEcidia hypophyllous or on the petioles, seated on yellow spots, in roundish or, on the petioles, elongated clusters, cup-shaped, yellow, with slightly torn, recurved margin ; spores delicately verruculose, pale-yellowish, 17 — 25 /z. Uredospores. Sori amphigenous, scattered or in rows, small, elliptic or oblong, long covered by the epidermis, pulverulent, yellow-brown ; spores globose to ovate, delicately echinulate, yellow or yellow-brown, 21 — 32 x 18 — 25 fi ; epispore 1| — 2 /j, thick, with 7 — 9 germ-pores ; paraphyses generally wanting. 1 12 « i UROMYCES Teleittospons. Sori generally hypophyllous, similar to the uredo-sori but more often confluent, always covered by the epidermis, compact . shining, black; spores ovate- oblong, occasionally ellipsoid or pyri- form, rounded above, rarely truncate, often slightly thickened (up to 4//,), Fig. 77. U. Dactylitis. Te- smooth, yellow-brown, darker only leutospores and the accom- along the summit, 18 — 30 X 14 — panying paraphyses. . ,. , ,. , 20//,; epispore H/u, thick; pedicels brownish, persistent, nearly as long as the spore; paraphyses numerous, brown, agglutinated, dividing the sori into compart- ments. . Ecidia on Ranunculus acris, R. bulbosns, R. repens, March — May: uredo- and teleutospores on Dactylis glomerate,, from July onwards, often covering the leaves, less often the sheaths and culms, and persisting through the winter, especially on the latter. (Fig. 77.) This species and the following (U. Poae) are very closely allied, and should possibly be united. Perhaps more experimental cultures have been carried out with these two species than with most other Uredines ; but the result is only a wild confusion of contradictory statements, from which one can infer, either that an immense number of intricately connected, but morphologically indistinguishable forms, inhabit the species of Ranun- culus and of Poa and Dactylis — or, preferably, that the factors which govern the success of an attempted infection are so numerous and so little known, that failure does not afford any ground for arriving at a definite conclusion. Those who wish to learn further should consult the long account of these results given in Sydow, Monographia, ii. pp. 312 — 16. In the British specimens of U. Dactylidis which I have examined, the paraphyses in the teleuto-sori, though often overlooked, are a conspicuous feature. But upon the question of paraphyses in the uredo-sori no agree- ment has been arrived at ; the various authorities flatly contradict one another. Either, therefore, the paraphyses occur differently in different countries, as Plowright suggests, or more than one species is included under this title, or their presence or absence is a matter of no importance. Against the latter suggestion, however, we must set the fact that in other cases, e.g. in Puccinia Sonchi, the paraphyses form a constant and dis- tinctive character. An secidium occurring on Ranunculus acris belongs to Puccinia perplexans Plowr., but cannot be distinguished from the present one, ON GRAMINE.E 127 except by cultures ; see also under U. Poae. The teleuto-sori of the present species are more numerous and much more conspicuous than those of U. Poae. They resemble the spots of Phyllachora graminis on the same host. Distribution : Europe only, not yet found in America. 38. Uromyces Poae Raben. JEcidium crassum var. (3 Ficariae Pers. ; Sow. pi. 397, f. 4. JE. Ranuncidacearum DC. ; Cooke, Haudb. p. 539 ; Micr. Fung. p. 196, pi. ii, figs. 12—14, p.p. Uromyces Poae Rab. Unio itin. 1866, no. 38. Plowr. Ured. p. 131. Sydow, Monogr. ii. 310. Fischer, Ured. Schweiz, p. 72, f. 55. Spermogones. Epiphyllous, honey-coloured, a few also on the lower surface among the aecidia. ^-Eddie-spores. iEcidia hypophyllous or on the petioles, in roundish clusters on yellow spots, which on the petioles are often elongated, cup-shaped, yellow, with torn, recurved margin ; spores delicately verruculose, clear-yellow, 17 — 25 x 12 — 20 /jl. Uredospores. Sori amphigenous, scattered or in rows, small, elliptic or oblong, at first covered by the epidermis, pulverulent, yellowish-brown ; spores globose to ovate, faintly echinulate, yellow, 14 — 25 x 14 — 20 yu, ; epispore 1| — 2 n thick, with 4 — 9 germ-pores ; a few paraphyses occasionally intermixed. Teleuto spores. Sori generally hypophyllous, similar to the uredo- sori, but always covered by the epidermis, compact and black ; spores oblong-ellipsoid to pyriform, rounded or truncate above, not much thickened (up to 4 or 5 ytt), smooth, yellowish-brown, apex alone darker, 17 — 28x14 — 20 fx ; epispore l|/x thick; pedicels brownish, as long as or shorter than the spore ; para- physes numerous, brown, agglutinated, as in U. Dactylidis. iEcidia on Ranunculus auricomus, R. bidbosus, R. Ficaria, R. repens, March — May ; uredo- and teleutospores on Poa annua, P. nemoralis, P. pratensis, P. trivialis, May — September. (Fig. 78; see also Fig. 211.) Fig. 78. U. Poae. Teleutospores and the accompanying para- physes, on P. trivialis. L28 IMTCINIA ■•■ Concerning the difference between this species and U. Dactylidu (apart from the habitat) little that is definite can ^ .Vl" a be said. Some authors unite them, but I find the teleutospores of U. Poae to be usually more oblong and often provided with shorter pedicels and the sori to be less conspicuous. The leaves of Poa are smaller than those of Dactylis, and the teleuto-sm-i do not cover them in such enormous numbers and never extend to the culms. Juel divides this species into 9 or 10 biological races, but there is the usual conflict between different experimenters as to their limits. Some would even deny, what has been proved several times, that an secidium on Pi. Ficaria belongs here: see U. Rumicis (p. 115). This secidium and U. Ficariae may occasionally be found on the same leaf of R. Fia.tr in. Kig. 7!).) It must not be forgotten that an secidium which occurs on R. bvlbosus and R. repens belongs to Puccinia Magnusiana : this is morphologically in- b Fi^'. 79. .Kcidia of U. Poae (b) and teleuto- sori of U. Ficariae (a), on a leaf of I?. Ficaria, nat. size. distinguishable but is said to be later in its appearance. The eecidia which can be seen on R. repens in July and August possibly all belong to this latter species. The teleutospores of U. Poae can most easily be found by looking on the lower leaves of Poa in June or July, along a damp road-side where It. rcpc/is is abundant. Distribution : Europe, Asia Minor, Nova Scotia. PUCCINIA Pers. Autcecious or hetercecious. Spermogones as in Uromyces. vEcidia with a peridium, which is occasionally evanescent ; spores as in Uromyces. Uredospores as in Uromyces. Teleutospores two-celled, rarely one- or several- celled, the upper cell usually with an apical pore, less often the pore is displaced to the side; the lower cells with a pore just beneath the septum or rarely at a lower level. Basidiospores as in Uromyces. ON COMPOSITE 129 This genus occupies an intermediate position between , Uromyces and the less advanced genera, Phragmidiurn, etc., as is shown by the fact that many species tend to form one-celled teleutospores (mesospores) indistinguishable from those of Uromyces, while others have spores with three or more cells, arranged as in Phragmidiurn, Triphragmium, Sphaerophrag- iii i a m, etc. The number of species is enormous, more than 1300 are already known. The genus must therefore be subdivided, but no quite satisfactory classification has as yet been discovered. Schroter's and Fischer's separate nearly allied species, while Arthur's is a pathless chaos. As a temporary measure, instead of adding a new imperfect scheme to those already existing, the species are here arranged on the plan adopted in Sydows' Monographia, in the order of the families and genera on which they are parasitic ; this has the advantage that it does, to a great extent, bring nearly allied species close together, while it is at the same time very convenient for consultation. In each family the genera are arranged in the order familiar to British botanists, but the fami- lies themselves are in the order usually adopted on the continent, because that will before long be accepted here also. 1. Puccinia Tripolii Wallr. Puccinia Tripolii Wallr. Fl. Crypt. Germ. ii. 223. Cooke, Micr. Fung. p. 207; Gre- villea, iii. 180. P. Asteris Duby, Bot. Gall. ii. 888. Plowr. Grevillea, ii. 48 ; Ured. p. 215. Sacc. Syll. vii. (587 p.p. Sydow, Monogr. i. 15 Fig. 80. P. Tripolii. Teleutospores (one Teleutospores. Sori amphigenous, con- abnormal), fluent into rather large, pulvinate masses, hard, compact, very dark-brown ; spores oblong-clavate or clavate, rounded above or somewhat narrowed and much thickened (as much as 8/f), slightly constricted, generally attenuated downwards, smooth,, g. u. 9 130 PUCCINIA pale clear-brown, 45 — 60 x 20—2.") ll : pedicels brownish, persis- tent, thick, about as long as the spore or longer; no mesospores were seen, but some irregular spores. On Aster Tripolivm. New Pitsligo, 1870 (Herb. Berk.); Wblferton Beach, King's Lynn, July — November, 1873 (Plow- right). (Fig. 80.) The greyish tinge mentioned by Plowright seems to be due to germ- tubes issuing from the spores of this Lejrtopuccinia. This species is decidedly different from the American forms with which it is united by Sydow ; there are no spots, the colour of the spore is paler, the apex not so much thickened, and the sori are not confined to the lower leaf-surface. M.my spores were observed bifid at the summit, but not owing to germi- nation, which had not taken place in them. Distribution : Northern and Central Europe, Siberia. 2. Puccinia Virgaureae Lib. Xyloma Virgaureae DC. Flor. fr. vi. 158. Puccinia Virgaureae Lib. Crypt. Arduen. iv. No, 393. Cooke, Handb. p. 500 ; Micr. Fung. p. 206. Plowr. Ured. p. 203. Sacc. Syll. vii. 679. Sydow, Monogr. i. 151. Fischer, Ured. Schweiz, p. 363, f. 264. Teleiitospores. Sori hypophyllous, minute, crowded in stellate or dendritic clusters, on round, yellowish, purple-centred spots, com- pact, shining, black ; spores oblong, clavate or fusoid, above rounded, attenuated or truncate, very much thickened (as much as 12 ll) and darker or with a paler hood-like cap, hardly constricted, tapering below, smooth, yellow-brown, paler down- wards, 30 — 56x12 — 20^; pedicels somewhat hyaline, half as long as the spore. On Solidago Virgcmrea. August and September. Un- common ; Surrey, etc. (Fig. 81.) The sori are arranged in a radiating fashion ; they are surrounded each by a thick fence of dark-brown paraphyses, remain long covered by Fig. 81. P. Virgaureae. Teleutospores. ON COMPOSITE 131 the epidermis and resemble rather a Dothidea or Asteroma than a Puccinia. The spores bear a slight resemblance to those of P. Poarum, but are more fusiform, and much more thickened at the apex. Mixed with them are sometimes a number of mesospores. Distribution : Europe, except the southern parts. Fig. 82. P. Millefolii. spores. Teleuto- 3. Puccinia Millefolii Fckl. Puccinia Millefolii Fckl. Symb. Myc. p. 55. Cooke, Micr. Fung. p. 207. Plowr. Urecl. p. 215. Sydow, Monogr. i. 2. Fischer, Ured. Schweiz, p. 296, f. 216. Teleutospores. Sori amphigenous, on indistinct spots, minute, roundish or irregular, gene- rally scattered, compact, dark- brown ; spores oblong or clavate, rounded or gently attenuated at the apex and thickened (4 — 9 /a), con- stricted, more or less tapering below, smooth, pale-brownish, 35 — 50 x 13 — 19 n ; pedicels yellowish above, thick, persistent, about 40 jx long; a few mesospores are sometimes intermixed. On Achillea Millefolium. For- den ; St Leonards. August — October. (Fig. 82.) Plowright demonstrated {I.e. p. 216) that this species is quite distinct from P. Tripolii, with which it had been previously united, and his con- clusion has been confirmed by Magnus and others. The species is rather uncommon ; there is a closely allied one on A. Ptarmica (P. Ptarmicae Karst.) which has not yet been seen in Britain. The spores of P. Mille- folii germinate as soon as mature. Distribution : Central and Western Europe. 4. Puccinia Chrysanthemi Roze. Uredo Chrysanthemi Roze ; Plowright in Trans. Brit. Myc. Soc. i. 98. Puccinia Chrysanthemi Roze, Bull. Soc. Myc. Fr. 1900, p. 92. Sacc. Syll. xvi. 296. Sydow, Monogr. i. 46, 854. McAlpine, Rusts of Australia, p. 153, f. 251—5, and pi. E, f. 21. Fischer, Ured. Schweiz, p. 190, f. 150. P. Chrysanthemi-chinensis Henn. in Hedwig. xl. 26 (1901). 9—2 132 PUCCINI \ Pig. 83. P. Chry- santhemi. Ure- dospore (Brit- ish). Uredospores. Sori generally hypophyllous, on irregular pallid-yellow or brownish spots, scattered or in clusters, about 1—1! nun. diam. often circi- oate, pulverulent, snuff-brown; spores globose to ellipsoid, delicately echinulate, brown, 24 — •">2 x 17 — 27 p, mostly with three germ-pores. [Teleutospores. Mixed with the undospores, oblong or ellipsoid, rounded and slightly thickened above, usually rounded or some- what tapering at base, scarcely constricted, delicately verruculose, chestnut-brown, 35 — 57 x 20 — 25 fi : pedicels thick, hyaline, persistent, 35 — 60 [i long; mesospores subglobose or pyriform, slightly thickened at the summit, 32 — 37 x 20—21 fi.] On leaves of Chrysanthemum in- dicum and C. sinense (not on other species of the genus, much less on other genera of Compositae), in greenhouses, all the year round. The leaves that are attacked soon flag and die. (Figs. 83, 84.) This species is said to be very common in Japan. It was first observed in England in 1895, and has been found in other Euro- pean countries and in North America ; in 1904 it reached Australia and New Zealand. In Japan it produces teleutospores in separate sori, which are hypophyllous, roundish, dark-brown and naked, but in Europe the teleutospores have been rarely seen, though mesospores occasionally occur. Abnormal and 2-celled uredospores (as well as 3- or 4-celled teleutospores) have been described and figured by Roze, Jacky and Eischer ; but these I have not seen in British specimens. Since, under the conditions in which the plants are grown here, the young shoots appear above ground before the old ones die away, it is probable that the parasite maintains itself by the uredospores alone ; the alternative would be the possession of a perennial mycelium, which has not Fig. 84. P. Chry santhemi. Teleutospores and uredo- spores, one abnormal (after Fischer). ON COMPOSITE 133 been found (Gibson, 1904, p. 188). If that is so, the disease can easily be kept in check by rigid cleanliness and by spraying at intervals with very dilute Bordeaux mixture or, better still, potassium sulphide solution. Remove and burn all attacked leaves as soon as seen, water carefully with- out wetting the leaves, choose resistant varieties {e.g. " October Sun " and " William Tricker" are said to be immune), and there will be little fear of an epidemic of the disease. Distribution : Europe, Japan, North America, Australia. 5. Puccinia Leucanthemi Pass. Puccinia Leucanthemi Pass, in Hedw. 1874, p. 47. Sacc. Syll. vii. 705. Sydow, Monogr. i. 116, f. 95. P. Asteris var. Chrysanthemi- Leucanthemi Massal. in Bull. Soc. Bot. Ital. 1900, p. 258. Sacc. Syll. xvi. 297. Trans. Brit. Myc. Soc. iii. 224. Teleutospores. Sori amphigenous, generally hypophyllous, often also on the petioles, scattered or often circinate on indistinct spots, or confluent into compact cushions 2 — 5 mm. wide, reddish-brown; spores oblong or subclavate, somewhat rounded or more often narrowed at the apex, much thickened above (up to 14 p), constricted, tapering down- wards, smooth, yellowish, 40 — 70 x 14 — 24 /x; pedicels hyaline, thick, about as long as the spore. On Chrysanthemum Leucanthe- mum. Very rare. Lamorna Cove, Cornwall, September, 1906 (F. J. Chittenden). (Fig. 85.) Only recorded for Britain and Italy. The similarity of the spores to the teleutospores of Puccinia JEcidii- Leucanthemi Fisch., which has its eecidiospores on C. Leucantliemuyn and its teleutospores on Carex montana, seems to indicate that this is one of those instances, like P. fusca and P. Pruai-spinosae, which give us a glimpse into the mode of evolution of the Uredinales. Fig. 85. P. Leucanthemi. Teleutospores, from an ori- ginal specimen issued by Passerini. 1 34 PUCCINI A (5. Puccinia Absinthii DC. Trichobasis Artemisiae Berk. ; Cooke, Micr. Fung. p. 223. Puccinia Discoideamm Link ; Cooke, Micr. Fung. p. 206. /'. Tanaceti Plowr. Ored. p. Is'-' p.p. Sacc. Syll. vii. 637 p.p. /'. Absinthii DC. Flor. fr. vi. 56. Sydow, Monogr. i. 11. Fischer, Ured. Schweiz, p. 188, f. 148. Uredospores. Sori generally hypophyllous, on yellowish- brown or indeterminate spots, scattered or aggregated, not confluent, minute, roundish, pulverulent, pale-brown; spores globose to ovoid, echinulate, pale yellowish-brown, 20 — 35 x 15 — 26 ft, mostly with three subequatorial germ-pores. Teleutospores. Sori amphigenous, but generally on the lower surface or sometimes on the stems, similar to the uredo-sori but occasionally confluent, soon naked, dark-brown or blackish ; spores oblong to oblong-clavate, rounded and thick- ened (3 — 7 yu.) above, constricted, slightly attenuated below, the upper cell punctate or verruculose, the lower frequently smooth, especially at the base, brown, 38 — 62 x 20 — 27 fu. : pedicels hyaline, thick, persis- tent, as much as 80 /j, long. On Artemisia Absintliivm, A. maritima, A. vulgaris. July — Sep- tember. Rather uncommon. Also recorded in Switzerland on A. campestris. (Fig. 86.) Fig. 86. P. Absinthii. Teleu- tospores, on A. Absinthium. The germ-pores of both kinds of spores are covered with paler, swollen caps. P. Tanaceti, which was formerly united with this species, possesses on the average narrower and shorter teleutospores, though in each these spores are said to be marked in a similar way, chiefly in the upper half. But the markings of P. Absinthii are stronger and less likely to escape observation, especially on the pore-caps of both cells. Distribution : Europe, Siberia, Japan, North America ON COMPOSITE 135 7. Puccinia Tanaceti DC. Puccinia Tanaceti DC. Flor. fr. ii. 222. Cooke, Micr. Fung. p. 207. Plowr. Ured. p. 189 p.p. Sacc. Syll. vii. 657. Sydow, Monogr. i. 161. Fischer, Ured. Schweiz, p. 185, f. 146. P. Discoidearum Link ; Cooke, Handb. p. 499 p.p. Uredospores. Sori amphigenous, scattered, minute, not confluent, orbicular, pallid-brown ; spores subglobose or broadly ellipsoid, echinulate, yellow-brown, 25 — 32x16 — 25 fi, with three germ-pores, each covered by a convex colourless cap. Teleutospores. Sori mostly hypophyllous, similar, but more compact and blackish, ^ — frara, diam. ; spores ellipsoid or Fig. 87. P. Tanaceti. Teleutospores. oblong, more or less rounded at both ends, thickened above (up to 7 fju), delicately verruculose(?) especially in the upper part, rich-brown, 32 — 44x16 — 24 p; pedicels hyaline, thick, per- sistent, as much as 120 /x long. On Tanacetum vulgare. Not common. August, September. (Fig. 87.) 136 PUCCINIA Formerly united with Puccinia Absinthii, from which it is distinguished 1 > v its (on tin' average) narrower and shorter teleutospores, with longer pedicels. I could not sec in any case that the teleutospores were verru- culose, as thej are Baid to be. |)isTKii:iTTiox: North- western and Central Europe. 8. Puccinia expansa Link. Puccini" expansa Link, Sp. PL vi. 2, p. 75. Sydow, Monogr. i. 146. Fischer, Ured. Schweiz, p. 182, f. 141—2. P. Senecionis Cooke, Micr. Fung. p. 207. Plowr. Ured. p. 209, (non Lib.). Teleutospores. Sori amphigenous,but generally hypophylli mis, on round yellowish or brownish spots, densely crowded and con- fluent in roundish clusters 5 — 8 ram. broad, blackish-brown; spores ovate or broadly ellipsoid, rounded at both ends, with a minute paler apical papilla, hardly constricted, smooth, brown, 30 — 40 x 19 — 30/*, but sometimes smaller; pedicels hyaline, very short ; an occasional mesospore is seen. On Senecio aquaticus, S. Jacobaea. Not common. July- September. (Fig. 88.) The sori of this species are chiefly on the underside of the leaf; they are collected into roundish or elongated groups and open by a small round pore, after the manner of some recidia, so that, when the spores have fallen out, the group looks like a honeycomb. P. glomerata is said to be distinguished by its paler, narrower and longer spores with a smaller papilla, and by having the sori on both leaf-surfaces. I think the two species are the same so far as our specimens are concerned, but have followed Sydows' Monographia pro tern. Plowright has confused the characters of the two, partly because he thought his species on S. aquaticus was P. Senecionis Lib., which is incorrect. The true P. Senecionis Lib. has not been found in Britain. Distribution : Central Europe, Holland, California. Puccinia glomerata Grew Puccinia glomerata Grew in Berk. Eng. Fl. v. 365. Cooke, Handb. p. 500 ; Micr. Fung. p. 206. Plowr. Ured. p. 209. Sj'dow, Monogr. i. 148, t. 8, f. 125. 88. V. expansa. Teleuto- spores and mesospore, on ,S'. Jacobaea. 9 ON COMPOSITE 137 Teleutospores. Sori amphigenous, on brown orbicular spots, densely gregarious and confluent in roundish clusters, 3 — 6 mm. broad, occupying the whole spot, often circinate, long covered by the epidermis, at length rather pulverulent, deep-brown ; spores ovate to oblong, with a very minute paler apical papilla, hardly constricted, rounded or slightly attenuated below, smooth, pale-brown, 30 — 45 x 16 — 24 /a; pedicels hyaline, very short. On Senecio Jacobaea. August- -November. Not common. Closely allied to the preceding species, but said to be distinguished by its paler, narrower and longer spores which ai'e provided with a smaller apical papilla. Moreover, the sori are said to be less dark in colour (not darker as Plowright states), and are distributed more uniformly on both leaf-surfaces. Nevertheless, I suspect they are identical. Distribution : Belgium. 10. Puccinia Carlinae Jacky. Puccinia Hieracii Mart. ; Plowr. Ured. p. 184 p.p. P. Carlinae Jacky, Compos. -Puce. p. 59, f. 14. Sacc. Syll. xvi. 297. Sydow, Monogr. i. 35. Fischer, Ured. Schweiz, p. 216, f. 167. Spermogpnes. Epiphyllous and on the petioles, singly or in groups. Uredospores. Sori amphigenous, without evident spots, scattered, not confluent, mi- nute, punctiform, pulverulent, brown ; spores globose or broadly ellipsoid, very deli- cately echinulate, pallid- brown, 24—30 x 20—25 fi; epispore uniformly thick, with three germ-pores. Teleutospores. Sori simi- lar, but blackish-brown; spores ellipsoid to obovate-oblong, rounded at both ends or more often attenuated downwards, not thickened above, hardly constricted, delicately verruculose, brown, 25 — 35x16 — 20 //,; pedicels hyaline, short. Fig. 89. P. Carlinae. Teleutospores. l:;s PUCCINI A On Carlinavulgaris. .Inly — October. Not common. (Fig. 89. ) Distinguished from /'. Cirsii, which it much resembles, by the larger uredospores, with very delicate hardly perceptible spines, and by the slightly smaller teleutospores which are more frequently narrowed below. But no cultures seem to have been made to prove their distinctness. Distribution : Central and Northern Europe. 1 1 . Puccinia Bardanae Corda, I'lic-iitiii llicrurii Plowr. Ured. p. 184 p.p. /'. H< i rib tine Corda, Icones, iv. 17. Sydow, Monogr. i. 113. Fischer,. Ured. Schweiz, p. 221, f. 173. Spermogones. Similar to those of P. Cirsii. Uredospores. Sori of two kinds ; primary epiphyllous, on- discoloured rounded or irregular spots, in concentric rings round the spermo- gones, sometimes confluent, 1 — 5 mm. broad, flat, surrounded by the cleft epidermis, pulverulent, cinnamon ; secondary amphigenous, without spots, about ^ — 1 mm. diam., scattered or gregarious; spores globose to ellipsoid, echinulate, pale-brown, 26 — 30 x 22 — '27 fx, with three (rarely four) germ- pores. Teleutospores. Sori amphigenous, generally hypophyllous, without spots, scattered or in places densely gregarious, \ — 1 mm. diam., rounded, pulverulent, black-brown ; spores ellipsoid, rounded at both ends, not thickened, gently constricted, delicately verruculose, dark-brown, 30—42 x 22 — 27 //, ; pedicels hyaline, short. On Arctium Lappa. September, October. A doubtful native ; I have seen no British specimens. Description after Sydow. (Fig. 90.) Distinguished from P. Cirsii by its larger teleutospores and by its two kinds of uredo-sori. Jacky demonstrated the distinctness of the two species by experimental cultures (Centralbl. f. Bakt. 2. ix. 796) ; also that P. Bardanae could not be transferred to Taraxacum. Distribution : Europe only. Fig. 90. P. Bardanae. Te- leutospores, on A. Lappa (Austria). ON COMPOSITE 139 12. Puccinia tinctoriae Magn. Puccinia Compositarum var. Serratulae Cooke, Exsicc. no. 33. P. tinctoriae Magn. in Abhand. Nat. Gesell. Niirnberg, 1900, xiii. 37 ; Ber. deutsch. Bot. Gesell. 1893, pi. 21, f. 27—8. Sydow, Monogr. i. 150, f. 127. Uredospores. Sori amphigenous, on inconspicuous pallid spots, scattered, minute, roundish, pulverulent, dark-cinnamon ; spores globose to ellipsoid, echinulate, brown, 24 — 40 x 19 — 26 /*; epispore with two germ-pores towards the summit. Fig. 91. P. tinctoriae. Teleutospores and uredospore (from Cooke's specimens). Teleutospores. Sori similar, but blackish ; spores ellipsoid to oblong, rounded at both ends, not thickened above, hardly constricted, delicately verruculose, brown, 27 — 60x19 — 30 yu,; pedicels hyaline, up to 14^ long; epispore thin. On leaves of Serratula tinctoria. Apparently very uncom- mon; Belton Wood, Highgate (M. C. Cooke). Teleutospores, July — September. (Fig. 91.) Distribution: France, Sweden, Germany, Austria, Italy, Siberia. 13. Puccinia Centaureae DC. Puccinia Compositarum Scb. ; Cooke, Handb. p. 499 ; Micr. Fung. p. 206 p.p. P. Centaureae DC. Flor. fr. vi. 595. Cooke, Micr. Fung. p. 207. Plowr. Ured. p. 186. Sydow, Monogr. i. 39. Fischer, Ured. Schweiz, p. 222, f. 174—5. 140 IM< i INI \ Spermogones. Chiefly epiphyllous, in little clusters, orange. Uredospores. Sori generally hypophyllous, on yellowish or brownish spots (spots sometimes aone), primary rather large, secondary very minute, much scattered, pulverulent, brown : spores globose to ellipsoid, acu- leolate, brown, 22—30 x 16 — 28 ix, with three (or two) germ- pores. Teleutospores. Sori similar, but blackish-brown ; spores ellipsoid or somewhat obovate, rounded at both ends, not thick- ened above, not or hardly constricted, delicately verruculose, chestnut-brown, 24—40x16 — 25 /*; epispore thin; pedicels hyaline, thin, generally very short. Fig <)'2. P. . Centaureae. Teleuto- spores and uredospore. On Centaurea nigra. Teleutospores, August- Common. (Fig. 92.) November. As Plowright says, there are two generations of uredospores. The primary uredospores, which are the equivalent of the secidiospores, accom- panied by spermogones, appear in May, and are followed by the secondaiy uredospores, which form much smaller sori. The former, however, do not occur on every infested plant, being only produced from direct infection by the basidiospores. The primary uredospores are widely scattered by the wind ; Plowright proved that they produce the secondary spores in about fourteen days. Fischer and Jacky record this also on ('. Scabiosa, but the Puccinise on other species of Centaurea arc considered to be distinct species or else biological races. Distribution: Europe. Asia Minor, Siberia, North America. 14. Puccinia Cyani Pass. Uredo Cyani Schleich. Plant. Helv. 95. Puccinia suaveolens var. Cyani Winter, Pike Deutschl. p. 190. Plowr. Ured. p. 183. Sacc. Syll. vii. 633. P. Cyani Pass, in Rab. Fung. Eur. 1767. Sydow, Monogr. i. 38. Spermogones. A few mixed with the primary uredo-sori. Uredospores. Sori generally hypophyllous, without spots, scattered or crowded, minute, roundish, pulverulent, cinnamon: ON COMPOSITE 141 spores globose to ellipsoid, delicately echinulate, yellow-brown, 22 — 30 x 19 — 24 /j,, with two germ-pores. Teleutospores. Sori amphigenous and on the stems, scattered, minute, punctiform, pulverulent, blackish-brown ; spores broadly ellip- soid, rounded at both ends, not thickened above, not constricted, very delicately verruculose, chestnut- brown, 30 — 35 x 22 — 27 //., sometimes longer ; pedicels hyaline, short. & t J Fig. 93. P. Cyani. On Centaurea Cyanus. Not un- Teleutospores. common in gardens. June — September. (Fig. 93.) There is a similar species, P. montana, which occurs on the continent on Centaurea montana ; it is distinguished by the much coarser warts of the teleutospores, which are also relatively longer. Both these species, like P. obtegens, permeate the whole plant with the mycelium of the primary uredospores, but such infested plants of Centaurea can flower freely, while those of Cirsiwm arvense never do. The mycelium of the secondary uredospores, mixed with teleutospores, is more strictly localised. Distribution : Western and Central Europe. 15. Puccinia Carduorum Jacky. Puccinia Hieracii Plowr. Ured. p. 185 p.p., see also p. 216. P. Carduorum Jacky, Composit.-Puccin. p. 58. Sacc. Syll. xvi. 297. Sydow, Monogr. i. 33. Uredospores. Sori generally hypophyllous, on very indis- tinct spots, scattered, minute, pulverulent, cinnamon ; spores globose or subglobose, densely echinulate, pale-brown, 22 — 28 p, with three germ-pores. Teleutospores. Sori similar, but dark-brown; spores variable, oblong to ovate, rounded above, not thickened, hardly at all constricted, rounded below, verruculose, brown, 25 — 38 x 17 — 24 fx ; epispore thin ; pedicels hyaline, short. On Carduus crispus, G. nutans. Not common ; Yorkshire, Clare Island, etc. 142 PUCCINIA This species was formerly undistinguished from the numerous forms on allied species of Composite, until Jacky proved experimentally (in l.s!»:i tliai ii i- confined to the sub-genus Carduus, and could doI be transferred to Cnicus or Cirsium. The uredo form described above (after Sydow) is the secondary uredo ; the primary form, which does not always occur, is described as ampbigenous and larger, often on the mid-ribs 2 — 4 mm. long, and remaining long covered by the epidermis. Distribution : Europe and Siberia. 16. Puccinia Cardui-pycnocephali Syd. Puccinia Cardui-pycaocepltaU Sydow, Monogr. i. 34, f. 35. Massee, Journ. Bot. xlvi. 152 ; Trans. Brit. Myc. Soc. (1909), iii. 123. Uredospores. Sori hypophyllous, without spots, scattered, minute, pale-brown; spores globose or subglobose, very deli- cately echinulate, pale-brown, 22 — 26 p,. Teleutospores. Sori similar, hidden in the tomentum of the host, brown ; spores oblong, rounded above and more or less so below, not thickened, hardly constricted, smooth, pale-brown, 38 — 50x16 — 23//,; epispore thin; pedicels thin, hyaline, up to 40 \i long, but deciduous. On leaves and stems of Carduus pycnocephalus. Very rare; Sidmouth (Dr Mayor), and between Eastbourne and Hastings (G. Massee). Only recorded for Britain and Italy. July, August. Distinguished from P. Carduorum by its longer teleutospores, which are smooth, and not plainly verruculose as in that species. But sometimes the spores are said to approach those of P. galatica Sydow, which occurs on the same host in Asia Minor, in being smaller, darker-coloured, thick- walled, and delicately punctate. According to Massee (I.e.) these are rather two poles of one species than two distinct si>ecies. 17. Puccinia Cirsii Lasch. Trichobasis Cirsii Lasch : Cooke, Micr. Fung. p. 224. Puccinia Cirsii Lasch, in Rab. Fung. Eur., no. 89. Cooke, Grevillea, iv. 109 ; Micr. Fung. p. 206. Sydow, Monogr. i. 55. Fischer, Ured. Schweiz, p. 217, f. 168 — 171. Uredospores. Sori epiphyllous or hypophyllous, on paler spots, scattered, minute, girt by the epidermis, pulverulent, cin- namon ; spores globose to ovate, echinulate, pale chestnut-brown, ON COMPOSITE 143 22 — 28 x 19 — 24 /*, with three more or less equatorial germ- pores each provided with a thickening. Fig. 94. P. Cirsii. Teleuto- spores, on C. palustre, from Hereford. Fig. 95. P. Cirsii. Teleuto- spore and uredospore, on G. lanceolatum. Teleutospores. Sori mostly hypophyllous only, similar, but blackish-brown or black ; spores ellipsoid or somewhat obovate, rounded at both ends, not thickened above, hardly constricted, verruculose or merely punctate, chestnut-brown, 25 — 38 x 17 — 25 /it ; epispore thin; pedicels hyaline, very short. On Cirsium, Dupplin Castle, Perth (M. C. Cooke). On G. pratense, Ballyquirke Lake, Co. Gal way (communicated by J. Adams); Earlswood Lakes, near Birmingham. On G. palustre, Hereford, Seckley Wood, Barnt Green, etc. Uredospores from March ; teleutospores, June — November. (Figs. 94, 95.) There is no mention of this in Plowright's Uredinese, or in the Trans. Brit. Myc. Soc. (Plowright's list), but it is probably not uncommon. It occurs frequently on the radical leaves, and can be easily distinguished from P. Cnici-oleracei by the presence of the uredospores and the non- circinate teleuto-sori, as well as by the absence of the apical thickening. Fischer records it from Switzerland on many species of Cirsium (but not on those mentioned here), and also assigns to it sperrnogones on the upper leaf-surface and petiole ; I have not been able to find any trace of these in our British specimens. The uredospores, seen in water, sometimes appear quite smooth, as Cooke describes them. The teleutospores have the upper pore at the summit or at the side, the lower pore just beneath the septum or lower down ; they are at times faintly granulated, at others distinctly verruculose. Mesospores are rare. I have also a number of specimens on C. lanceolatum from Droitwich, Wyre Forest, etc., bearing a great similarity to P. Cirsii-lanceolati Schrbt., but the differences from P. Cirsii are so slight and elusive that, as the secidial stage by which the former is distinguished has not yet been found in Britain, it is better for the present to place them under P. Cirsii. Distribution : Europe, Siberia, Japan, North America. 144 PUCCINI A L8. Puccinia Cnici-oleracei lVrs. Puccinia Cnici-oleracei Pers. in Desm. Cat. dea PI. omis., p. 24. Sydow, Monogr. i. 58, pi. 3, f. 17. Fischer, Ured. Schweiz, p. 292, 594, f. 213. /'. ( 'ardui Plowr. Ored. p. 216. Teleutospores. Sori hypophyllous, on discoloured roundish spots, minute, circinate, l»ut mostly confluent in large patches as much as 5 mm. diam. which remain for a long time covered by the epidermis, compact, blackish-brown; spores clavate or subfusiform, rounded or rarely somewhal conical above, very Fig. 96. P. Cnici-oleracei. Teleutospores, on C. lanceolatum. much thickened (5 — 10 /x), constricted, attenuated below, smooth, yellowish-brown, 38 — 56 x 14 — 21 /u,; pedicels yellowish- hyaline, thick, persistent, as much as 50 //. long. On Cirsium lanceolatum, Carduus crispus (?). August — October. Not uncommon on the former. (Fig. 96.) This is one of the species in which the upper cell separates from the lower one. It is doubtful if Puccinia Syngciiesiarum (Cooke, Handbook, p. 499 ; Micr. Fung. p. 206) belongs entirely here, as the figure in the latter work (pi. 4, f. 64) does not give the true form of the teleutospore ; but the majority of the specimens issued by him under that name are this species. It is very questionable if this species occurs on Carduus crispus, on which it is recorded by Plowright. Distribution : in a few places in Europe. ON COMPOSITE 145 19. Puccinia obtegens Tul. Caeoma obtegens Link, Obs. ii. 27. Trichobasis suaveolens Lev. ; Cooke, Handb. p. 530 ; Micr. Fung. p. 226, pi. vii. figs. 151—3. Puccinia obtegens Tul. Ann. Sci. Nat. ser. 4, ii. 87 (1854). Sydow, Monogr. i. 53. P. suaveolens Host. Bot. Zeit. 1874, p. 556. Plowr. Ured. p. 182 p.p. Sacc. Syll. vii. 633 p.p. Fischer, Ured. Schweiz, p. 219, f. 172. Spermogones. Chiefly hypophyllous, a few epiphyllous, crowded, covering the whole surface of the leaf, of a bright honey-yellow colour and a pleasant smell. Uredospores. Primary sori hypophyllous, occupying the whole surface of the leaf, minute, crowded, often confluent,, pulverulent, reddish-brown, then darker ; secondary, more scat- tered ; spores globose to broadly ellipsoid, echinulate, pale- brown, 21 — 28 fx, with three irregularly placed germ-pores. Teleutospores. Sori similar, always dark-brown ; spores ovate to ellipsoid, rounded at both ends, or somewhat tapering below, not thickened above but with a low flat pore-cap, hardly constricted, delicately verruculose, brown, 26 — 42 x 17 — 25/Z; epispore thin; pedicels hyaline, thin, short. Fig. 97. P. obtegens. On Cirsium arvense (Carduus arvensis). Teleutospores. Very common. (Fig. 97.) The life-history of this species is peculiar. In spring the mycelium permeates the host in every part. The affected plants can be recognised immediately by their pale-green colour and spindly appearance ; they never flower. The spermogones are first seen towards the end of April, and are easily detected by their bright colour, and their strong perfume, resembling that of privet-flowers ; the uredospores follow on the same leaves during May. From these primary uredospores, a second generation arises on other plants about July, and forms secondary uredospores and telexitospores in sori which are more scattered, never confluent, and darker brown. This generation is not accompanied by spermogones. The mycelium of these sori is localised to the infected spots and the host does not suffer so severely. The sori of the primary uredospores rarely bear a few teleutospores intermixed, but the secondary sori abound with them from .September to November, and it is from the germination of G. U. 10 L46 PUCCINIA these latter that new infections arise in the spring as 30011 as the shoots appear. The hibernation of the mycelium in the rhizome, which is si by Plowright, lias oot been proved. The LriTiu-pures arc very easj to see in the uredospores of this - ] .< •< 1 . - ami its allies. Each is often surrounded by a border like a bordered pit, an appearance caused by a thickening of tin- membrane around and over the pit. As visual the appearance of the spore changes according a- it is wet or dry ; if wet, it may appear merely punctate; if dry. it is seen to be densely ami coarsely echinulate. Distribution": Northern and Central Europe. Siberia, North America. 20. Puccinia Andersoni B. et Br. Puccinia Andersoni B. et Br. Ann. Xat. Hist, ser. 4, xv. 35. Cooke, Micr. Fung. p. 206; Grevillea, iii. 179. Plowr. Cred. p. 204. Sacc. Syll. vii. 710. Sydow, Monogr. i. 58. Teleutospores. Sori hypophyllous, seated on round yellow spots 1 — 1^ cm..diam. with a brown border, almost concealed by the pubescence of the leaf, minute, but densely crowded in flat circular clusters which are few on each leaf, compact, blackish-brown or purplish- black: spores oblong to clavate, rounded or conically thickened (8 — 10 /la) above, slightly constricted, smooth, pale-brown, 40 — 54 x 16 — 22 /x ; pedicels browTiish, stout, per- sistent, as long as the spore or longer; a number of mesospores are found intermixed. Fig. 98. P. Andersoni. Teleu tospores and mesospore. On Cirsium hetero'phyllum. June — October. A striking and rare species. Glen Ogle (Rev. M. Anderson). Den of Airlie (Mr Gardiner), Ingleton and Grassington, Yorkshire (H. T. Soppitt), Alston, Cumberland (J. G. Baker). (Fig. 98.) DISTRIBUTION : in a few places in Europe. ON COMPOSITE 147 21. Puccinia Lapsanae Fckl. JEcidium Lapsanae Schultz, Prod. Flor. Starg. p. 454. Cooke, Journ. Bot. ii. 38, pi. 14, f. 2 ; Handb. p. 543 ; Micr. Fung. p. 198. Trickobasis Lapsanae Cooke, Micr. Fung. p. 224. Puccinia iMpsanae Fckl. Synib. Myc. p. 53. Cooke, Micr. Fung, p. 207. Plowr. Ured. p. 149. Sacc. Syll. vii. 607. Sydow, Monogr. i. 112. Fischer, Ured. Schweiz, p. 203, f. 159. Spermogones. Crowded in little clusters, epiphyllous, honey- coloured. jEcidiospores. vEcidia amphigenous, somewhat crowded on large, roundish, purple spots, fiattish, with torn white rerlexed margin ; spores subglobose or ovate, nearly smooth, orange, 16—21 x 13—17 ft. Uredospores. Sori amphigenous, very minute, round, very numerous, often confluent, pulverulent, chestnut- brown ; spores globose to ovate, delicately echinu- late, pale-brown, 17 — 22 x 15 — 18 /n, with two germ-pores. Teleutospores. Sori amphigenous, minute, scattered, numerous, pulverulent, blackish-brown; Fig 99 F spores ellipsoid or ovate, rounded at both ends, Lapsanae. . Teleutospore. not thickened above, scarcely constricted, very delicately punctate, chestnut-brown, 22 — 33 x 17 — 26 /x; pedi- cels hyaline, slender, short, often oblique. On leaves and stems of Lapsana communis. Very common. iEcidia, March — May; uredospores, April — June; teleutospores, June — September. (Fig. 99.) The aecidium of this species shares with those on Ranunculus Fiearia and Potentilla Fragariastrum the distinction of being the earliest Uredine to make its appearance in the spring. It may be found on the young leaves of the seedlings almost as soon as they are formed. Plowright demonstrated (Expt. 499, May, 1885) that all three spore-forms belong to the same life-cycle. The mycelium of the aecidia, when occurring on the petioles, causes them to become pale and swollen ; on the leaves it often produces conspicuous purple spots which bear the spermogones on the upper surface. Distribution : Europe, Syria, Japan. 10—2 148 PUCCINIA 22. Puccinia Cichorii I'm 11. JJredo Cichorii l)('. Ror. fr. vi. 7 1. Puccinia Hieraoii Mart. ; Plowr. Ured. }>. 184 p.p. /'. Cirlmrii Bell, in Kickx, Fl. Fland. ii. (;.">. Sydmv, Monogr. i. 49. Fischer, Ured. Schweiz, p. 227, f. 17'.). McAlpine, Rusts of Australia, p. L54, f. 61 and pi. J), f. 18. Credosjiores. Sori amphigenous or on the steins, numerous, scattered, sometimes confluent, minute, pulverulent, surrounded lty the torn epidermis, cinnamon ; spores globose to ellipsoid, echinulate, yellowish-brown, 21 — 27 //.; epispore moderately thick, with two germ-pores. Teleutospores. Sori similar, but blackish-brown ; spores ellipsoid, rarely obovate, rounded above and not thickened, hardly constricted, generally rounded below, smooth or faintly verruculose, brown, 27 — 38 x 19 — 25 /x; epispore thin; pedicels hyaline, short, deciduous. On Cichorium Intybus. Rare. August — October. This species closely resembles Puccinia Hieracii, and there seem to have been no cultures which give any information about its distinctness. It is therefore separated merely on the ground of the difference of the genus of the host ; see under P. Hieracii. Distribution : Europe ; introduced into Australia. 23. Puccinia Hypochceridis Oud. Puccinia Hj/poc/iocridis Oud. in Nederl. Kruidk. Archief. ser. 2, i. 175. Sydow, Monogr. i. 100. Fischer, Ured. Schweiz, p. 232. McAlpine, Rusts of Australia, p. 159, f. 62 — 3. Uredospores. Sori amphigenous or on the stems, generally on minute spots, scattered, pulverulent, cinnamon, primary sori rather large, secondary minute ; spores globose to ellipsoid, echinulate, brown, 22 — 28 /x, with two germ-pores. Teleutospores. Sori similar, but punctiform (chiefly and larger on the stems), black; spores ellipsoid to ellipsoid-obovate, rounded at both ends or rarely attenuated downwards, not thickened above, hardly constricted, delicately verruculose- punctate (?), brown, 30 — 46 x 18 — 24//,; epispore thin; pedicels hyaline, short. ON COMPOSITE 149 On Hypochoeris radicata. Not uncommon. July — Sep- tember. The records from other countries include all the three British species of Hypochoeris. Only uredospores were seen in British specimens. (Figs. 100, 101.) Fig. 100. P. Hypoehoeridis. Uredospore (British) on H. radicata. Fig. 101. P. Hypoehoeridis. Teleuto- spores aud uredospore, on H. glabra (Berlin, ex herb. Sydow). This species, which has been often stated to be a Brachypuccinia, differs from most others of the type of P. Hieracii in having two kinds of uredo-sori — the primary ones 1 — 2 mm. wide, and only on the leaves, the secondary ones conspicuously smaller, almost punctiform. In this respect it approaches P. Cyani and P. obtegens, but Jacky showed that in his cultures it did not produce spermogones (Centralbl. f. Bakter. 1907, xviii. 83), as they do. The alleged punctations of the teleutospores were invisible in all the specimens I have seen. Distribution : Europe, Siberia, North America, Chili, Australia. 24. Puccinia Leontodontis Jacky. Puccinia Hieracii Plowr. Ured. p. 184 p.p. P. Leontodontis Jacky, Composit.-Puccin. p. 75. 114 Fischer, Ured. Schweiz, p. 231, f. 182. Sydow, Monogr. i. Uredospores. Sori amphigenous, but chiefly hypophyllous, scattered, not confluent, minute, punctiform, surrounded by the cleft epidermis, cinnamon ; spores globose to ellipsoid, echinu- late, brown, 25 — 32 ^ diam. or 27 — 35 x 24 — 27 /x, always with two opposite germ-pores above the equator. Teleutospores. Sori similar, but darker-brown; spores variable, ellipsoid to oblong or obovate, rounded at both ends, not thickened above, not constricted, delicately verruculose or 150 IT. (IMA faintly granulated, chestnut-brown, 30 — 42x21 — 27/u: epispore thin : pedicels hyaline, short- or nearly as long as the spore. Fig. 102. P. Leontodontis. Teleutospores and uredospore, on L. autumnalis. On Leontodon autumnalis, L. hispidus. August, September. Not uncommon. (Fig. 102.) The teleutospores of this species are said to be more than usually variable ; sometimes the sori, according to Sydow, are seated on coloured spots, but more often the spots are wanting. Up to the present, no experimental cultures appear to have heen carried out with this Puccinia, and it is separated from P. Hieracii mainly because it is parasitic upon a different genus. I have found it mostly upon old yellowing leaves. Distribution : Europe generally. 25. Puccinia Tragopogi Corda. J^idiiim Trtit/ii/i<>•">, pi. 1, f. 1—3. Puccinia Trugnpogi < 'onla. Icon. v. 50. Plowr. bred. p. 197. Saec. Syll. vii. 668 p.p. Sydow, Monogr. i. 167. Fischer, Ured. Schweiz, p. 215, f. 166. P. sparsa Cooke, Handh. p. 498 ; Micr. Fung. p. 205. Spermogones. Epiphyllous, honey-coloured, sometimes absent. sEcidiospores. iEcidia hypophyllous, without spots, scat- tered uniformly over the whole surface, and on other green parts, cup-shaped, with a white torn revolute margin ; spores globose to ellipsoid, densely verrucose, pale-orange, 20 — 30 x 18 — 24^, with three germ-pores. ON COM POSIT. K 151 Teleutos pores. Sori amphigenous, without manifest spots, scattered, occasionally aggregated, mi- nute, punctiform, long covered by the epidermis, at length pulverulent, dark- brown ; spores ellipsoid, rounded at each end, not thickened above, gently constricted, tuberculate, chestnut- brown, 26—45x18—32^; pedicels Fi8- ^ ?• Tragopogi. r L leleutospores. hyaline, short. On Tragopogon pratensis. Not uncommon. April — Sep- tember. (Fig. 103.) The mycelium arising from the infection of young plants by the basi- diospores permeates the whole of the host, so that secidia are produced on every part — stems, leaves, bracts, and receptacles — and the infected plants are noticeable for their paler colour and distorted form. It is believed to hibernate also in the upper part of the root-stock. The mycelium of the teleuto-sori is, on the contrary, strictly localised. Uredospores are not' produced in separate sori, but a few can be found intermixed with the teleutospores, as well as a few mesospores ; the former can be distinguished from the latter by * being delicately aculeate, not tuberculate. The teleutospores of this species are grossly and distinctly warted, reminding one of those of Uromyces tuberculattts ; they can be found at the same time as the fecidia, but are much rarer, or at least less frequently observed. Distribution : Europe, Asia Minor. 26. Puccinia Chondrillae Corcla. Mcidium Prenanthis Pers. Syn. p. 208 (non Cooke, Handb. p. 542 ; Micr. Fung. p. 198). Puccinia Chondrillae Corda, Icones, iv. 15. Fischer, Ured. Schweiz, p. 200, f. 156. P. Prenanthis Auct. p.p. ; Plowr. Ured. p. 148. Sydow, Monogr. i. 106. McAlpine, Rusts of Australia, p. 162 (?). uficidiospores. iEcidia hypophyllous or on the petioles, rarely a few on the upper surface, seated on large yellow and purple spots, in clusters as much as 6 — 8 mm. broad, at first hemispherical, opening by a pore, then flattened, whitish or yellow, sometimes with a purplish tinge ; spores globose to ellipsoid, delicately verruculose, pallid-orange, 13 — 24//,; peri- dium poorly developed. L52 PUCCINIA I redosjx.res. S«»ri hypophyllous, on pallid irregular spots, scattered, minute punctiforra, pul- verulent, pallid-brown : spores more or less globose, echinulate, yellow- brown, lb' — 24 fji. with three (rarely four) germ-pores. Teleutospores. Sori similar, surrounded bv the cleft epidermis, "tSU'JSSSS-.* blackish-brou",, : spores ellipsoid, rounded and not thickened above, not constricted, rounded below, very delicately verruculose, brown, 26 — 36 x 16 — 24 /x ; pedicels hyaline, very short. On Lactuca (Prenanthes) muralis only. Not common. April — September. All three spore-forms may be found on the same leaf. (Fig. 104.) The secidium of this species is not provided with a proper peridium and opens with a pore, never assuming the form of a cup ; moreover the secidiospores are more round than angular. The germ-pores of the uredo- spores are covered with a broad convex colourless cap, which swells up somewhat in water. The genetic connection of the three spore-forms has been proved experimentally by Jacky (Ceutralbl. f. Bakt. ix. 1902, p. 842), as also the fact that the parasite is not transmissible to other specie Lactuca. There is another secidium also found on Lactuca mnralis on the continent which possesses a well-developed peridium ; this belongs to Puccinia Opizii Bubak, and has its teleutospores on Carex muricata. Distribution : Europe generally. 27. Puccinia variabilis Grev. JEJcidium Grevillei Grove, Journ. Bot. 1886, p. 129. Puccinia variabilis Grev. Scot. Crypt. Fl. pi. 75. Cooke, Handb. p. 500 ; Micr. Fung. p. 207, pi. 4, f. 82—3. Plowr. Ured. p. 150. Sydow, Monogr. i. 163. Fischer, Ured. Schweiz, p. 202, f. 158. jEcidiospores. iEcidia amphigenous, on minute indetermi- nate yellow or purplish spots, solitary or a few loosely aggre- gated, cup-shaped, with whitish torn margin; spores subglobose or ovate, with orange contents, verruculose, 20 — 25 x 15 — 20 /x. Uredospores. Sori amphigenous, on very minute yellow or ON COMPOSIT/E 153 purplish spots, scattered, minute, punctiform, soon naked, brown; spores few, globose to ovate, echinulate, brown, 22— 32 x 19 — 26yti, with two germ-pores. Teleutospores. Sori similar, but darker ; spores ellipsoid to oblong, rounded at both ends, not thickened above, not con- Fig. 105. P. variabilis. /Ecidia, on leaf of Taraxacum, and teleutospore. stricted, delicately verruculose, brown, 28 — 40x18 — 25 /*; epispore thin ; pedicels hyaline, about as long as the spore, but deciduous. On Taraxacum officinale and its variety palustre. July — October. Not common. (Fig. 105.) Plowright and Soppitt both proved, Dy laying leaves affected with the aecidium of this species on healthy plants of Taraxacum, that the uredo- and teleutospores were produced in about fourteen days. In July the three spore-forms may be found on the same leaf. There are two forms of eecidium found upon Taraxacum ; one, jffcidium Grevillei Grove ( = jE. Taraxaci Grev. non K. et S.), spreads pretty uniformly over the whole leaf in "numerous little clusters with single ones scattered between them," as Greville describes it (Flor. Edin. p. 444) — the other, SE. Taraxaci K. et S., forms large round clusters, and belongs to P. silvatica. Fischer points out that the two recidia differ in the form of their peridium cells, those of P. variabilis having the membrane thickened on the inner side, while those of P. silvatica have the outer wall most strongly thickened. He states, furthermore, that it will be found that this difference is characteristic in general of autcecious and hetercecious species respectively. It is not, however, universally so, e.g. the secidium of P. albescens has the outer wall much more strongly thickened, although it is autcecious. Greville figures the teleutospores of his species (Scot. Crypt. Flor. pi. 75) as having either one or both of the cells sometimes divided by a vertical septum. Distribution : Switzerland, Sweden. L5 1 PI I 'IMA 28. Puccinia Taraxaci Plowr. Puccinia Phaseoli vax. Taraxaci Etebent. Fl. Neomarch. j>. 356. /'. variabilis n alpine plants), if they are the result of an adaptation to a shortened vegetative period. Uromycea Fica/riae may be taken as another instance of a similar character. 35. Puccinia albescens Plowr. JEcidium albescens Grev. Fl. Edin. p. 444. Cooke, Handb. p. 536 : .Micr. Fung. p. 194. Puccinia albescens Plowr. Ured. p. 153. Fischer, Ured. Schweiz, p. 144, f. 110. P. Adoxae Hedw. ; Cooke, Micr. Fung. p. 209 p.p. Sacc. Syll. vii. 612 p.p. Sydow, Moiio»t. i. 203 p.p. Spermogones. Scattered among the i fi) conical point at the apex, constricted, slightly tapering below, smooth, pale or dark chest- nut-brown, 28 — 40 x 14 — 19 /x : pedicels hyaline or somewhat yel- lowish, rather long, deciduous or persistent. On Veronica alpina, V. offici- nalis and perhaps other species. Not common. July — October. (Fig. 120.) There are two forms of spores in this species ; one, //•"//////".<, with deciduous pedicels ; the other, penlst <■,>.<, with persistent pedicels. In the former, the sori are soon naked and pulverulent, the spore-wall is thicker and darker-coloured, and germination takes place only after a winter's rest. In the latter, the sori are compact, the spores are thin- walled and paler and germinate as soon as mature, upon the living plant. But occasionally both forms may be seen in the same sorus. It is the latter form that spreads the parasite during the summer ; the former causes new infections in the following spring. Distinguished from P. Veronicae by the relatively broader spores, and the cone-shaped apical thickening. It is possible that the Puceiniae on other species of Veronica, included by the Sydows with this, are really distinct. But in Herb. Berkeley there is a Puccinia on Veronica alpina, from Ben Aulder, Invernessshire (and also from Perthshire) which is undoubtedly this species. The sori are hypophyllous only, but cover the larger part of each leaf. The spores are quite smooth, by which it is distinguished from the continental form on V. a/pixa, which has the upper spore-cell occasionally warted on the upper part ; this form (P. albulensis Magn.) has the sori densely gregarious on stems as well as leaves, some- times covering a whole internode. Distribution : Europe generally. 43. Puccinia Menthae Pers. JEcidium Menthae DC. Flor. fr. vi. 95. Cooke, Handb. p. 544 ; Micr. Fung. p. 199. Tridwbasis Labiatarum Lev. ; Cooke, Micr. Fung. p. 224 p.p. T. Clinopodii DC. ; Cooke, Micr. Fung. p. 224 (?). ON LABIATvE 171 Puccinia Menthae Pers. Syn. p. 227. Cooke, Handb. p. 496 ; Micr. Fung. p. 204 p.p., i>l. 4, f. 69, 70. Plowr. Ured. p. 157. Sacc. Syll. vii. 617. Sydow, Monogr. i. 282. Fischer, Ured. Schweiz, p. 168, f. 131. McAlpine, Rusts of Australia, p. 140, f. 250. P. Clinopodii DC. ; Cooke, Micr. Fung. p. 205 (?). Spermogunes. Scattered or arranged in little groups, honey- coloured. jEcidiospores. /Ecidia hypophyllous and often on the stems, arranged on the leaves in clusters on orange or purplish spots, or forming elongated patches on the stems and petioles which are much thickened and deformed, opening irregularly, margin scarcely torn, erect or even incurved ; spores verruculose, pallid-yellow, 24—40 x 17— 28/*. Uredospores. Sori hypophyllous, on yellowish or brownish spots (or without spots), minute, roundish, scattered or aggre- gated, soon naked, surrounded by the ruptured epidermis, sometimes confluent, cinnamon; spores globose to obovate, echinulate, pallid-brown, 17 — 28 x 14 — 19/x, with three equatorial germ-pores. Teleutospores. Sori similar, but dark-brown; spores subglobose to obovate, rounded at both ends, with a broad pale-coloured apical papilla, not or scarcely constricted, more or less in- distinctly verruculose, sometimes smooth, dark- brown, 26 — 35x19— 23 /*; pedicels hyaline, slender, longer than the spore. P. Menthae. • ht n/r Teleutospore On Mentha aquatica, M. arvensis, M. citrata, on M. aquatica. M. rotundifolia, M. silvestris, M. vir'idis, Ori- ganum vulgare, Galamintha Clinopodium, C. officinalis. May — October, teleutospores from August. Very common on garden Mint, rather common on some of the other species. (Fig. 121.) There can be little doubt that this is a collective species. Points of difference are found in the finer or coarser warts of the teleutospores and in the length of the pedicel ; but hitherto no certainty has been arrived at in delimiting the various forms. I have seen the warts quite distinctly on some of the teleutospores on M. aquatica and M. arvensis, when they are viewed dry, but other spores in the same sorus seemed perfectly smooth. The secidium seems not to occur on all species, though it is common on 172 PUCCINIA garden mint ; perhaps those forms which are without it may hereafter be separated. l>ut it has occurred on all the hosts mentioned above except mum ami M. rotundifolia ; it may, however, according to Sydow, be mcnly facultative. The formon C. < linopodium really shows less difference from that on Mentha aquatica than those on .species of Mentha do from one another. But Cruchet was unable to infect any one of the four M. arvensis, M. ,/,/,/, /tint, M. silvestris, C. C(inopodiu7n, except by spores from the same species. As the result of his experiments, he divides /'. Menthae into eight biological races, as it occurs on Mentha and CaZamintha; and the form "ii Origanum, is also biologically distinct. The Australian form of /'. Menthae, which is an introduced species on M. Pulegium and M. laxv- flora, has no known aecidiospores, but occasional mesospores. Nothing seems to be known about the form on Ajuga reptans mentioned by Plowright, from Johnston's Flor. Berwick. In garden mint (J/, viridis) the mycelium of the aecidial stage is spread throughout the whole plant, even in the rhizome ; Klebahn was aide to trace the hypha? in some cases nearly up to the growing point. It lasts for several years at least ; a bed of mint infested with it should be rooted tip ami burnt ; there is no cure for the disease, although I have found that cuttings taken from some of the more distant healthy-looking shoots and planted i/.tiiekere grow up without the parasite. The mycelium of the two other stages is purely local. I have known the secidia to occur for several years in a garden without being followed by uredo- or teleutospores so far as could be seen, and vice-versa, in another case, these spores occurred but no secidium was ever noticed. Distribution : Europe, Asia, Africa ; the American and Australian teleutospores are more strongly warted. 44. Puccinia caulincola Schneid. Puccinia caulincola Schneid. in Jahresb. Schles. Gesell. 1870, p. 120. Sydow, Monogr. i. 301. Fischer, Ured. Schweiz, p. 172, f. 133. P. Xrhwideri Schrot. in Herb. Schles. Pilze, no. 448. Plowr. Ured. p. 201. Sacc. Syll. vii. 677. Teleutospores. Sori on the stems and petioles, rarely on the leaves, scattered, occasionally confluent, minute, roundish or elongated, long covered by the bullate epidermis, at length pulverulent, black, then cinnamon-brown ; spores ellipsoid, rounded at both ends, apex sometimes thickened in a papilli- form fashion, rather constricted, smooth, pale-brown, 24 — 33 x 15 — 24 fx: pedicels hyaline, thin, rather long, not very ON LABIAT.E 173 deciduous; a few mesospores intermixed, each with a low broad pore-cap. On Thymus Serpyllum. June — October. Very rare ; links, Aber- deen (Prof. J. W. H. Trail). (Fig. 122.) The presence of the mycelium causes the stems to stand more upright ; the internodes are considerably lengthened, and the leaves fewer, so that the affected plants can be readily distinguished, as in P. Betonicae. The mycelium appears to be perennial, and the sori to be confined almost entirely to the stems, where they cause a slight thickening and are more frequent at the nodes than elsewhere. Distribution : Central and North-western Europe. Fig. 122. P. cauUncola. Teleutospores and mesospore. 45. Puccinia Glechomatis DC. P. Glechomatis DC. Encycl. viii. 245. Cooke, Handb. p. 496 ; Micr. Fung. p. 204, pi. 4, f. 73—4. Plowr. Ured. p. 214. Sacc. Syll. vii. 688. Sydow, Monogr. i. 277. Fischer, Ured. Schweiz, p. 327, f. 239. Teleutospores. Sori hypophyllous or on the petioles, on brownish spots or sometimes none, h — 1 mm. diam., roundish, solitary and scattered, or more often subcon- fluent into rounded clusters as much as 4 mm. diam., on the stem and petioles often elongated, pulvinate, at first yellowish, then chestnut, and at last blackish ; spores ellipsoid or oblong, with an acute or rounded horn-like process (8 — 12^ high) which is often obliquely placed and falls off on germination, faintly constricted, rounded below, smooth, pale and clear-brown, 30 — 48x15 — 24 /x ; pedicels hyaline, persistent, as much as 75 fi long; an occasional mesospore is found. Fig. 123. P. Glechomatis. Teleutospores and mesospore. 174 PUCCINI A On leaves, petioles and stems of Glechoma hederacea {Nepeta Glechoma). June — October. Not uncommon. (Fig. L23.) The sori are especially large, round and compact late in the season, when they produce spores which are darker and will not germinate immediately (as the others do), but only after the winter's rest. I have a specimen, resembling this species, on Prunclfn cidgari* from Sutton Park, Warwicks. ; Plowright mentions a similar one from Ben Lawers(2.c. p. 215). Distribution : Europe, Siberia, Japan. 46. Puccinia Betonicae DC. Puccini'/ Anemones var. Betonicae A. et S. Consp. p. L31. P. Betonicae DC. Flor. fr. vi. 57. Cooke, Handb. p. 4!)7 ; Micr. Fung. p. 205. Grove, in Card. Chron. xxiv. 180, f. 38 L885 . Plowr. Ured. p. 199. Sacc. Syll. vii. 677 p.p. Sydow, Monogr. i. 275. Fischer, Ured. Schweiz, p. 173, f. 134. Teleutospores. Sori hypophyllous, on pallid irregular spots, numerous, aggregated in patches, or more generally spreading over nearly the whole of a leaf, more or less crowded on the nerves, minute, perfectly round, surrounded by the torn erect epidermis, pulverulent, reddish-brown ; spores ellipsoid to Fig. 124. P. Betonicae. Normal and abnormal teleutospores. ovate, rounded above with a small paler hemispherical pore- cap, slightly constricted, rounded below, smooth, yellow-brown, 27 — 45 x 15 — 24 fj, ; pedicels thin, hyaline, deciduous, about as long as the spore ; a few oval mesospores intermixed. On Betonica officinalis (Stachys Betonica). Not common. May — September. (Fig. 124.) The affected leaves are paler, narrower, and stand more erect than the healthy ones. Besides the mesospores, other anomalies are occasionally ON LABIATE 175 met with, viz. spores with three or more cells variously arranged. See Grove, Gardener's Chronicle, xxiv (1885), p. 180, f. 38. The mycelium is probably perennial. Distribution : Central and Western Europe. 47. Puccinia annularis Schlecht. Uredo annularis Strauss in Wetter. Ann. ii. 106. Puccinia annularis Schlecht. Flor. Berol. ii. 132 (1824). Plowr. Ured. p. 217. Sacc. Syll. vii. 689. Sydow, Monogr. i. 300. Fischer, Ured. Schweiz, p. 329, f. 240. P. Scorodoniae Link. Spec. ii. 72 (1825). Cooke, Handb. p. 497 ; Micr. Fung. p. 205. Teleutospores. Sori hypophy lions, on indefinite yellowish or brownish concave spots, at first minute, roundish, covered by the epidermis, in orbicular clusters, then naked, con- fluent, and forming a thick pulvinate mass, ferruginous-brown ; spores ob- long, rounded or attenuated at the apex and much thickened (up to 8 /j,), slightly constricted, rounded or attenu- ated at the base, smooth, very pale yellowish-brown, 30 — 54 x 14 — 21 //,; pedicels hyaline, persistent, as much as 80 fi, long. On Teucrium Scorodonia. July — October. Not common. (Fig. 125.) Here, as in other Lepto-species, there seem to be two kinds of spores, those which germinate at once and those which will not germinate till the following year. The form on T. Chamaedrys (not yet found in Britain) has been proved to be biologically distinct. Distribution : Central and Southern Europe. P. annularis. Teleutospores. 48. Puccinia Convolvuli Cast. Credo Betae ft. Convolvuli Pers. Syn. p. 221. Puccinia Convolvuli Cast. Obs. i. 16. Plowr. Ured. p. 146. Sacc. Syll. vii. 610. Sydow, Monogr. i. 319. Fischer, Ured. Schweiz, p. 322, f. 234. L76 puccini \ \.l\r id inspires. . Kcidia hypophyllous, on brownish or pur- plish spots, more or less circinate, often on the petioles and then in elongated patches, cup-shaped, minute, with broad recurved born white margin; spores delicately verruculose, pallid-yellow, 17— 28/*.] Vredospores. Sori scattered or circinate, minute, often confluent, soon naked, brown ; spores more or less ellipsoid, rarely ovate, faintly echinulate, pale-brown, 22 — 30x18 — 26/*, with two or three germ-pores just above the middle. Teleutospores. Sori similar, but long covered by the grey epidermis, black- brown ; spores ellipsoid to oblong, ob- tuse or rounded above, more or less thickened (up to 9 /a), gently constricted, rounded below, smooth, chestnut-brown, 38— 66 x 18— 30 /*; with them are intermixed (according to Sydow) ovoid mesospores, much thickened at the Fig. 126. P. Convolvuii. apex, brown, 25 — 35x20 — 26//-; pedi- mlTsephlm, PoughCp- cels brownish, thick, persistent, up to sie, U.S.A. 35 yu, long. On Convolvulus sepiwm (Miss Jelly). June — October. Very rare. (Fig. 126.) The connection of the secidiospores with the teleutospores was experi- mentally demonstrated by Arthur. According to Fischer, the uredospores often have a smooth median equatorial zone, of which I could see no trace. I have not seen the eecidia. Distribution : Europe, Africa, Japan, North America. 49. Puccinia Vincae Berk. Uredo Vincae DC. Flor. fr. vi. 70. Trichobasis Vincae Berk. ; Cooke, Micr. Fung. p. 226, pi. 6, f. 130—1. Puccinia Vincae Berk. Engl. Fl. v. 364. Cooke, Handb. p. 497 ; Micr. Fung. p. 205, pi. 6, f. 132. Plowr. Ured. p. 161, pi. 2, f. 11—14 ; Card. Chron. 1885, xxiv. 108, f. 22—3. Sacc. Syll. ix. 310. Sydow, Monogr. i. 338. Fischer, Ured. Schweiz, p. 167, f. 130. Spermogones. Hypophyllous or ' amphigenous, minute, ON VINCA 177 punctiform, brownish, often very numerous, scattered over the whole leaf-surface, sweet-scented, spherical, about 175 /x diam. Uredospores. Sori hypophyllous, pallid-brown, of two kinds ; primary irregular, often elongated and curved, crowded and confluent, naked ; secondary scattered, on roundish dirty-brown spots, long covered by the epidermis; spores globose to pyriform, aculeate, pallid-brown, 20 — 32 /x diam. or 20 — 46x16 — 24 ll, with three germ-pores. Fig. 127. P. Vincae. a, teleutospore, seen dry ; b, the same, seen wet ; c, uredospore ; d, the so-called a?cidia. On Vinca major, all x GOO. Teleutospores. Sori hypophyllous, on scarcely perceptible or conspicuous spots, minute, scattered or in groups, roundish or irregular, surrounded or half-covered by the torn epidermis, pulverulent, dark-brown ; spores ellipsoid to oblong, rounded at both ends or attenuated downwards, hardly thickened at apex but with a pale papilla, not at all or faintly constricted, scrobi- culate, ochreous-brown, 35 — 54x18 — 27 /x ; epispore 3 — 4 ll thick ; pedicels hyaline, deciduous, rather long. On Vinca major, V. minor. Not common. Spermogones in April ; uredospores, May — June ; teleutospores, July — October. (Fig. 127.) This is one of the most remarkable species of Puccinia found in Britain. There is considerable difference of opinion about its structure. The bodies referred to in the description given by Plowright (I.e. p. 161) as "aecidia" are of a puzzling nature : they are not described by Sydow or Fischer, but are mistakenly considered by them as identical with the primary uredo-sori. They accompany the spermogones on the under side of the leaves, and are flatly pulvinate subepidermal erumpent sori, surrounded by the erect G. u. 12 L78 puccinia epidermis, and consisting of a floor of crowded erect narrow brownish hyphsB, each of which ahstricts from its apex a small, smooth, round, thick-walled, nearly colourless spore, 8 M>/x diam. Plowright describes and figures these spores as forming basipetal chains, Imt this I could not see. The sori arc numerous, about 1 mm. diam. and dark-brown with a greyish bloom, due apparently bo the overlying spores. Can they lie a parasitic fungus like Darluca Filum, which has itself occasionally been considered as an additional spore-form of the Uredine on which it prej ed ' The plants affected by the mycelium of the spermogones are permeated by it ; they grow taller and more erect, the internodes are longer, the Leaves paler, shorter, and thicker. Plowright considered this mycelium to be perennial, which is very probable. The spermogones have a distinct. but faint odour. The sculpture of the teleutospores is very striking and almost unique. According to Sydow, it consists of a network of warts, but, in the British specimens which I have seen, it would l>e more correctly described as a series of pits arranged more or less in longitudinal lines (about 12 across the spore). Fischer represents the network as much finer. As is frequently the case, these markings can only be seen properly on the dry spore ; they vanish almost completely when it is placed in water, unless it is emptied of its contents. The germ-pore of the lower cell is placed near the insertion of the pedicel, a very unusual position ; it is covered with a pale papilla like the upper one. Distribution : Central and South-western Europe. 50. Puccinia Gentianae Link. JEcidium Gentianae Jacz. Champ. Mont. p. 163 (1892), but \ Credo Gentianae Strauss in Wetter, Annal. ii. 102 (1811). Puccinia Gentianae Link, Spec. ii. 73. W. G. Smith, Card. Chron. xxiv. (1885) p. 372, f. 82. Cooke, (irevillea, xiv. 39. Plowr. Ured. p. 147. Sacc. Syll. vii. 604. Sydow, Monogr. i. 340. Fischer, Ured. Schweiz, p. 164, f. 126, [Spermogones. Honey-coloured, in small clusters. sEcidiospores. iEcidia hypophyllous or on the stems, on circular brown spots, in irregular clusters, cup-shaped, with torn white margin; spores delicately verruculose, orange, 16 — 23 x 14— 17 /x.] Uredospores. Sori ainphigenous, but oftener on the upper surface, scattered or circinate, minute, roundish, at first covered by the epidermis, pale-chestnut; spores globose to ovoid, ON GENTIANA 179 Fig. 128. P. Gentianae. Teleutospore, mesospore, and medospore, on G. acaulis. aculeolate, brownish-yellow, 20 — 30 x 18 — 24 /x, with two, rarely three, germ-pores. Teleutospores. Sori similar and also on the stems, but pulverulent and black-brown ; spores ellipsoid to ovoid, rounded at both ends, not thickened above, but some- times with a low broad papilla, not constricted, smooth, dark- chestnut, 28—38 x 24—30 p ; pedicels hyaline, thin, rather long, very deciduous ; occa- sionally there are a few mesospores intermixed. On Gentiana acaulis. Uredo- and teleutospores, Kew Gardens, August, 1905 (M. C. Cooke), and Horsham (W. G. Smith). (Fig. 128.) I have a specimen on 67. Andrewsii, from the United States, with exactly similar teleutospores. The aecidia are said to appear in April and June, and more frequently on the stems and peduncles than on the leaves. Jaczewski's JEcidium was on calyx, stem and leaf of 67. angustifolia, on pale orange spots. Distribution : Europe, Siberia, East Indies, North America. 51. Puccinia Primulae Duby. JEcidium Primulae DC. Flor. fr. vi. 90. Gooke, Handb. p. 544 ; Micr. Fung. p. 199. Uredo Primulae DC. Flor. fr. vi. 68. Trichobasis Primulae Cooke, Micr. Fung. p. 227. Puccinia Primulae Duby, Bot. Call. ii. 891. Cooke, Handb. p. 495 ; Micr. Fung. p. 204. Plowr. Ured. p. 159. Sacc. Syll. vii. 612. Sydow, Monogr. i. 348. Fischer, Ured. Schweiz, p. 161, f. 124. sEcidiosjjores. iEcidia hypophyllous, on yellowish spots, densely but irregularly clustered in roundish groups, shortly cylindrical, with a broad much cut revolute white margin ; spores verruculose, orange, 17 — 23x12 — 18 fx. 12—2 ISO IMVi'lMA Uredospores. Sori hypophyllous, minute, scattered or circinate, roundish, sood aaked, brown; spores subglobose to ovoid, echinulate, pallid-brown, 20 — 23 x 16 — 19 fi, with three germ -pores. Teleutospores. Sori similar. but long covered by the grey epidermis, often confluent, or standing in circles round the Fig. 129. P. Primulae. . ° . . Teleutospores and mesospore. secidia or uredo-SOri, blackish- brown; spores ovoid or oblong, rounded at both ends, hardly thickened above but with a broad colourless papilla on each germ-pore, gently constricted, smooth, pallid-brown, 22 — 30 x 15 — 18 /x ; pedicels hyaline, short, deciduous ; an occasional mesospore was seen. On Primula acaulis {vulgaris). Xot common. zEcidia in May; teleutospores, June — October. (Fig. 129.) All three spore-forms may be found on the same leaf. The teleuto- spores are rather irregular in shape ; mesospores are not infrequent, and Fischer describes and figures three-celled spores. This parasite is recorded on the continent also on Primula elatior and P. verts. Distribution : Central Europe. 52. Puccinia Soldanellae Fckl. .■Ecidlum Soldanellae Hornsch. in Rab. Krypt. Fl. p. 18 (1844 . Cooke, Handb. p. 537 ; Micr. Fung. p. 195. Credo Soldanellae DC. Flor. fr. vi. 85 (1815). Puccinia Soldanellae Fckl. Symb. Myc. Nacht. iii. 14. Plowr. Ured. p. 159. Sacc. Syll. vii. 618. Sydow, Monogr. i. 349. Fischer, Ured. Schweiz, p. 159, f. 123. Spermogones. Hypophyllous, numerous, punctiform, spheri- cal. JScidiospores. zEcidia hypophyllous, scattered uniformly over nearly the whole leaf-surface, shortly cylindrical or urceo- late, with a white denticulate revolute margin ; spores delicately verruculo.se, yellowish. 18 — 26 fi. [Uredospores. Sori generally epiphyllous, without spots, ON PRIMULACE^ 181 scattered or circinate, minute, surrounded by the torn epidermis, brown ; spores globose to ellipsoid, echinulate, pale-brown, 20—32x18—28/*; epispore thick (2— 3/*), with three germ- pores. Teleutospores. Sori similar, but black-brown; spores ellipsoid to ovate-oblong, with a broad paler apical papilla (5 — 8 \x high), gently constricted, usually rounded below, smooth, brown, 35 — 55 x 20 — 34 ^ ; pedicels hyaline, deciduous, up to 50 /x long.] On Soldanella alpina (Botanic Garden, Glasgow). All three spore-forms are said to occur together on the same plant in July and August. The secidium stage is most common ; its mycelium is perennial and diffused through the plant, and causes a conspicuous change in the leaves ; they become smaller, paler, and longer-stalked. Only this stage seems to have been met with in Britain ; no doubt on imported plants, as has happened with P. Gentianae, P. Pazschhei, etc. Description of the uredo- and teleuto-stages after Sydow. Distribution : Central Europe. 53. Puccinia Hydrocotyles Cooke. Caeoma Hydrocotyles Link, Sp. Plant, ii. 22. Trichobasis Hydrocotyles Cooke, Journ. Bot. ii. 344 ; Handb. p. 530 ; Micr. Fung. p. 225, pi. 8, f. 168—9. Puccinia Hydrocotyles Cooke, Grevillea, ix. 14. Ph. et PL Grevillea, xiii. 53. Plowr. Ured. p. 195. Sacc. Syll. vii. 641. Sydow, Monogr. i. 388. \jEcidiospores. iEcidia amphigenous, distributed pretty uniformly over the whole leaf, rarely single, cup-shaped, with a yellowish deeply cut revolute margin ; spores punctate, pale- yellowish, 19 — 26 /i.] Uredospores. Sori amphigenous, scattered, now and then confluent, often circinate round a central larger one, very minute, long covered by the epidermis, at length naked, pul- verulent, cinnamon; spores subglobose or ellipsoid, echinulate, brownish, 24 — 34 x 20 — 27 /x, with two conspicuous germ- pores. L82 PUfcCJNIA Teleutospores. A Fig. 130. P. Hydrocotyles. Teleutospore and uredospore. very few are found in the uredo-sori, ellipsoid to oblong, rounded at both ends, hardly thickened above, gently constricted, smooth, brown, 30 — 44 x 18 -N/z : pedicels hyaline, I liin. deciduous. On Hydrocotyle vulgaris. Rare : Kew Gardens; Epping Forest; Ireland, Co. Dublin. Uredospores, July — Sep- tember; teleutospores, October. (Fig. 130.) This species is very imperfectly known. The secidiun) is recorded only from South America ; in Europe the uredo-form alone has been observed, except for a few teleutospores in the uredo-sori. Lindroth describes the teleutospores (which are rare everywhere) as furnished with large isolated depressed and rounded warts, while those I have seen are perfectly smooth and with long and persistent pedicels. Cooke describes both uredo- and smooth teleutospores, intermixed, on Hydrocotyle, from Natal (Grevillea I.e.). Sydow states that the uredospores from all localities agree perfectly ; those that I have examined from the Hawaian Islands agree exactly with ours, having the same peculiar colour resembling a strong wash of ''raw sienna." This species illustrates, in the Epping Forest locality, what can be frequently observed : — that, so long as the surroundings are undisturbed by man, many species of Fungi occur year after year continuously in the same spot. It is recorded as found there in 1863, 1864, 1871, 1882, and 1906, etc., and no doubt could have been found, or was found, there equally in all the intervening years. Distribution : France, Holland, Italy, North and South America, Natal, Pacific Islands. 54. Puccinia Saniculae Grev. JEddium Sanicidae Carm. in Cooke's Brit. *Ecid., Journ. Bot. ii. 39, pi. 14, f. 1. Cooke, Handb. p. 543 ; Micr. Fung. p. 198. Puccinia Saniculae Grev. Fl. Edin. p. 431. Cooke, Handb. p. 502 : Micr. Fung. p. 208. Plowr. Ured. p. 160. Sacc. Syll. vii. 618. Sydow, Monogr. i. 413. Fischer, Ured. Schweiz, p. 122, f. 93. Spermogones. Ainphigenous, in little groups. /Ecidiospores. /Ecidia hypophyllous or on the petioles, on ON UMBELLIFER.E 183 brown or purple spots, in small round clusters 2 — 4 mm. diam., elongated on the nerves and petioles, cup-shaped, with a whitish lobed revolute margin ; spores tending to become ellipsoid, delicately verruculose, hyaline, 18 — 26 x 15 — 22 fx. Uredospores. Sori hypophyllous, on pale minute spots 2 — 3 mm. diam., scattered, rarely aggre- gated, minute, punctiform, pale-cinna- mon ; spore globose to ellipsoid, echinu- late, not thickened above, brown, 25 — 38 x 18 — 27 /j, ; epispore thick (up to 3£ fi), with two (rarely three) germ-pores, with slightly swollen caps. Fig. 131. P. Saniculae. m j , o • • -i i i Teleutospore and uredo- 1 eteutospores. Son similar, but dar- Spore. ker ; spores ellipsoid or oblong, rounded at both ends, sometimes slightly thickened above and provided with a small papilla, hardly constricted, smooth, brown, 26 — 45 x 18 — 26 fx; pedicels hyaline, thin, deciduous. On Sanicula europaea. Common. iEcidia, May and June; teleutospores, August — October; but I have found all three kinds of spores, in sheltered places, as early as April. (Fig. 131.) Distribution : Central and North-western Europe. 55. Puccinia Cicutae Lasch. Puccinia Cicutae Lasch in Klotzsch, Herb. Myc. no. 787. Sydow, Monogr. i. 372. uEcidiospores. iEcidia on the nerves of the leaves, on petioles and stems, in roundish or oblong groups as much as 1^-cm. long, pustular, with a feebly developed peridium, golden- yellow ; spores globose to ellipsoid, delicately punctate, sub- hyaline, 17— 26x10— 20 p. Uredospores. Sori generally hypophyllous, scattered, mi- nute, punctiform, pulverulent, cinnamon ; spores subglobose to ovate, echinulate, yellow-brown, 18 — 28 x 14— 22 yu,, with three germ-pores. Teleutospores. Sori similar, but blackish-brown ; spores ellipsoid or oblong, rounded at both ends or rarely attenuated downwards, not thickened above, gently constricted, somewhat L84 prccixiA verruculose or distinctly eerrucosi'-ivticulated, even sometimes nearly smooth, brown. 28 — 4o x 18 — 30 //,; pedicels hyaline, thin, short, deciduous. On Cicuta virosa. Rare. Mentioned by Plowrighl in his arrangement of the British species, Trans. Brit. Myc. Soc. ii. 26; he assigns sponnogones to it as well. I have seen no specimens. The fungus is similar to /'. Smyrnii in its secidia and teleutospores, but differs in the presence (if uredospores. Description after Sydow. Distribution: Central and Northern Europe, Siberia, Japan, North America. 56. Puccinia Apii Desm. Trichobasis Apii Wallr. ; Cooke, Micr. Fung. p. 224. Puccinia Apii Desm. Cat. des PI. omis. p. 25. Cooke, Handb. p. 502; Micr. Fung. p. 208. Plowr. Ured. p. 156. Sydow, Monogr. i. 359. Fischer, Ured. Schweiz, p. 119. Spermogones. Hypophyllous, mostly surrounded by the fficidia, often circinate, shining reddish-brown. sEcidiospores. yEcidia hypophyllous or on the petioles, on minute, irregular, conspicuous, 3^ellowish spots, in roundish or on the petioles elongated clusters, shortly cylindrical, with white torn margin; spores delicately verruculose, orange, 17 — 24 yu.. Fig. 132. P. Apii. Teleutospores and uredospore. Uredospores. Sori hypophyllous, scattered, here and there confluent, surrounded by the epidermis, pulverulent, cinnamon- brown ; spores ellipsoid to obovate or even subclavate, shortly echinulate, slightly thickened above (3 — 5 pu), brownish-yellow, 24 — 35 x 20 — 26 p,, with three equatorial germ-pores. Teleutospores. Sori hypophyllous, rarely epiphyllous, if on ON UMBELLIFER.E 185 the petioles sometimes very large, scattered or confluent, roundish, pulverulent, blackish-brown ; spores ellipsoid to ob- long, rounded above, not thickened, hardly constricted, rounded or gently attenuated below, smooth, brown, 30 — 50 x 15 — 23 /a; pedicels hyaline, thin, deciduous, about as long as the spore. On Apiam graveolens. Not common. iEcidia in May and June ; teleutospores September — November. Distinguished from many of its close allies by the possession of an aecidium. (Fig. 132.) Distribution : Central and Northern Europe, East Indies, Japan, Tasmania. 57. Puccinia iEgopodii Mart. Credo JEgopodii Schum. Plant. Sail. ii. 233. Puccinia JEgopodii Mart. Fl. Mosquen. p. 226. Cooke, Handb. p. 502 ; Micr. Fung. p. 208. Plowr. Ured. p. 201. Sacc. Syll. vii. 678 p.p. Sydow, Monogr. i. 353. Fischer, Ured. Schweiz, p. 105, f. 79. Teleutospores. Sori amphigenous, but chiefly on the petioles and nerves, on thickened yellowish spots, small, but collected into dense irregular clusters and confluent, at first black, covered by the shining epidermis which splits in places longitudinally, soon naked, pulverulent, blackish-brown ; spores ob- long1 to ovoid, often irregularly angled and oblique, usually rounded above and with ' Teleutospores'. a pale wart-like apiculus 2 — 3 u high, hardly or not at all constricted (often broadest at the septum), more or less rounded below, smooth, clear chocolate-brown, granular, 28 — 48 x 15 — 22 /x; pedicels hyaline, short, deciduous. On jEgopodium Podagraria. April — August. Rather com- mon. (Fig. 133.) According to Tranzschel, a few isolated uredospores are to be found in the young sori; they are almost colourless, aculeate, 20 — 22x18/^. Semadeni proved that the spores of this fungus would infect only Mgopodium, and not any of the allied Umbellifers. In this species, 186 PUCCINIA I find tin- septum of the teleutospores almost always comparatively l>road and dark, far more so than in the majority of Puccinias. Distribution : Kun>pe generally. 58. Puccinia Bulbocastani Fckl. Mcidium. Bulbocastani Cumino, Fung. Vail. Pis. lso4 — 5. ./-.'. llinu'i DC. Syii. \>. ")l. Cooke, Eandb. p. 540 p.p. ; Micr. Fung. p. 196. Plowr. Ured. p. 270(?). Puccinia Bulbocastani Fckl. Symb. Myc. p. 52. Cooke, Micr. Fung. p. 209. Sydow, Monogr. i. 363. Fischer, Dred. Schweiz, p. 133, f. 100. Spermogones. Few, scattered amongst the secidia, pale- yellowish. .h'ridiosjjures. .Ecidia rarely on the leaves, hypophyllous, more often on the petioles and stems, densely crowded, caus- ing considerable hypertrophy and curva- ture, between cup-shaped and pustulate, whitish, with a white irregularly torn margin ; spores delicately verruculose, yellowish, 15 — 22 fx. Teleutospores. Sori amphigenous, scattered, minute, roundish, sometimes on the petioles confluent and elongated, long covered by the epidermis, black: spores ellipsoid to obovate-oblong, gener- ally rounded at both ends, not thickened above, hardly constricted, minutely reti- culate, brown, 25 — 42 x 14 — 24/m: pedicels hyaline, thin, deciduous. On Co ft hi (Bh ilium) Bulbocastanum. Very rare. Dunstable (W. G. Smith). (Fig. 134.) This species has no uredospores. Plowright confused together this and the Puccinia tumida on Conopodium denudatum (see his synonymy on pp. 206, 270). The latter species has no secidia ; this partly explains his remarks that he was unable to obtain any evidence of the connection between the secidium and the Puccinia. Nevertheless, it appears not yet Fig. 134. V. Bulbocas- tani. Whole plant of C. Bulbocastanum, vtith aecidia. nat. size. (Dun- stable, April, 1*96.) ON UMBELLIFER.E 187 to have been shown experimentally that the Eecidia and the teleutospores described above are genetically connected. The markings on the teleuto- apore are really little, round, densely crowded pits, not actual reticulations as in P. Chaerophylli. - Plowright's statement {I.e. p. 270) that this secidium was found by him on Conopodium denudatum is a mere slip of the memory, as I am informed by Mr W. G. Smith, in whose company it was found at Leagrave, near Dunstable, on the date mentioned. Distribution : Western and Central Europe, Algeria. 59. Puccinia tumida Grev. Puccinia tumida Grev. Flor. Edin. p. 430 (1824). Sydow, Monogr. i. 376. P. Bunii Winter ; Plowr. Ured. p. 206. P. Umbelliferarum DC. ; Gooke, Micr. Fung. p. 208 p.p., pi. 4, f. 71 — 2. Uredospores. Very few, oval, pale yellow, sparsely verrucu- lose, 20 — 25 x 15 — 18, mingled with the teleutospores. Teleutospores. Sori on the leaves, more often on the petioles and nerves, minute, but many crowded together and confluent in thickened elongated masses (up to 1 cm. long), covered by the ash-coloured epider- mis, for a considerable time, black- brown ; spores ellipsoid to ovate, rounded at both ends, not thickened Fig. 135. P. tumida. Teleu- above, hardly constricted, smooth, 40 p.p. V redo Pititju',,' I!,,. Strauss in WCtter, Annal. ii. 102 (1811). Trichobasis Pimpinellae Cooke, Micr. Fung. p. 224. Pucrinln I'iiiiniaellae Mart. Fl. Mosquen. ed. ii. p. 226. Cooke, Micr. Fung. p. 209. Plowr. Ured. p. 155 p.p. Sacc. SylL vii. 616 p.p. Sydow, Monogr. i. 408. Fischer, Ured. Schweiz, p. 127, f. SJ7. Spermogones. Amphigenous, mostly scattered amongst the aecidia, pale-yellowish. ^Ecidiospores. ^Ecidia hypophyllous, in smaller or larger groups, often along the nerves and causing slight hypertrophy, between cup-shaped and pustulate, with a whitish irregularly cut margin; spores verruculose, subhyaline, 20 — 28 /z. Uredospores. Sori hypophyllous, scattered, minute, pul- verulent, cinnamon; spores globose to ellipsoid, echinulate, brown, 22 — 32 x 20 — 26 /i, with two (rarely three) germ- pores. Teleidospores. Sori similar, but blackish-brown; spores ellipsoid, rounded Fig. 136. P. Pimpinellae. a^ both ends, not thickened above, hardly Teleutospore, on P. Saxi- constricted, reticulate, brown, 28 — 37 x fraga (British); a, the same, on P. magna (after 19 — 2o fi ; pedicels hyaline, deciduous, lecher). rather ghort On Pimpinella magna, P. Saxifraga. Xot common. zEcidia, May and June ; teleutospores, July — October. (Fig. 136.) Very similar to P. Ch.aeroph.ylli ; distinguished from it especially by the uredospores, which have for the most part a thicker and darker membrane with only two germ-pores. The peridium of the acidia is ON UMBELLIFER.E 189 better developed, and the teleutospores are plainly but not so densely reticulate. Klebahn proved by cultures that P. Pimpinellae is distinct from P. Chaerophylli, and Semadeni similarly proved its difference from that species and from P. Heraclei ; the latter showed (Centralbl. f. Bakt. 2. xiii. 215) that it could be transferred to other species of the genus Pimpinella, but not to other genera of the Umbelliferae. Distribution : Europe, Asia Minor, East Indies, Algeria. 61. Puccinia Bupleuri Rud. JEcidium Falcariae /3. Bupleuri-falcati DC. Flor. fr. vi. 91. Puccinia Bupleuri Rud. in Linnsea, iv. 514 (1829). Cooke, Grevillea, vi. 47. Plowr. Ured. p. 154. P. Bupleuri-falcati Wint. Pilze, p. 212 (1884). Sydow, Monogr. i. 364. Fischer, Ured. Schweiz, p. 123, f. 94. [Spermor/ones. Hypophyllous, numerous, generally scattered over the whole surface among the gecidia. zEcidiospores. iEcidia hypophyllous, or a few epiphyllous, uniformly scattered, cup-shaped, with a torn white revolute mar- gin ; spores globose or ellipsoid, punctate, yellow, 16 — 24 /*.] Uredospores. Sori amphi- genous, scattered or occasion- ally circinate, on minute paler spots, small, rounded, cinna- mon ; spores globose to ellipsoid, echinulate, yellow-brown, 19 — 24 x 17 — 22 n, with three, four, or even five germ-pores. Teleutospores. Sori amphigenous, minute, scattered, round- ish, on the stems often larger and oblong, occasionally confluent, covered by the epidermis, at length naked, blackish -brown ; spores oblong to clavate, rounded at both ends, not thickened above, hardly constricted, smooth, brown, 25 — 44 x 16 — 30 yu, ; pedicels hyaline, thin, short, deciduous. On Bupleurum tenuissimum. Very rare. Uredo- and teleutospores, Walton-on-the-Naze, August, 1887. (Fig. 137.) The uredospores were found in small quantities among the teleuto- spores. Fischer says that the affected plants usually bear tecidia on every leaf, the leaves are narrower and paler, and the plants do not flower ; the Fig. 137. P. Bupleuri. Teleuto- spores (Walton-on-the-Naze). |!)() PUCCINIA Becidium-stage appears in Maj and June, often abundantly. This applies especially to the parasite on Bupleurum falcatum, which is probably identical with that on B. tenuissimum, but the Walton plants bore flowers. Distribution: Europe, Asia Minor, East Indies, Yunnan. f>2. Puccinia iEthusae .Mart. Uredo Petroselini DC. Flor. fr. ii. 597. Trichobasis Petroselini Berk. ; Cooke, Micr. Fung. p. 223 (?). T. Cynapii DC. ; Cooke, Micr. Fung. p. 224. Puccinia JEthusae Mart. Flor. Mosq. ed. ii. \>. 22-"> 1817 . Cooke, Micr. Fung. p. 209. p. bullata Wiut. ; Plowr. Ured. p. 185 p.p. /'. Petroselini Lindr, Faun, et Flor. Fenn. xxii., no. 1, p. 84 (1902 . Sydow, Monogr. i. 399, 889. Fischer, Ured. Schweiz, p. 1 1 2. f. 86. Spermogones. Hypophyllous, yellow-brown, or almost hya- line. Credo* pores. Sori generally hypophyllous, scattered or in small clusters, very small, occasionally confluent and larger, pulverulent, cinnamon; spores globose to ellipsoid, echinulate either all over or only in the upper part, thickened above (5 — 6 jjl), yellowish or brownish-yellow, 22 — 29 x 21 — 25 fi, with three (rarely two) equatorial germ-pores with conspicuous caps. Teleutospores. Sori similar, but dark-brown, on the petioles and stems often larger, confluent and elongated ; spores ellipsoid or ovate rounded at both ends or slightly at- tenuated below, not thickened above, hardly constricted, smooth or nearly so, brown. 28 — 48 x 18 — 25 /x : pedicels hyaline, thin, short, deciduous. On JEthusa Cynapium, Petroselinum sativum. Not common. June — Oc- tober. (Fig. 138.) It is possible that the forms on these two hosts are distinct species, or at least biological races. Semadeni showed (Centralbl. f. Bakter. 2. xiii. 443) that, while he could infect several (non-British) species of Umbelliferae with uredospores from dSthusa Cynapium, he could not infect Petroselinum sativum ; at the same time he could find no morphological difference Fig. 138. P. Mthusae. Teleutospore and uredospore on AZthusa. ON UMBELLIFERjE 191 between the two forms. Conium maculatum took the infection very weakly. Lindroth and Fischer both describe the teleutopores as furnished with numerous minute embedded granules, otherwise even or with low rounded undulations, but Fischer figures them as perfectly smooth, as they certainly are in the cases I have seen. The uredospores are spiny in the upper part, nearly smooth below. When the few spines on the basal part are not to be discerned (as sometimes happens), they closely resemble those of P. Conii, except in being relatively broader ; these two species are closely allied. Distribution : Central and Northern Europe. 63. Puccinia Silai Fckl. Puccinia bullata Wint. Pilze, p. 191 p.p. Plowr. Ured. p. 183 p.p. Sydow, Monogr. i. 403 p.p. Fischer, Ured. Schweiz, p. 119, f. 91 b. P. Silai Fckl. Syrnb. Myc. p. 53. Cooke, Grevillea, xiv. 39. Spermogones. Scattered, pale yellowish, accompanying the primary uredo-sori. Uredospores. Primary sori generally on the nerves and petioles, elongated and con- fluent up to 3 cm. long, dark cinnamon ; secondary hypo- phyllous or occasionally epi- phyllous, scattered, minute, punctiform, brown; spores glo- bose to ovate, more or less thickened above (4— 5/*), echin- Fig 139 P_ silai. Teleutospores. ulate, brown, 25 — 40 x 18 — 28 /x, with three (rarely four) germ-pores. Teleutospores. Sori minute, similar, but sometimes con- fluent on the stems, blackish ; spores obovate or oblong, rounded at both ends or gently attenuated below, not thick- ened above, but often with a papilla, slightly constricted, smooth, brown, 28 — 42x18 — 32 /j,; pedicels hyaline, rather short, deciduous. On Silaus pratensis. Rare ; Pontrilas ; Kew Gardens. August, September. (Fig. 139.) L92 PUCCINI A This species was included by Sydow under the old P. bullata of Winter ; bul since then it has been rendered probable by the experiments of Seiiiadeiii that it is distinct from most of the forms still remaining under thai collective name (Centralbl. f. Bakter. 190 1, 2. xiii. 530). Schrbter proved that the spore-forms described above are genetically connected. The uredo- and teleutospores are generally found intermixed. 64. Puccinia Angelicae Fckl. Vmln Ai>idia hypophyllous, frequently on the [m tidies and especially on the nerves of the leaves, on thickened yellowish spots, densely crowded in irregular clusters, often causing distortion, between cup-shaped and pustulate, sometimes almost spherical and superficial; peridium feebly developed, opening by a rounded pore ; spores delicately verruculose, yellowish, 21—32x18—28^. clei. " Keidia on U redospures. Sori amphigenous, scattered, Beracleum (re- minute, chestnut-brown ; spores globose to ellipsoid, densely echinulate, pale-brown, 25 — 32 x 19 — 27 fji, with three or four germ-pores. Teleutospores. Sori similar or more or less confluent on the nerves, surrounded by the ferruginous epidermis, pulverulent, blackish ; spores ellipsoid, rounded at both ends, hardly con- stricted, reticulated, brown, 26 — 37 x 18 — 27 /j. ; pedicels hya- line, short, deciduous. On Heracleum Sphondylium. zEcidia, March — June; teleu- tospores, August. Not common. (Fig. 142.) This species closely resembles P. CkaerophyUi, but is distinguished by its less densely reticulated teleutospores. Seruadeni proved by experi- ment that they are distinct species, but no one has as yet reared all the spore-forms of P. Heraclei from the basidiospores, as has been done for P. Chaerophylli. The aecidia of this species are more conspicuous than those of its allies ; they occur in swollen patches, reminding one of the eecidia <>n .Siiu/rniun, <)1 uxntrum, and being sometimes almost spherical and superficial might be compared to a group of miniature Peziza vesiculosa. Distribution : Central and North-western Europe. ON UMBELLIFER/E 195 67. Puccinia Chaerophylli Purt. Puccinia Chaerophylli Purton, Midi. Flor. iii. 303. i. 367. Fischer, Ured. Schweiz, p. 129, f. 98. P. Pimpinellae Str. ; Plowr. Ured. p. 155 p.p. Sydow, Monogr. Spermogones. Pale-yellow, roundish. JZcidiospores. /Ecidia on the leaves and petioles, on the leaves scattered or circinate, on the petioles and nerves in dense elongated clusters and causing a slight hypertrophy, between cup-shaped and pustulate, yellowish ; peridium poorly developed ; spores verruculose, orange, 18 — 35 x 16 — 26 p. Uredospores. Sori hypophyllous, scattered, minute, roundish, pulverulent, cinnamon ; spores globose to obovate, echinulate, pale brownish-yellow, 20 — 30 x 18 — 25 p ; with three usually equatorial germ-pores. Teleutospores. Sori similar, but black-brown, on the petioles more elongated ; spores ovate to oblong, rounded at both ends or gently at- tenuated below, not thickened above, slightly constricted, reticulated, yellow- ish-brown or brown, 24 — 36 x 16 — 25 p\ pedicels hyaline, thin, as long as the spore or shorter. On leaves, petioles, and stems of Anthriscus silvestris, Chaerophyllum temulwm, Myrrhis odorata. Not com- mon. ^Ecidia, May and June ; teleuto- spores, July — October. (Fig. 143.) Fig. 143. P Chaerophylli. Teleutospores, a, on Anthriscus, b, on Myr- rhis. It was proved experimentally by Klebahn that this parasite is not identical with P. Pimpinellae, and by Semadeni that it is not identical with that or with P. Heraclei. The latter also reared spermogones and secidia from the basidiospores, and uredo- and teleutospores from the Becidiospores, of P. Chaerophylli; thus proving that all the spore-forms described above belong to the same species. The feeble development of the peridium and the pustule-like, not cup-shaped, aecidia are paralleled by those of P. Heraclei. The markings on the teleutospore are formed by a network of low ridges, with small polygonal or roundish meshes. The spore-reticulation is exactly of the same character as that of P. Pimpinellae except that the meshes are a little smaller and not quite so easily seen. 13—2 1 96 PUCCINI. \ Semadeni showed that the spores from .1. aUvestris infected M. inhtrnta readily (Centralbl. f. I'.akt. pt. 2, xiii. 217—9), but whether the form on C. /'nullum belongs to the same species (or is a biological race of it) seems at present to he undetermined. Distribution: Central and Northern Europe, Siberia. 68. Puccinia Conii Fckl. Uredo Conii Strauss in Wetter. Ann. ii. 96. Trichobasis Conii Cooke, Micr. Fung. p. 225. T. Umbellatarum Lev. ; Cooke, Micr. Fung. p. 22~> p.p. Puccinia Conii Fckl. Symb. Myc. p. 53. Cooke, Handb. p. 209 p.p. Sacc. Syll. xiv. 30-2. Sydow, Monogr. i. 375. Fischer, I'red. Schweiz, p. 114, f. 87. J', bulla ria Link ; Cooke, Handb. p. 503 p.p. P. bullata Wint. ; Plowr. Ured. p. 183 p.p. Uredospores. Sori hypophyllous, occasionally on the peti< >les, scattered, minute, rarely confluent, pul- verulent, cinnamon ; spores ellipsoid to obovate, thickened (up to 7 /a) above and echinulate in the upper part only, pale-brownish, 24—36 x 17 — 26 /z, with three germ-pores. Teleutospores. Sori similar, but blackish-brown, on the stems and petioles often larger and long covered by the grey epidermis ; spores ovate or ovate- oblong or even clavate, rounded at both ends or attenuated below, not thickened at the apex but with a small papilla or pore-cap, hardly constricted, nearly or quite smooth, pale-brown, 30 — 48 x 20 — 28 /x; pedicels hyaline, short, deciduous. On Conium maculatum. Not common ; England, Wales, Ireland (Clare Island). August, September. (Fig. 144.) Distinguished by the uredospores, which are echinulate only in the upper half ; the spines gradually diminish in size downwards and the lower half is quite smooth. The teleutospores which I have examined are quite smooth when empty, even under the highest power, but the protoplasm is very granular and presents a misleading effect at first sight. Fig. 114. P. Conii. Te- leutospore and uredospore. ON UMBELLIFER.E 197 69. Puccinia Smyrnii Corda. JEcidium Bunii var. Smyrnii- Olusatri DC. Flor. fr. vi. 96. Trichobasis Petroselini Berk. ; Cooke, Handb. p. 529 p.p. Puccinia Smyrnii Corda, Icon. iv. 18, f. 67 (1840). Cooke, Handb. p. 503 ; Micr. Fung. p. 209, pi. 3, f. 55—6. Plowr. Ured. p. 199. Sacc. Syll. vii. 670. P. Smymii-Olusatri Lindr. Faun, et Flor. fenn. xxii. no. 1, p. 9 (1902). Sydow, Monogr. i. 416. Spermogones. Epiphyllous, on sunken spots. sEcidiospores. iEcidia hypophyllous or occasionally epi- phyllous, in rather large irregular clusters, or on the petioles and stems in elongated groups, on yellow spots, hemispherical, yellow, opening by an irregular pore with a nearly entire margin; spores globose or ovate to fusiform or pyriform, delicately verruculose, yellowish, 16 — 40 x 16 — 20 p. Teleutospores. Sori hypophyllous, on small yellow spots, scat- tered or a few together, minute, pulverulent, dark-brown; spores ellipsoid to oblong, rounded at both ends, not thickened above, hardly constricted, coarsely and remotely reticulate and tuber - culate, brown, 30 — 48 x 17 — 26 fi ; pedicels hyaline, thin, de- ciduous, up to 60 jx long; epi- spore rather thick. Fig_ U5 p SmyrniL Teleuto. . _. spores; a, as really sculptured; On Smyrnium UiUSatrum. b, as seen when wet. Rather common near the coast. iEcidia, April — June; teleutospores, June — August. (Fig. 145.) The secidia and teleuto-sori may occur on separate plants or on the same. The markings on the teleutospore form a wide-meshed network, which bears wart-like tubercles at the angles of the meshes. The aecidio- spores bear more resemblance in form to uredospores than to what they really are ; but they are produced in chains with intercalary cells in the usual way : they are not echinulate, but delicately verruculose ; the markings can easily be seen on an empty spore. The peridium-cells are grossly verrucose on the inner surface, and are not arranged in regular rows. It is this irregularity which causes the peridium (as is usual in such cases) not to split into lacunae, but to open by a pore. IDS PUCCINIA Distribution: Central and Eastern Europe, Crete, Cyprus, Asia M nmr Algeria. 70. Puccinia Circaeae Pe rs. Puccinia Circaeae Pers. Disp. Meth. ]>. 39, pi. 3, f. 4. Cooke, Handb. p. 507; Micr. Fung. p. 211. Plowr. Ured. p. 213. Sacc. Syll. vii. 686. Sydow, Monogr. i. 4-2-2. Fischer, Ured. Sdiwfiz, p. 319, f. 232. Teleutospore*. Sori hypophyllous, on sunken yellowish or purplish round spots, minute, pulvi- nate, brown, then with a greyish bloom, scattered or circinate and at length confluent in a thick crust ; spores generally fusoid, rounded or conically attenuated above and much thickened (up to 1 2 /a), gently constric- ted, attenuated downwards, smooth, yellowish-brown or brown, 25 — 40 x 9 — 13/a; pedicels hyaline, persistent, Fig. 14(). P. Circaeae. Teleutospores. about as long as the spore. On Circaea alpina, 0. lutetiana. Rather common. August — October. (Fig. 146.) The sori of this species present two different forms : the first-formed are roundish, clear-brown, solitary or circinate and confluent ; the later- formed, which appear round the others or on the stem and on the nerves of the leaves, are darker-brown and never greyish. All the spores are of the same shape, but the paler ones can germinate at once, in the sorus, while the darker ones rest until the following spring, as in P. Veronicarum. There is another Uredine on the same host, even occurring on the same leaf : see Pucciniastmm, Circaeae. Distribution : Central and Northern Europe, North America, East Indies. 71. Puccinia pulverulenta Grev. jEcidium Epilobii DC. Flor. fr. ii. 238. Cooke, Handb. p. 536 ; Mier. Fung. p. 195. Uredo vagans var. Epilobii-tetragoni DC. Flor. fr. ii. 228. Trichobasis Epilobii Berk. ; Cooke, Micr. Fung. p. 226. ON EPILOBIUM 199 Puccinia pulverulenta Grew Fl. Edin. p. 432 (1824). Cooke, Handb. p. 507 ; Micr. Fung. p. 211, pi. 4, f. 78—9. Plowr. Ured. p. 151. P. Epilobii DC. ; Sacc. Syll. vii. 608 p.p. P. Epilobii-tetragoni Wint. Pilze, p. 214 (1884). Sydow, Monogr. i. 424. Fischer, Ured. Schweiz, p. 152, f. 118. McAlpine, Rusts of Australia, p. 170, f. 79—81. Spermogo7ies. Scattered among the gecidia, honey-coloured. sEcidiospores. iEcidia hypophyllous or, when very abundant, also epiphyllous, scattered rather closely over nearly the whole surface of the leaf, cup-shaped, with a white torn revolute margin; spores very delicately verruculose, orange, 16 — -2(j p. Uredospores. Sori hypophyllous, scattered or circinate, sometimes confluent, pulverulent, chestnut-brown ; spores globose to ovoid, remotely echinulate, brown, 20 — 28 x 15 — 25 fi, with two germ- pores. Teleutospores. Sori hypophyl- lous, often circinate, soon naked, Fig. 147. P. pulverulenta. Te- pulverulent, dark-brown ; spores leutospores and uredospore. ellipsoid or ovoid, rounded at both ends, somewhat thickened above (up to 5 fi) with a broad low cap-like addition, gently constricted, smooth, brownish, 24 — 35 x 14 — 20 fx ; pedicels hyaline, slender, deciduous. On Epilobium hirsutum, E. montanum, E. tetragonum. zEcidia, May and June ; teleutospores, June — November. Common. (Fig. 147.) The eecidium-forming mycelium appears (but perhaps falsely) to be perennial, for the same plants are attacked year after year. The aecidia appear in May and cover leaf after leaf, as they are developed. The affected plants are easily recognisable by their much paler and yellowish colour. Soon the sori of uredo- and teleutospores begin to appear, at first on the same leaves, but afterwards on the later-formed leaves higher up the plant. In September and October the small last-formed leaves are thickly covered by the teleutospores ; it is probably from the germination of these in spring that the next attack proceeds. The mycelium of the uredo- and teleuto-sori is strictly localised. Plowright states (I.e.) that the ajcidiospores sown on young seedlings of E. hirsutum gave rise to aecidiospores in seventeen days, but very possibly 200 PU<< IMA there Lb some oversighl here, for Dietel obtained the wedo by sowing the secidiospores from K. tetragonum on that plant ; he also suggests that the form on K. tetragonum is biologically distinct from that on E. hirsutum, since on the latter he obtained no result. DISTRIBUTION: Central and Northern Europe, Nov. i Z.mbla, Siberia, North America. 72. Puccinia Epilobii DC. /''/ccinia Epilobii DC. Flor. fr. vi. 61. Cooke, Handb. p. 506; Micr. Fung. p. 211. Plowr. Ured. p. 202. Sydow, Monogr. i. 127. Fischer, Ured. Sehweiz, p. 155, f. 120. Teleutospores. Sori hypophyllous, scattered or rather crowded, often uniformly distributed over the whole surface of the leaf, rarely confluent, sur- rounded by the torn epidermis, pulverulent, reddish-brown ; spores ellipsoid, oblong or pyriform, rounded at both ends, hardly thickened above, but with a minute papilla, much constricted, minutely verruculose, Fig. 148. P. Epilobii. ' J Teleutospores. clear brown, 2t — 48x16 — 25 fx : pedicels hyaline, 10 — 16 /x long. On Epilobiii in palustre. Rare. May — August. A sub- alpine and arctic species. (Fig. 148.) Distinguished from P. pulrrrvJenta not only by its spores and the absence of the secidium- and uredo-stages, but also by the smaller teleuto- sori which are scattered pretty uniformly over the leaf-surface. The mycelium seems to be perennial ; it permeates the whole plant, and somewhat deforms the shoots, making the leaves smaller and thicker. The warts of the epispore are sometimes hardly perceptible, but can usually be seen if the spore is squeezed in the way recommended, see p. 83. Distribution : Central and Northern Europe. 73. Puccinia Violae DC. JEcidium Violae Sebum. PI. Sail. ii. 224. Cooke, Handb. p. 543 ; Micr. Fung. p. 198. Tri'Jiobasis Violarum Lev. ; Cooke, Micr. Fung. p. 226. ON VIOLA 201 Puccinia Violae DC. Flor. fr. vi. 62 (1815). Plowr. Ured. p. 152. Sydow, Monogr. i. 439. Sacc. Syll. vii. 609. Fischer, Ured. Schweiz, p. 139, f. 106. P. Violarum Link, Sp. Plant, ii. 80 (1824). Cooke, Handb. p. 504 ; Micr. Fung. p. 210 ; Grevillea, iii. pi. 49, f. 5, 10 a, b. Spermogones. Crowded in little clusters, yellowish. JZcidiospores. iEcidia on all the green parts of the host, on the leaves often forming swollen yellowish spots, generally in roundish, or irregularly expanded, groups, on the stem some- times scattered, flat, with a white irregularly torn revolute margin ; spores delicately verruculose, orange, 16 — 24 x 10 — 18 yli. Fig. 149. P. Violae. a, leaf of V. silvatica with recidia ; b, teleutospore, seen wet ; c, the same, seen dry ; d, a mesospore. Uredospores. Sori hypophyllous, scattered or circinate, minute, soon naked, pulverulent, cinnamon-brown ; spores glo- bose to ellipsoid, echinulate, brownish, 20—26 x 17 — 23 fi, with two germ-pores. Teleutospores. Sori similar, but darker ; spores ellipsoid to oblong, rounded at both ends or gently attenuated below, thickened and paler above, hardly constricted, faintly punctate, chestnut-brown, 20 — 40x15 — 23/a; pedicels hyaline, deciduous, rather long ; a few mesospores are found. On all green parts of Viola canina, V. hirta, V. odorata, V. Riviniana, V. silvestris. Very common. ^Ecidia, April- June ; teleutospores, August — November. (Fig. 149.) The punctation of the teleutospores is very delicate, like little pin- pricks, and can only be seen when they are dry ; these spores are generally described as smooth and appear so except in the most favourable circumstances. The germ-pore of each of the cells is covered with a paler convex cap. The connection of all the spore-forms of this species was 202 I'l ( < INI \ experimentally proved by De Bary andJacky. Bock (Centralbl. fiir Bakt. •_'. x.\. 586) showed that what appeared to be identical with /'. Violae could also !"• artificially produced on I', comma, V. Intra, I', tricolor; the secidia formed small or Large crowded well-defined groups, and thereby differed from those of /'. aegra which spread widely over the stems and leaves of their hosts, mostly standing alone and not collected in groups. In the latter, moreover, secidia and teleutospores can be found in July at the same time (or even on the same leaf), which is not usually the case with /'. Violae. This species can attack cultivated violets, but rarely does much harm ; burning the infected plants is a sufficient remedy. Distribution : World-wide, except Australia. 74. Puccinia aegra Grove. .Eciih'nm (l<'i>au{jcra>ix\T\Y.e, Gard. Ghron. Is7<'> pp. Cooke, Micr. Fung. p. 195 ; Grevillea, v. 57. Puccinia argra Grove, Journ. Bot. xxi. (1883) p. 274. p. 158. Sacc. Syll. vii. 614. P. depauperans Sydow, Monogr. i. 442 (1903). 175, 361, 437. Plowr. Ured. Uredospores. JEcidiospores. ^Ecidia on all green parts of the host, particularly on the stems, not in clusters, but spread pretty uniformly over the whole surface, especially of the stems which are swollen and distorted, cup-shaped, with a white torn revolute margin; spores globose to ellipsoid, smooth, orange, 17 — 21 x 14— 16 fx. Sori amphigenous, without spots, irregularly scattered or aggregated, occasionally confluent, long covered by the lead- coloured epidermis, cinnamon ; spores globose to ellipsoid, delicately echinu- late, brownish 20 — 28 /x diam. Teleutospores. Sori similar, but darker; spores ellipsoid or ovoid, rounded or gently attenuated above, where it is slightly thickened or rather surmounted by a minute subhyaline papilla, hardly constricted, usually rounded below, very delicately punctate or quite smooth, brown, 22 — 34 x 16 — 20 ix ; pedicels hyaline, short ; a few mesospores are found. Fig. 150. P. argra. Teleuto- spores and mesospore. ON VIOLA 203 On Viola cornuta, V. lutea, V. tricolor, and most of the many hybrid Violas and Pansies now cultivated in gardens. Fresh ascidia are formed right through the summer till August, while those of P. Violae cannot usually be found after early June. (Fig. 150.) The mycelium of the ajcidia is perennial in the underground parts ; all the shoots which arise from the affected plant are deformed, the internodes are lengthened, the leaves become smaller, paler and often twisted. In P. Violae this is not the case ; only a slight swelling arises at the part where the localised mycelium is producing its eecidia. Bock (Centralbl. fiir Bakt. 2. xx. 586) found that he could produce the pecidia of P. Violae on the three species of Viola named above, by artificial infection ; but like others he still considered the two species as distinct on account of their different habit. Liro, on the contrary, considers them as the same. In my experience, the uredo- and teleuto-sori of P. aegra are larger, and remain longer covered by the epidermis, but the spores are identical. This species can do considerable harm if allowed to spread ; there is no remedy, but all infected plants should be carefully uprooted and burnt. A Viola, badly attacked by the recidium, was once sent to the Gardener's Chronicle by a correspondent as " a hybrid between a fern and a violet." Distribution : Germany, Denmark. 75. Puccinia Fergussoni B. et Br. Puccinia Fergussoni Berk, et Br. Ann. Nat. Hist. 1875, p. 35. Cooke, Grevillea, iii. 179, pi. 49, f. 10 c ; Micr. Fung. p. 210. Plowr. Ured. p. 207. Sacc. Syll. vii. 682. Sydow, Monogr. i. 444. Grove, Journ. Bot. 1912, p. 10. Teleutospores. Sori hypophyllous or on the petioles, on large roundish or irregular yellow spots, in suborbicular or (on the petioles) elon- gated clusters up to 1| cm. long, densely crowded and confluent, long covered by the epidermis, then pulverulent, choco- late-brown; spores irregular, generally oblong, attenuated or rarely rounded at both ends, thickened above in a conical form (up to 6 ll), gently constricted, smooth, pale-brown, 26 — 45 x 12 — 18 //.; pedicels hyaline, thin, deciduous, up to 30 fx long ; an occasional mesospore is found. Fig. 151. P. Fergussoni. Teleutospores and me- sospore. 204 PUCCINIA On Viola palustris. Bare; Wales, Scotland, and near Birmingham. The beginnings of the sori may be seen by the middle of May. (Fig. 151.) This species is easily recognised by its large and pulvinate groups of sori. The mycelium spreads considerably beyond the part occupied by th^ spores, and consequently causes large yellow patches, usually only one or at most two on each leaf, each the result of a separate infection by the basidiospores. /'. asarina Cooke, Handbook, p. 504, Plowright, Uredinese, p. 202 (non K. et S.), is this species, a mistake having been made in identifying the • i ilant. In continental specimens of P. asarina Kunze, so far as I have seen, the sori on the lamina are as frequent on the upper leaf-surface as on the lower, whereas in /'. Fergussoni they are entirely hypophyllous. Distribution : Northern Europe, North America (?). 7(>. Puccinia argentata Wint. JEcidivm argentatum Schultz, Prod. Flor. Starg. p. 454 (1806 : teleuto- spores, on Impatiens). Trichnbad* Impatientis Rab. ; Cooke, Micr. Fung. p. 225. Puccinia Noli-tangeris Corda, Icon. iv. 16, pi. 5, f. 57 (1840). Cooke, Handb. p. 504 ; Micr. Fung. p. 210. P. argentata Winter, Pilze. p. 194 (1884). Plowr. Fred. p. 193. Sacc. Syll. vii. 639. Sydow, Monogr. i. 450. Fischer, Ured. Sehweiz, p. 143, f. 109. [Spermogones. Hypophyllous, scattered among the secidia, honey-coloured. JEcidiospores. iEcidia hypophyllous, pretty uniformly dis- tributed on discoloured swollen spots, on the petioles and stems more scattered, white, with a deeply- cut revolute margin ; spores 18 — 22 x 13 — 20 ft ; contents golden-yellow.] Lredospores. Sori hypo- phyllous, scattered or circinate, rig. 152. P. argentata. leleutospores, . . . from the original specimen (L. Jenyns). sometimes on minute yellowish spots, often confluent, covered by the silvery epidermis, then pulverulent, roundish, ochra- ceous; spores globose to broadly ellipsoid, delicately echinulate, ON IMPATIENS 205 pale-yellowish, 16 — 22x14— 20 fi, with 3—5 (usually four) germ-pores. Teleutospores. Sori similar, but chestnut-brown; spores ellipsoid to subclavate, with a colourless conical cap to each germ-pore, rounded or slightly attenuated at both ends, hardly constricted, smooth, pale-brownish, 25—38 x 12 — 22 /z; pedicels hyaline, slender, short. [zEcidia on Adoxa Moschatellina, April — June;] uredo and teleutospores on Impatiens fulva, I. Noli-tangere, May, August- October. Very rare; Albury, Surrey, October, 1864 (Rev. L. Jenyns), Guildford (Rev. W. A. Vize), Shere (M. C. Cooke), Kew Gardens (G. Massee). (Fig. 152.) The teleutospores are at first produced in the same sori as the uredo- spores ; the eecidial stage is not recorded, probably because it has been confounded with the secidium of P. albescens, from which it is distinguished, according to Bubak, by its gold-coloured spores. I do not think this distinction will hold good ; many of the specimens found in this country, which appear to be P. albescens, have golden-yellow spores. See under that species (p. 163). The description given above of the spermogones and secidia is taken from Bubak, who showed (Centralbl. fiir Bakt. 2. xii. 413) that they could produce the other stages on Impatiens in about ten days. He also showed that the secidium could be produced on the Adoxa by over-wintered teleutospores ; the incubation period was as long as one month, probably because the mycelium first permeated the whole plant, from the leaf to the stem, before producing spores. Afterwards he proved, contrary to his former opinion, that the mycelium does not perennate in the rhizome, but fresh infection must take place each spring {ibid. xvi. 150). Distribution : Central and Northern Europe, North America, Japan. 77. Puccinia Buxi DC. Puccinia Buxi DC. Flor. ft', vi. 60. Cooke, Handb. p. 508 ; Micr. Fung. p. 212. Plowr. Ured. p. 217. Sacc. Syll. vii. 688. Sydow, Monogr. i. 453. Fischer, Ured. Schweiz, p. 316, f. 228—30. Teleutospores. Sori amphigenous, on indefinite spots, scat- tered or confluent, hemispherical, pulvinate, hard, compact, soon naked, dark chestnut-brown or purplish-brown ; spores 206 PUCCINI A oblong i" clavate, rounded above and not thickened, evidently constricted, usually attenuated below, smooth, brown, 55 — 90x20 — 35 yu. (or even 1 00 yu, long); pedin -Is hyaline, per- sistent, very long, (reaching 160 /*). On Buxu8 8emperviren8. Rather common. September, October, lasting through the winter and following spring. (Fig. 153.) The spores of this species easily fall apart into their component cells. Ed. Fischer proved that it has only the one spore-form : he gives (I.e.) figures showing the effect upon the leaf of an infection by the basidiospores. Accord- ing to him, the teleutospores germinate in spring, and infect the delicate young leaves. The mycelium grows slowly. During the summer and autumn the infected spot becomes much thickened : the sori are produced in late autumn or during the following winter. This is exactly in accordance with the suggestion made by Plowright (I.e.), without any experi- mental evidence being; at that time available. Fig. 153. P. Bu.ri. Teleutospores. I Distribution : Europe and Persia. < 8. Puccinia Malvacearum Mont. Puecinia Malvacearum Mont, in Gay, Hist. Chile, viii. 43. Cooke, Micr. Fung. p. 20-"> ; Grevillea, ii. 47, 137 and iii. pi. 35, pi. 49, f. 1. Plowr. Ured. p. 212. Sacc. Syll. vii. 686. Sydow, Monogr. i. 476. Fischer, Ured. Schweiz, p. 313, f. 227. McAlpifle, Rusts of Australia, p. 178, f. 99, 100, 123—130, and pi. F, f. 28. Teleutospores. Sori hypophyllous or amphigenous, and on the petioles and stems, on conspicuous yellow or orange spots, scattered but close together, small, hemispherical or on the stems elongated, pulvinate, compact, hard, at first pale-reddish, then reddish-brown ; spores oblong to subfusoid, attenuated at both ends or rarely rounded above, thickened at the apex, gently constricted, smooth, yellowish-brown, 35 — 75 x 12 — 26 //.; pedicels hyaline, persistent, short or as much as 150 /x long; <>ne-, three-, or even four-celled spores also occur. ON MALVACEAE 207 Fig. 154. P. Malvacearum. Teleutospores and meso- spore. On many species of Malvaceae (of the subfamily Malveae), especially on Malva moschata, M. silves- tris and Althaea rosea. Very common. May — October (also in April and Novem- ber). (Fig. 154.) This is one of the most noticeable of the Uredinales. It is truly plurivorous ; so far from being confined to a species, it is not even confined to a genus. In botanic gardens, where species of the family Malvaceae are grown, side by side, in the same plot, the disease can be seen to spread to plants of all the allied genera — Malva, Lavatera, Althaea, Kitaibelia, Malope, Abutilon, Sida, Sidalcea, Anoda, Malvastrum, etc., have been recorded. A list containing many (nearly forty) species of these genera is given by Sydow, to which more are added by Fischer, Mc Alpine, and Dandeno. On all these it appears to be identical ; artificial infections have proved that it can be transferred from Malva to Althaea, and vice-versd. It was first made known in 1852 by Montague from a specimen found in Chili. It was observed in Australia in 1857 (McAlpine). In Europe it appeared in 1869, in South Africa in 1875, and it is now spread all over the world. It is believed that Chili was its native home ; the rapidity of its distribution to other countries has few or no parallels among plant diseases. It has been proved by many experimenters that it produces only the one kind of spore, which is capable of germinating at once when mature, though some can hibernate. It is a disputed point whether the mycelium can pass the winter in the plant or in the seed : the balance of evidence goes to show that fresh infections arise each year by the germination of over-wintered teleutospores, which can be found on all green parts, even on the fruits. See p. 48. I 79. Puccinia Pruni-spinosae Pers. jEcidium punctatvan Pers. in Uster. Annal. Botanik, xx. 135. Plowr. Ured. p. 268. JE. quadrifidum DC. Flor. fr. vi. 90. Cooke, Handb. p. 536 ; Micr. Fung. p. 194. Puccinia Pruni-spinosae Pers. Syn. p. 226 (1801). Sydow, Monogr. i. 484. Fischer, Ured. Schweiz, pp. 157, 547, ff. 21, 122, 340. McAlpine, Rusts of Australia, p. 171, f. 83—6 and pi. D, f. 19—20. 208 PUCCINIA /'. Pruni DC. Flor. IV. ii. 222 (1805). Plowr. Ured. p. 192. Sacc. Syll. vii. 648. /'. Prunorum Link, Sp. PL ii. 82 (1825). Cooke, Handb. p. 507; Micr. Fun-. ]». 211 : Grevillea, iii. pi. 49, f. 11. Tranzschelia punctata Arthur, North Americ. Fl. vii. 151. Spermogones. Amphigenous, scattered, brown or blackish, very shallow, punctiform. /Ecidiospores. /Ecidia hypophyllous, scattered over the whole surface, flat, with a broad revolute margin which is torn into few (3 — 5) lobes; spores roundish, pale yellow- ish-brown, finely verruculose, 16—24 fi. Uredospores. Sori hypo- phyllous, generally on minute coloured spots, scattered, but often crowded and confluent, soon naked, pulverulent, cin- namon-brown ; spores ellip- soid to fusiform, ovoid-oblong or sub-pyriform, smooth and more or less thickened at the summit in a conical shape and darker, paler and narrowed below, where they are sharply Fig. 155. P. Pruni-spinosae. iEcidia on A. coronaria (slightly reduced) ; a, an secidium on A. Aemorosa, York- shire, the normal form, and b, two less usual forms, x 30. Fig. 156. P. Pruni-spinosae. Two teleutospores ; a, lower half of a teleuto- spore ; b, uredospore ; c, paraphysis ; d, a cluster of teleutospores. On Wild Plum. (All x 600, except d, which is x 360.) verrucose or echinulate, pale-brown, 20 — 35 x 10 — 18 /x (with three or four equatorial germ-pores, Arthur), mixed with yellowish-brown or pale capitate paraphyses more or less thickened at the apex. ON PRUNUS 209 Teleutospores. Sori similar but blackish-brown ; spores ellipsoid to oblong, composed of two globose cells which readily separate (or lower cell often narrower, paler and imperfect), not thickened above, densely and coarsely verrucose, brown, 30 — 45 x 18 — 25 fi ; pedicels hyaline, very short, deciduous, spring- ing in clusters of about 10 — 20 from a common base. iEcidia on Anemone coronaria, A. nemorosa, April and May; uredo- and teleutospores on Primus domestica, P. insititia, P. spinosa, also on cultivated species and varieties of Primus and its allies, August — October. Common in certain districts. (Figs. 155, 156.) The discovery of the hetercecism of this parasite is comparatively recent. Tranzschel first showed (in 1904) that it was hetercecious, using Anemone coronaria and Amygdalus communis as the alternate hosts. F. T. Brooks, at Cambridge in 1911, laid fresh secidiospores from the same species of Anemone on both sides of certain leaves of a "Victoria" Plum, leaving others uninoculated. The plant was enclosed by a bell-jar, and three weeks later twenty-three of the inoculated leaves were found to bear on their under-surface uredo-sori of P. Pruni-spinosae, while the control leaves showed no signs of the rust (New Phytologist, x. 207). Arthur, in the United States, proved a similar fact, but in that case the host of the alternate phase was Hepatica acutiloba, a very close ally of Anemone. The a;cidium is also reported on other species of Hepatica, Anemone ranunculoides and other species of Anemone, Eranthis hiemalis, and various species of Thalictrum. Scribner (Report of the Dept. Agric. U.S.A. 1887) describes the Puccinia as found on Cherry, Apricot, and Peach : it is recorded by McAlpine on leaves, fruit and stems of Peach and Nectarine, and leaves and fruit of Almond and Apricot. Trichobasis llhamni of Cooke (Seem. Journ. Bot. ii. 344, iv. 104) on " Ithamnus catharttcus," which was afterwards referred by him to this species (Handb. p. 508), is probably an error due to a mistake in the identification of the host. He states that the Puccinia on the same leaf was absolutely identical with P. pruni-spinosae., and omits the reference in the fourth edition of " Microscopic Fungi." No one else has found such a Puccinia on Rhamnus. The leaf in " Micr. Fung." ed. i. 210, is no doubt a Primus leaf. The mycelium of the pecidial stage is perennial ; it penetrates in spring into the growing shoots which become deformed, the affected leaves are narrower and paler, and the flowers are usually imperfect or altogether wanting. These plants it is which cause fresh infections of the Plum-trees G. u. 14 210 PTJCCINIA every year; infection l>y over-wintered uredospores has been proved by Tranzschel to be possible, l>ut as Brooks shows I. c. it id probably rare, because the plum-leaves are generally aol affected until summer is well advanced. Fresh infections of the Anemone can, of course, be produced by the basidiospores of the over-wintered teleutospores. The distinction usually made, by describing the uredospoie> a> " echinulate," and the teleutospores as " verrucose," does not convey the exact truth ; the markings on both art- very similar, but the warts of the uredospores are sharp-pointed and usually turned downwards, while those of the teleutospores are often blunted, and always darker and more crowded. The brothers Sydow describe a second form of uredospore, which I cannot find. The two cells of the teleutospore separate with the greatest readiness, and the lower cell which is very often paler and imperfect, could then be easily mistaken for a uredospore and has been so described and figured. The true uredospores, mentioned in the description, are very similar to amphispores, and have been mistaken at times for paraphyses. The teleutospores are attached by short fragile pedicels in bunches to a common basal cell. This is one of the characters of Arthur's genus, Tramschelia. Arthur describes (North Americ. Flora, p. 150) a second species of Tramschelia (P. cohaesa Long, from Texas), agreeing in almost every minute detail with P. Pruni-spinosae, but having all its four spore-forms upon Anemone decapetala. In its teleutospores P. fusca (q.v.) agrees exactly with both of these, so that /'. cohaesa may be regarded as a primitive form, from which both the others have been evolved. See Grove, New Phytologist, 1913, p. 89. Jacky (Centralbl. f. Bakter. 2. viii. 658) divides P. Pruni-spinosae into two forms : /. typiea, in which the teleutospore has both cells alike, on the three species of Prunes mentioned ; and /. discolor, in which the teleuto- spore is thickened above, and the lower cell is paler, narrower and imperfect, on Amygdalus communis and P. Persica, less often on the other species. On P. Armeniaca both forms are found. This difference is by no means constant, however, and is hardly worthy of mention. Distribution : Europe, North and South America, Africa and Australia. 80. Puccinia Rhodiolae B. et Br. /'»ccinia Rhodiolae B. et Br. Ann. Nat. Hist. ser. 2, v. 462. Cooke, Handb. p. 505 ; Micr. Fung. p. -111. Plowr. Ured. p. 207. Sacc. Syll. vii. 701. Sydow, Monogr. i. 491, f. 401. ON CRASSULACE/E 211 Teleutospores. Sori araphigenous or on the stems, but generally hypophyllous, scattered or crowded and confluent, minute, roundish, surrounded by the torn epidermis, pulverulent, dark- brown ; spores broadly ellipsoid, depressed, rounded at both ends, scarcely thickened above, not constricted, smooth, dark chest- nut-brown, 20—35 x 17—24 p : pedicels hyaline, about as long as the spore ; spores occasionally three-celled, like Triphragmium. On Seduvt Rhodiola (roseum). Very rare. Glen Callater, July, 1844 (W. Gardiner). (Fig. 157.) Distribution : Norway. Fig. 157. P. Rhodiolae. Teleuto- spores, from Glen Callater (ex herb. Berkeley). 81. Puccinia Umbilici Guep. Puccinia Umbilici Guep. in Duby, Bot. Gall. ii. 890. Cooke, Handb. p. 505 ; Micr. Fung. p. 211, pi. 4, f. 80—1. Plowr. Ured. p. 204. Sacc. Syll. vii. 700. Sydow, Monogr. i. 492, f. 403. Teleutospores. Sori amphigenous or on the petioles, on yellowish spots, minute, roundish, usually circinate, at length confluent and forming large orbicular clusters up to 1 cm. diain., at first compact, then pulverulent, dark reddish-brown; spores broadly ellipsoid or subglobose, rounded at both ends, not thickened above but surmounted by a minute subhyaline apiculus or pore-cap, not constricted, smooth, bright chestnut- brown, 28 — 32 x 18 — 26 [x, the cells often depressed {i.e. broader than long) and frequently oblique ; pedicels short, hyaline. On Cotyledon Umbilicus. Locally common. May and June ; in mild localities it can be found even as early as January. (Fig. 158.) Fig. 158. P. Umbilici. Teleutospores. Distribution : France, Belgium, Portugal. 14—2 ■212 PUCCINIA 82. Puccinia Ribis 1 >< '. Puccinia Ribis 1><'. Flor. IV. ii. 221. Gard. Chron. 1894, xvi. L35. Trans. Brit. Myc. Soc. i. 57. Sacc. Syll vii. 679. Sydow, Mqnogr. i. 496. Teleutospores. Sori epiphyllous, orbicular, surrounded by a discoloured yellow zone, circulate and often confluent, pulverulent, rich chestnut-brown; spores oval or oblong, rounded above and be- low, apex thickened slightly and hooded, hardly constricted, verru- culose or rather punctate, chest- nut-brown, 20—30 x 15— 20 /a; pedicels hyaline, thin, deciduous, about as long as the spore; a few mesospores intermixed. I )n leaves of Ribes rubrum. Very rare. Dallas Manse ( tarden, Elginshire, July 16, 1894 (Rev. Dr Keith). (Fig. 159.) The pore of the lower cell is always towards the base, near the insertion of the pedicel. Eriksson showed that the teleutospores do not germinate until they have passed through the winter. He considers the form on Ribes rubrum as biologically distinct from that on It. nigrum or It. Grossularia. Distribution: Central and Northern Europe, North America. F.ig. 159. P. Ribis. Teleutospores and mesospore. 83. Puccinia Saxifragae Schlecht. Puccinia Saxifragae Schlecht. Flor. Berol. ii. 134. Plowr. Ured. p. 208. Sacc. Syll. vii. 678. Sydow, Monogr. i. 500. Fischer, Ured. Schweiz, p. 151, f. 117. /'. Saxifragarum Cooke, Micr. Fung. p. 209 (non Handb. p. 506). Teleutospores. Sori generally hypophyllous, on discoloured spots, round, scattered or aggregated and confluent and then irregular, soon naked, pulverulent, dark-brown : spores ellipsoid or oblong, rounded at both ends or slightly attenuated below, slightly constricted, often surmounted by a rather large -pale ON SAXIFRAGACE^i 213 conical papilla, marked with faint, sometimes curved, longitu- dinal stria?, pale-brown, 26 — 45 x 14 — 20 fju\ pedicels hyaline, slender, deciduous, not as long as the spore. On leaves and petioles of Saori- fraga granulata, S. stellaris, S. um- brosa. Rare. August. (Fig. 160.) The markings on the teleutospores are perfectly invisible when wet. This species has no connection with Caeoma Saxifragae which is also found on S. granulata. I have specimens from both Scotland and Ireland, the latter on S. umbrosa from Clare Island, and the former on S. stellaris from Lochnagar. Fig. 160. P. Saxifragae. Te- leutospores, a, on .S'. umbrosa (Ireland), b, on S. stellaris (Lochnagar). Distribution : Central and Western Europe. 84. Puccinia Pazschkei Dietel. Puccinia Pazschkei Diet, in Hedwig. 1891, p. 103 ; Ber. deutsch. Bot. Gesell. ix. 44, pi. 3, f. 15. Sacc. Syll. xi. 185. Sydow, Monogr. i. 503, f. 411. Fischer, Ured. Schweiz, p. 148, f. 113. Trans. Brit. Myc. Soc. iii. 123. Teleutospores. Sori epiphyllous, about \ — 1 mm. wide, scattered or more often in orbicular groups 2 — 3 mm. diam., a few occasion- ally hypophyllous, surrounded by the swollen and torn epidermis, pulverulent, dark reddish-brown ; spores ellipsoid or oblong, rounded at both ends, very slightly thickened above or with a minute flat papilla, gently constricted, faintly and irregularly verruculose, pale clear-brown, 25 — 35 x 13 — 18 fi; pedicels hyaline, short, deciduous; a few mesospores intermixed. On leaves of Saxifraga longifolia, Kew Gardens (G. Massee), Journ. Bot. xlvi. 152. On a hybrid between S. Cotyledon and S. aizoon, Sutton Coldfield, April and May, 1911—12. (Fig. 161.) A parasite doubtless introduced into this country with the plants. In the Sutton example, the sori form two or three perfectly round groups Fig. 161. P. Pazschkei. Te- leutospores and mesospore, (Sutton Coldfield). 214 l'i CCINIA towards the tip <>f each affected leaf, on the upper side. Fischer records it on S. aizoon and S. elatior. Distribution : Switzerland, Austria. Fit:. 162. P. Chrysosplenii. Teleutospores ; a, the form fragilipes. 85. Puccinia Chrysosplenii (Jrev. Puccinia Chrysosplenii Grev. in Engl. Flora, v. 367. Cooke, Handb. p. 506 ; Micr. Fung. p. 210. Plowr. Ured. p. 211. Sacc. Syll. vii. 685. Sydow, Monogr. i. 493. Fischer, Fred. Schweiz, p. 318, f. 231. Teleutospores. Sori amphigenous, but generally hypo- phyllous, small, scattered or con- fluent, often circinate, roundish, pulvinate, pale-brown; spores of two kinds — (forma persistens) broadly fusoid, strongly thickened and more or less conical at the apex, rounded or slightly attenuated below, gently constricted, smooth, very pale-brown, 32 — 46 x 10— 1") /n: pedicels hyaline, rather long, persistent ; (forma fragilipes) ob- long-ellipsoid, with a conical papilla, distinctly constricted, yellowish-brown, with faint longitudinal ridges(7), 35 — 42 x 14— 19 fi ; pedicels very deciduous. On Chrysosplenium alternifolium, G. oppositifolium. No1 common. End of March to August or September. (Fig. 1<>2.) The two kinds of spores are similar in form and function to those of /'. Veronicarum ; forma persisteas consists of spores which germinate as -non as mature, forma fragilipes of spores which rest during the winter. It is said that the latter are scarce, but when present they form smaller sori which are less confluent, often solitary, and are frequently found on the upper leaf-surface. Compare the two similar kinds -of spores in P. Circaeae. Distribution: Central and Northern Europe, Eastern Asia. 86. Puccinia Thalictri Chev. Puccinia Thalictri Chevallier, Flor. Paris, i. 417. Plowr. Ured. p. 206. Sacc. Syll. vii. 680. Sydow, Monogr. i. 550. Fischer, Ured. Schweiz, p. 94, f. 72. OX RANUNCULACE.E 215 Teleutospores. Sori hypophyllous, scattered or gregarious, often occupying the whole leaf, roundish, soon naked, pulverulent, dark-brown; spores much constricted, not thickened at the apex, the upper cell nearly globose, the lower globose, obovoid or clavate, generally narrower ; the spores separate readily into their component cells, are covered with large pointed warts, dark-brown (the lower cell paler), 26 — 52x18 — 30//,; pedicels hyaline, deciduous. Fig. 163. P. Thalictri. Te- leutospores, from Prof. Trail's specimens. On Thalictrum flavum, T. minus. Very rare; Kinloch Rannoch, Perthshire (Prof. Trail); Kew Gardens. Autumn. (Fig. 163.) This species has all the marks of a perennial mycelium. The same plants are attacked by it year after year ; they are somewhat deformed and taller, with longer internodes, smaller and paler leaves and narrower segments. There is in Fischer a figure of a teleutospore with three cells, looking much like that of a Phragmidium. Cf. Puccinia fusca. Distribution : Northern and Central Europe, Siberia, North America. 87. Puccinia fusca Wint. Mcidium fuscum Pers. in Linn. Syst. Veg. p. 1472 (teleutospores). Puccinia Anemones Pers. Obs. ii. 24. Cooke, Handb. p. 503; Micr. Fung. p. 209, pi. 4, f. 65—6. P. fusca Wint. Pilze, p. 199. Plowr. Ured. p. 205. Sacc. Syll. vii. 669. Sydow, Monogr. i. 530. Fischer, Ured. Schweiz, p. 95, f. 73. Spermogones. Hypophyllous, mixed with the teleuto-sori, blackish. Teleutospores. Sori hypophyllous, rarely on the upper side, generally spread uniformly over the whole surface of the leaves, here and there confluent, small, round, pulverulent, dark-brown ; spores very much constricted, composed of two almost globose or oblong cells which easily separate, densely 216 PUCCINIA covered with large waits, brown, 30—55x15 — 2(> : pedicels hyaline, up t * > 40/z long: occasionally a few one- or three-celled spores are intermixed. On Anemone nemorosa. Common. March June. (Fig. 104.) It has been shown by De Bary and Ed. Fischer that the mycelium is perennial in the rhizome. The attacked plants are deformed and never flower; they bear paler and narrower leaves which are much thickened. The eecidia on the same host are not connected with this species (see Ochropsora Sorbi and P. Pruni-spinosae) ; in fact they do not appear until some time later than the teleutospores oiP.fusca begin to show. Distribution: Europe, Siberia, North America. Fig. 164. P.fusca. TeleUt<'-]i in -. 88. Puccinia Calthae Link. JScidium Calthae Grew Flor. Edin. p. 440. Cooke, Handb. p. 539; Micr. Fung. \>. 196. Puccinia Calthae Link, Sp. Plant, ii. 79. Cooke, Handb. p. 504; Micr. Fung. p. 210. Plowr. Ured. p. 145. Sacc. Syll. vii. 602. Sydow, Monogr. i. 540. Fischer, Ured. Schweiz, p. 310, f. 225. Spermogones. In little clusters, honey-coloured. jEcidiospores. iEcidia hypophyllous, in little clusters on roundish yellowish spots, or on the stems in elongated swellings, cup- shaped, with a torn whitish recurved margin ; spores delicately verrucu- lose, orange, 21 — 28 /x. Uredospores. Sori generally hy- pophvllous, minute, scattered, round- ish, pulverulent, chestnut; spores globose to ellipsoid, echinulate, pale- chestnut, 22—30 x 20—25 fi, with two germ-pores in the upper half. Teleutospores. Sori amphigenous, small, irregularly scat- tered or often circinate, pulverulent, but persistent, black- brown ; spores oblong-clavate or fusoid, generally with a paler Fig. 165. P. Calthae. Teleutospores. ON RANUNCULACEtE 217 conical papilla at the apex, hardly constricted, perfectly smooth, clear chestnut-brown, 30 — 44x13 — 22 /jl; pedicels hyaline, thick, persistent, up to 75 /x long. On Caltha palustris. Rather rare. iEcidia, May and June ; teleutospores, July — October. (Fig. 165.) There is another British species on Caltha palustris, P. Zopfii, which has been usually confounded with the present one in herbaria. It differs in having its teleutospores broader, darker, more oblong, and covered here and there with minute warts, but is otherwise similar in appearance. There are three others found on species of Caltha in North America, all so nearly allied that they are difficult of discrimination ; but of these one has no uredospores, and the others have, so far as is known, teleutospores only. Some of these may be found in Britain. Distribution : Europe, Siberia, North America. 89. Puccinia Zopfii Winter. Puccinia Zopfii Wint. in Hedw. 1880, p. 39, 107. i. 542. Fischer, Ured. Schweiz, p. 91, f. 70. Sydow, Monogr. Spermogones. Amphigenous or on the petioles, in little clusters of 6 — 10, brownish when old. sEcidiospores. J^cidia hypophyllous or on the petioles, usually surrounding the groups of spermogones in scattered Fig. 166. P. Zoj>fii. Leaf of Caltha with aecidia ; two teleutospores. roundish clusters 1 — 2 mm. diani. (elongated on the petioles), at first hemispherical, then shortly cup-shaped, Hattish, yellow, with a short torn scarcely reflexed margin ; spores delicately verruculose, yellow, 20 — 28 /j,. 2 IS PUCCINIA Urcdosporcs. Sori generally hypophyllous, minute, scattered, puncl id in 11, round, soon naked, surrounded by the erect epidermis, chestnut; spores ellipsoid, echinulate, brownish-yellow, 22 — 30 x 2<) — 25 [i, with two or three germ-pores. Teleutospores. Sori similar, but darker; spores oblong to obovate, rounded at both ends, sometimes truncate above or slightly narrowed below, scarcely thickened but with a broad flat papilla at the apex, gently constricted, delicately verruculose, dark chestnut-brown, 35 — 60 x 24 — 35 /x ; pedicels nearly hyaline, short, deciduous. On Caltha palustris. Rather rare; Shropshire, Scarborough, Rhydd-y-fen, Gullan Loch, Ireland (co. Dublin). /Ecidia in May; teleutospores, August — December (June — September, Fischer). (Fig. 166.) Distinguished from P. Ca.lthae essentially by its teleutospores which are provided with a few. rather distant, minute warts, mostly towards the upper end ; these are difficult to see except when viewed dry. The spores are also relatively much broader and not narrowed towards the summit, and are therefore easily seen to be distinct in shape ; they are darker in colour and have shorter pedicels. The secidia are not known to be different from those of P. Calthae ; those described above probably belong to P. Zopfii, because they were found in the same neighbourhood as the teleutospores in Ireland : the cells of the peridium agreed with those figured by Fischer (I.e.). Krieg showed (Centralbl. f. Bakt. 2. xv. 259) that P. Zopfii is autoecious, like P. Calthae. The two species have been frequently confounded in herbaria ; but, if I may judge by the specimens I have seen, the teleuto-sori of P. Calthae are larger, more crowded, more often circinate, more compact, and remain longer covered by the epidermis than those of /'. Zopfii, though this is not always so well marked. Distribution: Central Europe. 90. Puccinia Lychnidearum Link. Puccinia Lychnidearum Link, Sp. PI. ii. 80 (1825). Cooke, Handb. p. 505 ; Micr. Fung. p. 210 p.p. Plowr. Ured. p. 196. P. Arenariae Winter, Pilze, p. 169 (1884). Plowr. Ured. p. 210. Sacc. Syll. vii. 683. Sydow, Monogr. i. 053. Fischer, Ured. Schweiz, p. 307, f. 224. McAlpine, Rusts of Australia, p. 177, f. 97 (introduced on SteUaria media). ON CARYOPHYLLACE.E 219 P. Sperguhie DC. ; Cooke, Micr. Fung, p. 210. Sydow, Monogr. i. 560. P. A/oehrinaiae Fckl. ; Cooke, Micr. Fung. p. 210. Teleutospores. Sori hypophyllous or rarely on the stems, scattered or circinate, on pale spots, sometimes confluent, pulvinate, pallid-brown, then darker, greyish-pulverulent from the numerous basidiospores ; spores oblong-fusoid or clavate, Fig. 167. P. Lychnidearum. Teleutospores, on Lychnis diurna. Fig. 168. P. Lychnidearum. Teleutospores, on Arenaria trinervis. rounded or somewhat pointed above and more or less thickened (up to 10 /a), gently constricted, rounded or attenuated below, smooth, yellowish-brown, 30 — 50 x 10 — 20 /a; pedicels hyaline, persistent, 60 — 85 /x long. On various Caryophyllacere, such as Dianihus barbatus, Lychnis diurna, L. vespertina, Arenaria trinervis, Gypsophila elegans, Sagina nodosa (?), Spergula arvensis (?), Stellaria Holo- stea, S. media, 8. uliginosa. May — November. It is most common on Lychnis diurna, on which as also on Dianthus the sori are remarkably circinate. (Figs. 167, 168.) The two species, P. Lychnidearum and P. Arenariae, are united in Sydows' Monographia, on the ground that our present knowledge, derived from cultures, is insufficient to separate them, and that any apparent morphological distinctions break down completely when a long series of •220 PUCCINIA specimens is examined. The spores are, however, variable in form and colour, and the sori differ in appearance and arrangement ; no doubt the future will timl this Puccinia divided into several biological races, if not into distinct species. Meanwhile, on morphological grounds alone, our British forms may be arranged under three beads: 1. forma Lychnidearum 'Link), on Lychnis; sori medium-brown in colour, often greyish, remarkably circinate, on conspicuous yellow and purple spots. 2. forma Dianthi (DC), on Diantkus; sori larger, darker, and more pulvinate, usually somewhat circinate; =P. DianthiDC. 3. forma Arenariae (Schurn.), on Armaria and Stellaria ; sori paler, not so circinate. spores paler ; =/'. Moehringiae Fckl. The form on Sagina procumbens is so different that it is here reckoned as a separate species. /'. Saginae K. et S. (q.v.). It will be noticed that Plowright separated P. L;i<-hnido of the spores were in this state. Such spores have some- times been wrongly described as having digitate processes like those of /'. coronata. — It was stated by De Bary that he had seen the germ-tubes of the basidiospores of P. Dianthi enter the host-plant through the stomata : no similar case has been detected by any other observer (see p. 38). Distribution : Europe, Siberia, East Indies, North and Smith America. ON CARYOPHYLLACE.E 221 91. P. Saginae K. et S. Pucdnia Saginae K. et S. Exsicc. no. 221. Fuckel, Symb. Myc. p. 51. P. Arenariae Wint. ; Plowr. Ured. p. 210 p.p. Sydow, Monogr. i. 553 p.p. P. Lyehnidearum Link; Cooke, Handb. p. 505 ; Micr. Fung. p. 210 p.p. Teleutospores. Sori chiefly on the stems, compact, confluent, forming blackish-brown masses, about h cm. long, encrusting at intervals the flowering steins which are slightly thickened at that spot, rarely on the leaves ; spores oblong to oblong-clavate, Fig. 169. P. Saginae. Teleutospores, from the Ercal specimen. rounded and not thickened above, slightly constricted, usually somewhat tapering below, smooth, very dilute brown, almost hyaline, 36 — 41 x 15 — 19 /x ; pedicels hyaline, rather stout, persistent, about as long as the spore. On Sagina procumbens. Rare ; specimen in Herb. Phillips (Brit. Mus.) from Ercal, near Wellington, under the name Puce. Lyehnidearum. (Fig. 169.) The >spores of this species, as well as the character of the sori, render it widely distinct from the various forms included under the name P. Lyehnidearum ; the most obvious difference is in the total want of thickening at the apex, unless this was due merely to the fact that all the spores had germinated, which did not seem to be the case. 222 PUCCINIA !'2. Puccinia Behenis Otth. j/Ecidium Behenis 1>('. Flor. fr. vi. HI p.p. Puccinia Behenis Otth, Mitth. Naturforsch. Gesell. Bern ls7. Plowr. Ured. |i. L94. Sacc. Syll. vii. 642. Sydow, Monogr. i. 567. Fischer, [Jred. Schweiz, p. 135, f. 102. ('ri'dnspures. Sori araphigenous, generally hypophyllous, on minute purplish spots, scattered or aggregated, rounded, sometimes confluent, surrounded by the cleft epidermis, cinna- mon ; spores globose to ovate, delicately echinulate, yellow- brown, 23—30 x 20—26 p. Fig. 172. P. Oxyriae. Teleutospores (from a Swiss specimen). Teleutospores. Sori on the leaves similar, but also on the petioles and peduncles and then elongated, pulverulent, black- brown; spores oblong to obovate, rounded and slightly thick- ened or hooded above, constricted, generally rounded below, marked in the upper part with little unevennesses, almost as if corroded, brown, 30 — 46x15 — 25 /x: pedicels nearly hyaline, deciduous, rather long. On Oxyria digyna (reniformis). Rare ; Skye (Buchanan White), Braemar (Prof. Trail). August. (Fig. 172.) The teleutospores were described by the older authors as smooth ; but, as Lagerheim first pointed out, they are really provided with faintly perceptible markings, especially visible towards the summit. Distribution : Switzerland, Norway, Iceland, Colorado. ON POLYGONACE/E 225 95. Puccinia Conopodii-Bistortae Kleb. Puccinia Bistortae DC. Flor. fr. vi. 61. Cooke, Micr. Fung. p. 204 ; Grevillea, ii. 161. Plowr. Ured. p. 192 (?). Sydow, Monogr. i. 571 (all pro parte). Soppitt, Grevillea, xxii. 45 ; Card. Chron. 1895, xviii. 773. P. Conopodii-Bistortae Kleb. Wirtsw. Rostp. p. 318. ^Kcidivspores. /Ecidia on large swellings of a bright orange colour, immersed, spherical or flat, not at all cup-shaped, margin not projecting; spores delicately verruculose, orange, 15 — 20 pu. Uredospores. Sori hypophyllous, minute, roundish, yellowish- red, soon naked; spores globose to shortly ellipsoid, finely echinulate, pale yellowish-brown, 21 — 24 p,. Telmtospores. Sori hypophyllous, scattered or united in roundish groups, soon naked and pulverulent, dark- brown ; spores oblong or subclavate, rounded at both ends or slightly oblique at the apex, faintly constricted, clear yellowish-brown, 28 — 42 x 16 — 25 p, ; epispore equally thick, smooth, but some- times marked with a very few longitudinal or m^ oblique rows of delicate warts ; no papilla on P. Conopodii- the germ-pores ; pedicels hyaline, short, very Teleutospoi-e. deciduous. iEcidia on Conopodium denudatum, Yorkshire, May; teleuto- spores on Polygonum Bistorta, June — August, not common. (Fig. 173.) The Puccinias on Pol. Bistorta are not yet well known. There appear to be at least four (or five) distinct forms, which are divisible morphologically into two groups — (1) those whose teleutospores have no apical papilla, (2) those which have a small hemispherical papilla at the summit ; the latter are called by Sydow in the Monographia P. mammillata Schrot. (Pilz. Schles. p. 340). They are all heteroscious. Of the former group, one form has an ajcidium on Conopodium, and its other spore-stages on P. Bistorta. It is the one mentioned above ; the life- history was first demonstrated by Soppitt (I. c). It is remarkable as being the first instance known of a hetercecious Puccinia that had its teleutospores on a Dicotyledon. The second form has its secidium on Angelica silvestris and Carum Carui and its teleutospores on Pol. Bistorta and Pol. cieiparum. It would perhaps be better divided into P. Angelicae- Bistortae Kleb. (= P. Car i- Bistortae) and P. Polygoni-vivipari Karst. The connection in the former has been demonstrated by Klebahn and Fischer ; o. u. 15 22li PUCCINIA as regards Pol. viviparum Kiebahn expresses doubts, but Semadeni was able successfully to infect that species with uredospores from Pol. Bistorta. The form mi Pol. viviparum is here kept provisionally distinct, since the secidium on Angelica has not been found in Britain. Of the second group, P. mammillata, there are two biological races — (1) P. Mei-mammillata Semadeni, on Meum, and (2) /'. Angelicae-mam- millata EOebahn, on Angelica. Neither of these has beep found in Britain. All the four of these are closely allied ; in Sydows' Monographia it is suggested that possibly in all these cases the secidium on the Umbellifer is merely facultative and the Puccinia can maintain itself without that aid. 96. Puccinia Polygoni-vivipari Karst. Puccinia Bistortae DC. ; Cooke, Grevillea, ii. 161 ; Micr. Fung. p. 204 p.p. Plowr. Ured. p. L92. Sacc. Syll. vii. 0.38. Sydow, Monogr. i. 571 p.p. P. Polygoni-vivipari Karst. Enum. Fung. Lapp. p. 2-2]. Fischer, Ured. Schweiz, p. 100, f. 76 (?). U redospores. Sori hypophyilous, scattered, small, roundish, cinnamon-brown, girt by the erect epidermis ; spores roundish to ellipsoid, finely echinulate, pale-brown, 20 — 23 x 16 — 17 /a. Teleutospores. Sori similar, but blackish-brown ; spores elliptical to obovate-oblong, rounded and not thickened above, Fig. 174. P. Polygoni-vivipari. a, two teleutospores from liar Lodge; b, uredo- and teleutospore from a specimen issued by Fuckel, Symb. Myc. 57 ; both on Polygonum viviparum. hardly or not at all constricted, rounded or slightly tapering below, smooth, brown, 20 — 30 x 15 — 20/z; epispore very thin and translucent ; no papilla on the germ-pores ; pedicels deciduous, short. On Polygonum viviparum. Very rare. Xear Mar Lodge, August, 1822 (Dr Greville). Braemar, August, 1882 (Prof. Trail). (Fig. 174.) This species is closely allied to P. Conopodii- Bistortae (q. v.), but in the absence of all biological information is best kept distinct. I have ON POLYGONACE^E 227 seen no British specimens but those collected at Mar Lodge and Braemar, which are described above ; the spores of these agree with those of a specimen on Pol. riviparum issued by Fuckel (see Symb. Myc. p. 57). It is a purely Alpine species. 97. Puccinia Polygoni-amphibii Pers. JEcidium Geranii DC. ; Cooke, Handb. p. 543; Micr. Fung. p. 199 p.p.;(?). JE. sanguinolentum Lindr. Myk. Notiz. in Bot. Notis. 1900, p. 241. Uredo Polygonorum DC. ; Grev. Sc. Cr. Fl. pi. 80. Trichobasis Polygonorum Berk. ; Cooke, Micr. Fung. p. 226 p.p. Puccinia Polygonorum Link ; Cooke, Handb. p. 495 ; Micr. Fung. p. 203 p.p. P. Polygoni Pers. ; Plowr. Ured. p. 188 p.p. Sacc. Syll. vii. 636 p.p. P. Polygoni-amphibii Pers. Syn. p. 227 ; Sydow, Monogr. i. 569 p.p. Fischer, Ured. Schweiz, p. 301, f. 220. [Spermogones. Few, amphigenous. jEcidiospores. yEcidia hypophyllous, mostly in concentric groups, on well-marked spots which are deep-red or purplish and often surrounded by a conspicuous green- ish-yellow zone, sometimes occupying the greater part of a leaf, cup-shaped, with a much cut recurved margin; spores finely punctate-verruculose, yellowish, 18— 28 ^.] Uredospores. Sori amphigenous, more often hypophyllous, scattered, roundish, soon naked, pulverulent, brownish; spores ellipsoid to obovate, faintly echinulate, yellowish-brown, 25 — 28 x 18 — 21 ix, with two germ- pores in the upper half. Teleutospores. Sori chiefly hypophyllous, not projecting, long covered by the epidermis, surrounding the uredo-sori in a circular fashion, dark-brown ; spores oblong to clavate, rounded and thickened above (5 — 12 /x), often obliquely conical at the apex, more or less tapering below, gently constricted, smooth, yellow-brown, 35—52x16 — 22 /j,; pedicels nearly hyaline, persistent. 15—2 Fig. 175. P. Polygoni-amphi- bii. Teleutospores, on P. amphibium. 22S PUCCINI A [JScidia on Geranium protease, G. silvaticum\\ uredo- and teleutospores on Polygonum amphibium, I', lapathifolium, July- October; nut uiieumnion. Klobnhn and Bubak report the secidium oe G. /nolle, G. phaeum and other species. (Fig. L75.) Teleutospores are found not only in the separate sori described above, but also in the <>ld uredo-sori. These latter spores arc rounded at the apex ; those which grow in distinct sori often have the thickening cap forced to one side, presumably by the pressure of the persistent epidermis. For the distinctions of the tecidiuin of this Puccinia from that of Uromyces Gt r'i mi which grows upon the same hosts, see under that species (p. 104). The proof of the connection of the secidium with the Puccinia was first given byTranzschel in Russia, and has since been confirmed by Bubak. I have not seen any British specimens of JScidium sanguinolentum ; the description given above is taken from Lindroth. Similarly in Switzerland, Fischer records only the uredo- and teleutospores. On Polygonum amphibium they seem to be confined to the terrestrial form : I have never seen them on the floating leaves. Distribution : World-wide. 98. Puccinia Polygoni-Convolvuli DC. Puccinia Polygoni-amphibii Pers. ; Sydow, Monogr. i. 569 p.p. P. Polygoni-Convolvuli DC. Flor. fr. vi. 61. P. Polygoni Plowr. Ured. p. 188 p.p. Sacc. SylL vii. 63G p.p. Fischer, Ured. Schweiz, p. 303, f. 221. Spermogones and jEcidia. Presumably similar to those of the preceding species. Uredospores. Sori hypophyllous, roundish, soon naked, brown ; spores ellipsoid to obovate, echinulate, clear yellowish-brown, 24 — 30 x 15 — 20 /x, with two opposite germ-pores. Teleutospores. Sori hypophyllous, scattered, roundish, compact, pulvinate, about \ mm. diani., soon naked and pulverulent, black ; spores oblong to clavate. rounded or occasionally conical above and thickened (up to 8/a), faintly constricted, gently attenuated below, smooth, brown, 32 — 45 x 18 — 21 fx : pedicels clear yellow-brown. persistent, nearly as long as the spore. Fig. 17G. P. Polygoni-Con- volvuli. Teleutospores (ex herb. Cooke). ON THESIUM 229 /Ecidia on Geranium pusillum, and possibly also on G. molle, G. rot a ml! folium; uredo- and teleutospores on Polygonum Con- volvulus, August and September. Uncommon. (Fig. 176.) The connection of the secidium on the first-named host and the Puccinia on Pol. Convolvulus has been experimentally demonstrated by Tranzschel. It is possible that the same parasite also attacks P. dume- toruin, P. Persicaria and others. The recidium is not known for certain to have occurred in Britain. The teleuto-sori of P. Polygoni-Convolvuli are distinguished from those of P. Polygo?ii-amphibii by their compact pulvinate form, and by being soon uncovered by the epidermis, while the spores (perhaps in consequence of that) are much darker at the summit, and the apex, if conical, is less often oblique. According to Sydbw these distinctions, however true they may be of the European forms of the species, do not avail when the extra-European forms are considered. In the Monographia, therefore, the two species are united, and only culture experiments will be able to decide the question. 99. Puccinia Thesii Chaill. yEcidium Thesii Desv. in Journ. de Bot. ii. 311 p.p. Cooke, Handb. p. 537 ; Micr. Fung. p. 195, pi. 3, f. 50—1. Puccinia Thesii Chaill. in Duby, Bot. Gall. ii. 889. Cooke, Handb. p. 495 ; Micr. Fung. p. 204. Plowr. Ured. p. 145. Sacc. Syll. vii. 602. Sydow, Monogr. i. 585. Fischer, Ured. Schweiz, p. 300, f. 219. Spermogones. Amphigenous, numerous, amongst the aecidia. Fig. 177. P. Thesii. Teleutospores, from a Surrey specimen. jEcidiospores. /Ecidia amphigenous, scattered uniformly and rather thickly over the whole leaf-surface, seldom in roundish 230 puccin] \ or oblong groups, between cylindrical and cup-shaped, with a white born recurved margin; spores delicately verruculose, orange, 16 — 24^. Uredospores. Sori amphigenous or on the stems, distributed irregularly, minute, roundish, long covered by the epidermis, brown : spores globose to broadly ellipsoid, verruculose, yellow- brown, 20 — 28 fi, with four or five germ-pores. Teleutospores. Sori similar, but more compact, brown-black ; spores oblong to clavate, generally rounded above and slightly thickened, hardly or not at all constricted, rounded or attenuated below, smooth, uniformly brown, 35 — 54x16 — 24 //. ; pedicels brownish, thick, not very persistent, short or as much as 95 //, long ; a few mesospores intermixed. On Thesium hv/mifusum. Rare; Surrey, Dorset, Wilts., Hants., Cambs., etc. ^Ecidia, May — August (also recorded for October); teleutospores, August — October. (Fig. 177.) It was maintained by Vuillemin (Bull. Soc. Mye. France, 1894, p. 107 ff.) that this species has no aecidiuni-stage, the Mcvdium Thesii being referred by him to another species, P. Desvauxii (= P. Passerinii), distinguished by its broader and warted teleutospores ; but both Sydow and Fischer have disproved this contention. The secidia are rather more common than the teleutospores, and are sometimes found among the teleuto-sori on the same leaf, in October. Distribution : Europe, Eastern Siberia. 100. Puccinia Iridis Wallr. Uredo Iridis DC. Encycl. viii. 224. Plowr. Ured. p. 257. Trickobasis Iridis Cooke, Micr. Fung. p. 227. Puccinia Iridis Wallr. in Rab. Krypt. Fl. ed. i. p. 23 (1844). Plowr. Ured. p. 189. Sacc. Syll. vii. 657. Sydow, Monogr. i. 598. Fischer, Ured. Schweiz, p. 236, f. 186. P. truncata B. and Br. Ann. Nat. Hist. ser. 2, xiii. 461 (1854). Cooke, Handb. p. 494 ; Micr. Fung. p. 203. Uredospores. Sori amphigenous, solitary or somewhat aggregated, roundish or elongated, minute, long covered by the epidermis, not pulverulent, reddish -brown; spores globose to ON IRIS 231 ovate, ochraceous-brown, echinulate, 20 — 35x16 — 26 jjl; epi- spore thick, with 2 — 4 (or more) germ-pores. Teleutospores. Sori hypophyllous, few and irregularly scat- tered, sometimes confluent, oblong, compact, persistent, soon naked, black : spores clavate or oblong, much thickened (up to 14 //,) at the summit and rounded, less often conical or truncate, gently constric- ted, usually attenuated below, smooth, fuscous-brown, darker above, 30 — 52x14 — 22yx; pedicels brown- ish, thick-walled, persistent, about as long as the spore. Fig. 178. P. Iridis. Teleuto- spores, on I. Pseudacorus. On Iris foetidissima, I. Pseuda- corus, and on many cultivated species of Iris. Not common. May — October ; the teleutospores may be found on the old leaves until the following spring. (Fig. 178.) The uredo-stage is most abundant and assumes various characters, especially as regards the formation of spots ; when present these are conspicuous, oblong or oval, and yellowish, often with a greenish-brown circumference. In that case the leaf looks remarkably variegated. Plowright considered that the form which occurs on our cultivated Irises is different from that on our native species, because he could not find any teleutospores in the former ; other authors consider them as the same, because the teleutospores on many species are difficult to find, and appear only on dying leaves, especially towards the base, at the end of the season. One can easily recognise them by their being naked ; for in this species, contrary to the usual state of things, the uredo-sori remain long concealed by the epidermis and the teleuto-sori soon become uncovered. The uredospores are very thick-walled ; they can survive the winter and reproduce the fungus in the spring. This species might be hetercecious : no experimental cultures appear to have been made. It will probably turn out to possess several biological races, for it has been recorded on more than thirty-five species of Iris. Distribution : Europe, Asia, North America. 232 PUCCiNiA 101. Puccinia Schroeteri Pass. Pucdnia Schroeteri Pass. Nuov. Giorn. Bot. Ital. vii. 2.")."). W. (;. Smith. Gard. Chron. L889, v. 725, f. 118. Wolley-Dod, Journ. li'ny. Hurt. S(ic. xii. p. liii. Plowr. Trans. Brit. Myc. Soc. i. 57. Sacc. Syll. vii. 732. Sydow, Monogr. i. 608. Fischer. [Jred. Schweiz, p. 78, f. 59. Teh ntnspores. Sori amphigenous, chiefly epiphyllous, large, oblong or elliptic, surrouni led by a brownish-violet discoloration, 1 — 3 mm. long, solitary or in small clusters, long covered or half uncovered and surrounded by the lead-coloured epidermis, o©0 q G (3® vD / Pig. 179. P. Schroeteri. a, meso- and teleutospore, from the original specimen of W. G. Smith ; I, teleutospore and sori on leaf of daffodil, from Cooke's specimen. blackish-brown : spores ellipsoid or oblong, rounded at both ends, not thickened above, hardly constricted, obscurely reticu- lated, golden-brown, then chestnut, 40 — 60 x 25 — 20 /x : pedicels hyaline, short, thick, deciduous ; raesospores also occur. On Jonquil and Narcissus poetic as. Very rare: Malpas, May, 1880 (Rev. C. Wolley-Dod) ; in Gard. Chron. and Journ. Roy. Hort. Soc. (I.e.) it is stated to have been found also on the "common double Narcissus" (? X. telamonius plenus). There .-in- some specimens in Herb. Brit. Mus. on daffodil leaves, sent to the Gard. Chron. by a correspondent, May, 1804, and labelled P. Liliacearum by Cooke, which on examination prove to be this species. (Fig. 170.) on liliacea: 233 Plowright observed that the spores would not germinate at once, but, by securing the affected leaves during the winter near some plants of N. poeticus, he found the Puccinia reproduced next year and for eight or nine years afterwards, though only on the tips of the leaves. The reticulation of the spores varies in character, sometimes resolving itself into longitudinal ridges or rows of warts. Mesospores and other abnormal spores are recorded by Fischer. Distribution : Belgium, Italy, Carniola. 102. Puccinia Asparagi DC. Puccinia Asparagi DC. Flor. fr. ii. 59-5. Cooke, Handb. p. 494 ; Micr. Fung. p. 203. Plowr. Ured. p. 144. Sacc. Syll. vii. 601. Sydow, Monogr. i. 015. Fischer, Ured. Schweiz, p. 235, f. 185. Spermogones. In little clusters, honey-yellow. ^Ecidiospores. zEcidia in oblong groups on the stems, for a long time closed, then shortly cup-shaped, with a whitish, erect, torn margin; spore delicately verruculose, orange, 15 — 28 p. Fig. 180. P. Asparagi. Teleutospores anchmesospore. Uredospores. Sori oblong, narrow, flat, long covered by the epidermis, cinnamon-brown ; spores globose to ovate, delicately and densely echinulate, pale-brown, 20 — 30x17 — 25/i, with four germ-pores. 234 lMCCINIA Teleutospores. Son oblong or linear, often confluent, generally on the stems, rarely on the leaves (phylloclades), blackish-brown; spores ellipsoid to clavate, thickened (up to 8/*) and rounded above, very gently constricted, rounded below, smooth, brown, 35 52 x 17 — 2(i^; pedicels hyaline or brown- ish, persistent, as long as or longer than the spore; a few obovate mesospores intermixed. On Asparagus officinalis. iEcidia, not common, May; undo- and teleutospores, September — December, rather fre- quent. (Fig. 180.) Fischer points out that the connection between the secidium and the other spore-forms has not yet been demonstrated. This disease is often very destructive to asparagus beds ; all diseased shoots should be gathered and burnt. The best means of prevention is by the selection of resistant varieties, and the avoidance of overcrowding. Distribution : Europe, Abyssinia, North America. 103. Puccinia Liliacearum Duby. Puccinia LUiacearum Duby, Bot. Gall. ii. 891. Plowr. Ured. p. 107. Sacc. Syll. vii. 668. Sydow, Monogr. i. 627. Fischer, Ured. Schweiz, p. 76, 545, f. 57. Spermogones. Numerous, especially at the apex of the affected leaves, yellowish, conical. [JHcidiospores. ^Ecidia few, scattered, minute, deeply im- mersed, whitish, the narrow opening only pro- jecting; spores minutely verruculose, orange, about 15— -20 p.] Teleutospores. Sori amphigenous, embedded in the dried yellowish parts of the leaf, hemi- spherical, densely crowded, often confluent, long covered by the ashy-grey epidermis which at length opens by a cleft, then naked, pulverulent, reddish-brown ; spores oblong-fusiform or cla- vate, not thickened above and rounded or fre- quently somewhat narrowed, not constricted, indeed broadest at the septum, attenuated bel< >w, Tde utospore smooth' pallid-brown, 40-75 x 22-35 p. ; pedi- (Lytham). eels hyaline, thick, rather long. ON LILIACE.*: 235 On Ornithogalum umbellatum. Rare; Lytham and near Carlisle (Rev. Hilderic Friend). March— May. (Fig. 181.) As usual, only spermogones and teleutospores are present on these specimens. Fischer records that he found the fungus on 0. umbellatum in .March and April in great plenty, with spermogones and teleutospores, but no secidia ; the infested parts of the leaves were swollen, compact, and harder than the healthy portions. He suggests (Centralbl. f. Bakt. 2. xv. 230) with great probability that the recidia with spermogones which are also found on Ornithogalum (JEcidhom omithogaleum Bubak, Annal. Myc. iii. 222) belong to some heteroecious species. The fungus on 0. umbellatum is a biologic race ; it does not attack 0. nutans, much less other allied species of Liliacere, such as Muscari and Hyacinthus, although P. Lilia- rearuiK is recorded on them. Fischer adds that the mycelium is not perennial, but infection takes place afresh each spring by the basidiospores of the overwintered teleutospores, from leaves lying on the ground. Distribution : Central Europe. Cooke, Handb. p. 528 ; Micr. Micr. Fung. p. 212. 104. Puccinia Porri Wint. Credo Porri Sow. Engl. Fl. pi. 411. U. Alliorum DC. Flor. fr. vi. 82 p.p. Fung. p. 217 p.p. Uromyces Alliorum DC. ; Cooke, Handb. p. 518 Plowr. Ured. p. 137 p.p. Puccinia Porri Wint. Pilze, p. 200. Plowr. Ured. p. 148. Sacc. Syll. vii. 605. Sydow, Monogr. i. 610. Fischer, Ured. Schweiz, p. 80, f. 61. Uredospores. Sori amphigenous, on indeterminate pallid spots, scattered or more or less in rows, minute, at first covered by the swollen epidermis, yellowish or reddish-yellow ; spores globose to ellipsoid, very delicately echinulate, yellowish, 20—30 fi. Teleutospores. Sori amphigenous or caulicolous, generally without spots, scattered, minute, oblong or roundish, about 1 mm. wide, but sometimes confluent into larger patches, long covered by the lead-coloured epidermis, black- brown ; spores oblong or clavate, rounded or rather truncate Fig. 182. P. Porri. Teleuto- spores and mesospore, on A. Schoenoprasum. 236 PUCCINIA above, scarcely or slightly thickened, gently constricted, rounded or attenuated below, smooth, brown, 2.x - 52 x20 — 26 n : pedicels hyaline, short, deciduous; mcsospoivs numerous, obovate or pyriform, very irregular, 22 — 36x15 — 23 fx. On leaves, sheaths and stems of various species of Allium, A. Cepa, A. Scliuen<>]>r>r',i. C. stricta, ami ('. resicaria among British species. It must not be assumed, however, without trial that these are all identical ; the forms on C. hirta, ' '. v< siearia seem to be biologically distinct (Klebahn, Fischer), and others may belong to /'. Pri)igsheimiana, etc. C. binervis is included by Plow-right, hut tin- fungus on this host has not yet been experimentally shown to belong to jE. Urticae. This parasite has been the subject of numerous investigations, since Magnus in 1872 first showed that the recidium on the Nettle belonged to the life-cycle of P. Caricis on Carex hirta. Plowright, Se-hroter, Klebahn, and many others have followed in his steps. It may !>•■ remarked here that few species of Puecinia on Carex can he determined with certainty until the secidium-stage belonging thereto is known. The teleutospores of this species germinate after the winter's rest : they succeed most easily during April. They may be found on new or old leaves of Carex all the year round. It is almost in vain to look for the secidium on the Nettle except in the vicinity of water where Sedges are growing ; but where such a conjunction occurs, the parasite may often be found upon both hosts in abundance every year. Klebahn records the secidium on Urtica wrens ; Magnus states that the fungus can winter on C. hirta by means of its uredospores. Distribution : Europe, Siberia, Japan, America, Australia. 109. Puecinia Pringsheimiana Kleb. JEcidivra Orossulariae DC. Flor. fr. vi. 92. Cooke, Handb. p. 541 ; Micr. Fung. p. 197. Plowr. Ured. p. 263. Puecinia Pringsheimiana Kleb. in Zeitschr. f. Pflanzenkr. v. 76. Soppitt in Card. Chron. 1898, xxiv. 145, f. 38. Sydow, Monogr. i. 652. Fischer, Ured. Schweiz, p. 268. Massee, Diseases of cultivated PI. p. 300, f. 88. ON CYPERACE/K 243 JEcidiospores. /Ecidia hypophyllous, crowded on red and yellow spots, roundish, also in elon- gated clusters on the young branches, petioles, and nerves, sometimes en- tirely covering the young fruits, shortly cylindrical, with broad, re- curved, white, much torn margin ; spores orange, verruculose, 15 — 21 x 14—18/*. Uredospores. Sori hypophyllous, punctiform, about ^ mm. long, on yellowish spots ; spores more or less globose, pale-brown, echinulate, 18 — 22 x 17 — 21 fi, with three, rarely four, germ -pores. Teleutospores. Sori amphigenous, linear or punctiform, up to 1 mm. long, pulvinate, brownish-black; spores resembling those of P. Caricis, 40—58 x 15—22 p. Fig. 187. P. Pringshe imi 'ana. Leaf of Ribes Grossularia with groups of ascidia. ^Ecidia on Ribes Grossularia, R. nigrum (?), May and June, common; uredo- and teleutospores on G. acuta, G. caespitosa, G. Goodenovii, G. stricta. (Fig. 187.) The teleutospore-hosts are those given by Klebahn and Fischer. Klebahn first suggested the connection of the secidium with a Puccinia on Carex, and has since demonstrated the truth of this idea by many culture experiments. Soppitt also showed the same for Carex acuta and C. Goode- novii. The secidium is said to attack R. alpinvm, R. aureum, R. rubrum, R. sanguineum, but less frequently. Plowright records the secidium on leaves of Ribes nigrum (Norfolk, June, 1890), but there is no proof that it belonged to this species. This species is one of those forms originally named by Klebahn P. Ribesii- Caricis ; he has since divided them under five heads which can scarcely be reckoned anything but biological races : — P. Pringsheimiana, P. Ribis- nigri-Acutae, P. Ribis-nigri-Paniculatae, P. Pseudo-cyperi, and P. Magnum (the latter on C. acutiformis and C. riparia). The same species of Ribes serve as alternate hosts in each case, in varying degrees of susceptibility, except that P. Magnusii is not recorded for R. rubrum and R. Grossularia. The morphological differences between these forms are slight and elusive. P. Pringsheimiana can be distinguished from P. Caricis by its nearly 16—2 244 PUCCINIA round uredospores, but most of the other biological races of /'. Riberii Caricis have them oval <>r oblong. There is no remedy for this disease on the ( Juoseberry '".' t<> gather and burn all diseased leaves and fruit, etc., and even this will be of no avail so long as the affected Carices continue to exist. Luckily the disease rarely does much harm. 110. Puccinia dioicae Magn. .Kctdiuni Cirsii DC. Flor. fr. vi. 94. Puccinia dioieae .Magn. Tag. Nat. Vers. Miinchen, L877, p. 200. Plowr. Ured. p. 173. Sacc. Syll. vii. 629. Sydow, Monogr. i. 653. Fischer, Ured. Schweiz, p. 283, f. 208. Spermogones. /Ecidiospores. In little clusters, honey-coloured. iEcidia hypophyllous, on roundish yellow or brownish spots, in clusters 2 — 5 mm. diam., cup-shaped, with torn white margin : spores delicately verruculose, orange, 18 — 25 fj.. [Uredospores. Sori scattered, minute punctiform, brown ; spores globose to ellipsoid, echinulate, pale-brown, 18 — 25 ft. Teleutospores. Sori scattered, roundish or oblong, 1 mm. long, soon naked, sur- rounded by the cleft epidermis, pulvinate, black ; spores clavate, rounded or conical and much thickened (up to 14 fj,) above, gently constricted, tapering below, smooth, brown, darker at the apex, 35 — 56 x 14 — 20 /a, occasionally 70 /z lung; pedicels brownish, persistent, as much as 50 /u. long.] iEcidia on Girsium palustre, C. pratense, and (on the conti- nent) on other species of Girsium, •June, July; uredo- and teleutospores on Gareoc dioica, C. DavallianaiJ). Very rare; Scotland, Ireland. I have not seen the teleutospores. (Fig. 188. ) Fig. 1S8. P. dioii .Kcidia on C. pratense., co. Mayo (J. Adams), nat. size. Distribution: Northern parts of Europe. OX CYPERACE.E 245 111. Puccinia silvatica Schrot. jEddium Taraxaci K. et S. Myk. Heft. i. 85. Puccinia silvatica Schrot. in Cohn, Beitr. iii. 68. Plowr. Ured. p. I 72. Sacc. Syll. vii. 627. Sydow, Monogr. i. 056. Fischer, Ured. Schweiz, p. 289, f. 211. Spermogones. In little clusters, yellowish. jEcidiospores. /Ecidia hypophyllous or amphigenous, on roundish yellow or brown spots, in crowded clusters 2—5 mm. wide, rarely solitary, occa- sionally on the peduncle which they distort, cup-shaped, with torn whitish revolute mar- gin; spores nearly smooth, orange, 14 — 21 /x. [ Uredospores. Sori hypophyllous, scat- tered, minute, oblong, brown; spores globose to ovate, echinulate, brown, 20 — 27 x 15 — 22yCA. Teleutospores. Sori hypophyllous, scat- tered, minute, roundish or oblong, reaching 1 mm. in length, pulvinate, black ; spores cla- vate, rounded and much thickened (up to 11 fi) above, rarely conic-ally attenuated, gently con- stricted, tapering below, smooth, pale-brown, darker at the apex, 35 — 55 x 12 — 18 /a; pedicels brownish, persistent, as much as 40 /x long.] iEcidia on Taraxacum officinale, June. July. I have seen specimens from both England and Ireland. Teleutospores on species of Garex. (Fig. 189.) The question whether /'. silvatica occurs in Britain is still in the same state as in Soppitt's time. The secidium agrees with the one assigned to that species, but the teleutospores have not been found ; there is no evidence that those found on Garex remota at Kew belonged to this species. The distinction of this recidium from that belonging to P. variabilis lies chiefly in the clustered peridia, which are situated on a thickened part of the leaf, the peridium-cells are arranged in evident rows and the spores form, according to Juel, longer chains. Fischer insists that the peridium cells are thickest on the outer side, while those of P. variabilis are thickest on the inner side : he maintains that this difference is characteristic of Fig. 189. P. sil- vatica. iEcidia on leaf of Tarax- acum, from one of Soppitt's spec- imens, Saltaire. 246 PUCCINIA heteroecioua and autoecious species respectively, though this is certainly qo1 always true The only British species of Carex which have been proved by cultures to be connected with this secidium are ('. caryophyllea (praecox) and 0, armaria, but it is supposed to grow also on a large number of other species. The matter is, however, complicated \>y the fact that there are two other Pucciniae /'. arenariicola and P. Schoeleriana) which are very closely allied and differ chiefly in having their secidia on other hosts. DISTRIBUTION : Europe and Siberia. 112. Puccinia Schoeleriana Plowr. et Magn. JScidium Jacobaeae Grev. Flor. Edin. p. 445. JE. Compositarum var. Jacobaeae; Cooke, Handb. p. 542 ; Micr. Fung. p. 198 p.p. ./•,'. Senecionis Fischer, Fred. Schweiz, p. 534, f. 335(?). Puccinia Schoeleriana P. et M. Quart. J. Micr. Sci. xxv. (new ser.) pp. 107, 170. Phill. et Plowr. Grevillea, xiii. 54. Plowr. Journ. Linn. Soe. xxiv. 91 ; Ured. p. 171. Sacc. Syll. vii. 027. Sydow, Monogr. i. 659. Spermogones. Epiphyllous, orange. JEcidiospores. iEcidia hypophyl- lous, clustered on roundish yellow or brownish spots as much as 1 cm. diam., cup-shaped, with a torn white reflexed margin ; spores delicately echinulate, orange, 15 — 21 /x. Uredospores. Sori generally hypo- phyllous, on yellowish spots, scattered, minute, roundish or oblong, pulveru- lent, surrounded by the cleft epidermis, pale-brown ; spores globose to ovate, echinulate, yellow-brown, 24 — 30 x 16 — 25 fl. Teleutospores. Sori hypophyllous, scattered or aggregated, \ — H mm. long, oblong, surrounded by the turn epidermis, pulvinate, black ; spores Fi,. 190. p. schoei^nu, clavate or fusoid, rounded or conically Teleutospores, from one of attenuated above and much thickened Plowright's cultures on C. ^ ^ ^ ^^ constricted| tapering ON CYPERACEyE 247 downwards, smooth, brown, darker at the apex, 45 — 80 x 18 — 22 yu- ; pedicels brownish-yellow, persistent, 25 — 40 /u, long. iEcidia on Senecio Jacobaea, May and June ; uredo- and teleutospores on Garex arenaria, July — May. Rare ; Norfolk, Lincolnshire, Aberdeen, Burntisland. (Fig. 190.) The life-history of this parasite was investigated by Plowright, who at the same time demonstrated by his experimental cultures its distinctness from Puccinia Cartels and P. arenariicola. Fischer records (I.e.) an fecidium on Senecio Jacobaea, S. aquaticus and S. erucifolius closely resembling this, but points out at the same time that C. arenaria does not grow in Switzerland ; if, therefore, it is the same fungus, as it seems without doubt to be, its alternate stage must occur there on some allied species of Carex, such as C. disticha. Distribution : Germany, Holland, Switzerland (?), Russia. 113. Puccinia arenariicola Plowr. Puccinia arenariicola Plowr. Journ. Linn. Soc. Bot. xxiv. 90 ; Ured. p. 170. Sacc. Syll. ix. 311. Sydow, Monogr. i. 661. • JEcidiospores. zEcidia generally hypophyllous, in round clusters on circular yellow spots which are as much as 1 cm. diam. and margined with purple, cup-shaped, yellowish, with torn revolute margin ; spores nearly smooth, yellow, 15 — 20 /t. Uredospores. Sori on yellowish spots, linear or oblong, surrounded by the torn epidermis, brown ; spores globose to ovate, delicately echinulate, pale-brown, 18— 22/i. Teleutospores. Sori generally hypo- phyllous, scattered or sometimes aggre- gated, oblong, as much as 1 mm. long, pulvinate, black ; spores clavate or ob- long-clavate, rounded above where they are darker and much thickened (14//.), gently constricted, tapering downwards, smooth, brown, 40 — 65 x 14 — -22 //, ; pedicels brownish, persis tent, as much as 40 //, long. Fig. 191. P. arenariicola. iEcidia on leaf of Cen- taurea nigra, produced artificially in one of Plow- right's cultures (reduced). 24s PUCCINIA ^Ecidia od Gentaurea nigra, May and .June: undo- and beleutospores on Carex armaria, July — April. Very rare; on the sea-shore, Hemsby, Norfolk. (Fig. 191.) It has not been found anywhere else: but in all probability it could fairly be regarded as merely a well-marked biological race -a- mutation <>f /■. Schoeleriana (q.v.). Plowright's suggestion that it is identical with luistipes Rost. seems less likely, but it is closely allied to P. Caricis- montanae Fisch. 114. Puccinia extensicola Plowr. Puccinia extensicola Plowr. I'red. p. 181. Sacc. Syll. ix. 311. Sydow, Monogr. i. 667. JZcidiospores. ^Ecidia amphigenous or on the stems, seated on paler spots, scattered or in clusters, cup-shaped, whitish- yellow, with torn margin; spores delicately verruculose, pale- orange, 16 — 22 ll. ('li'i/nspores. Sori on extensive pale spots, scattered, mi- nute, oblong or linear, reddish-brown; spores subglobose or ovate, irregular very delicately echinulate, yellowish-brown, 22-30x16— 22 ll. Teleutospores. Sori oblong, ^ — 1 mm. long, covered for a long time by the epidermis which at length splits, pulvinate, black ; spores subclavate, rounded or truncate above, rarely hooded, thickened (up to 8/i), gently constricted, tapering downwards, smooth, brown, 40 — 60 x 18 — 24 m ; pedicels short, hyaline ; a few mesospores occasionally intermixed. Jv'idia on Aster Tripolium, June, July ; uredo- and teleu- tospores on Carex extensa, August — June. Rare; Wells, Norfolk. The life-history of this species was worked out by Plowright in 1888 ; it has hardly been found elsewhere except in Istria. 115. Puccinia paludosa Plowr. JEcidium Pedicularis Libosch. Mem. Moscou, v. 76. pi. 5, f. 1. Cooke, Handb. p. :>44 : Micr. Fung. p. 199. Puccinia paludosa Plowr. IT red. p. 174. Sacc. Syll. ix. 311. Sydow, Monogr. p. 671. Fischer. Fred. Schweiz, p. :>73. f. 203. ON CYPERACE.E 249 Spermogones. In little clusters, honey-coloured. sEcidiospores. iEcidia hypophyllous or on the more or less swollen petioles and stems, clustered in round elongated or irregular groups, cup-shaped, with torn white revolute margin ; spores delicately verruculose, pallid-orange, 15 — 25 /x. Uredospores. Sori hypophyllous, on yellowish spots, scat- tered or in little groups, very minute, roundish or oblong, soon naked, pulverulent, yellowish-brown; spores more or less globose, delicately echinulate, brownish, 20 — 26 /u. Fig. 192. P. paludosa. Teleutospores on C. vulgaris, Norfolk. Teleutospores. Sori hypophyllous, minute, scattered or arranged in long lines, pulvinate, quite black; spores clavate, rounded above and strongly thickened (up to 11 fi), constricted, tapering below, smooth, brown (upper cell very dark), 45 — 60 x 16 — 22 n : pedicels brownish, persistent, rather long. ^Ecidia on Pedicularis palustris, June, July ; uredo- and teleutospores on Care® fulva, G. panicea(l), G. stricta, C. vulgaris (Goodenovii), from July onwards. As usual on Carex, the teleutospores can be found throughout the winter. Rare : Norfolk, Orkney. (Fig. 192.) Distribution : Central and Northern Europe. 116. Puccinia uliginosa Juel. JEcidium Parnassiae Gravis, in Bot. Gall. ii. 904. Cooke, Micr. Fung. p. 198; Greviilea, i. 8 and ii. 161. Plowr. Ured. p. 129. Irish Naturalist, Oct. 1907, p. 321. "250 PUCCINIA Puccinia uliginosa Juel, Ofvers. k. Vetensk.-Akad. F6rh., L894, no. 8, 1>. 410. Sacc. Syll. \i. L98. Sydow, Monogr. p. HT-'i. Fischer, Qred. Scbweiz, p. 267, f. 202. /Eddiospores. .Ecidia hypophyllous, clustered on vdlow, then brown circular spots 2 — 5 mm. diam., or sometimes occupying the whole leaf-surface, cup-shaped, with torn recurved yellowish margin ; spores delicately verruculose, orange, 14— 1 8 fl. [ Uredospores. Sori amphigenous, scattered, very minute, rounded or oblong, pulverulent, yellow-brown; spores globose to ovate, echinulate, brownish, 21 — 25 /x, with three germ-pores. Teleutospores. Sori amphigenous or in little groups, minute, punctiform, roundish or oblong, pulvinate, black; spores oblong or somewhat clavate, rounded above and thickened (up to 8/u,), gently constricted, rounded or attenuated below, smooth, brown, darker at the apex, 30 — 38x12 — 18 /u.; pedicels subhyaline, persistent, 15 — 32 /x long.] .Ecidia on Parnassia vulgaris, June ; uredo- and teleuto- spores on Carex vulgaris (Goodenovii) and its var. juncella. < )nly the secidium recorded for Britain ; Glasgow, Aberdeen, Ireland. 117. Puccinia graminis Pers. JEcidiv/m. Berberidis Grnel. ; Cooke, Handb. p. 538 ; Micr. Fung. p. 195. Trichobasis linearis Lev. ; Cooke, Micr. Fung. p. -I-23, pi. 7, f. 144. P. graminis Pers. Disp. Meth. p. 39. Cooke, Handb. p. 493 ; Micr. Fung. p. 202. Plowr. CJred. p. 162. Sacc. Syll. vii. 622. Sydow, Monogr. i. 692. Fischer, Ured. Schweiz, p. 243, f. 190. McAlpine, Rusts of Australia, p. 120, and many figures. Sperniogones. In little clusters, honey-coloured. JEcidiospores. zEcidia hypophyllous, often also on the fruit, on roundish, often thickened spots, 2 — 5 mm. diam., which are margined with reddish-purple or yellow, clustered or scattered, cylindrical, white, with a cut and somewhat erect margin; spores appearing smooth, verging on orange, 14 — 26 /a diam. Uredospores. Sori amphigenous, often also on the sheaths and culms, scattered or arranged in rows, linear, 2 — 3 mm. long, ON GRAMINE.E 251 often confluent and reaching a length of 1 cm. or more, surrounded by the cleft epidermis, pulverulent, yellow-brown ; spores ellipsoid or ovate-oblong, echinulate, yellow-brown, then yellowish, 22 — 42 x 16 — 22 /x, generally with four equatorial germ-pores. Fig. 193. P. grami nis. a, aecidia on Berberix ; b, nredo- and c, teleutospores on wheat. Fig. 194. P. graminis. Teleuto- and uredo- spore on Aira caexpi- toxa . Teleutospores. Sori similar, but forming long lines, soon naked, pulvinate, black ; spores oblong-clavate, rounded or atten- uated at the summit, much thickened (6 — 13 fi), slightly con- stricted, attenuated below, smooth, chestnut-brown, 35 — 60 x 12 — 22/*; pedicels brownish, persistent, thick, as much as 60 //. long ; paraphyses always absent. /Ecidia on leaves and fruits of Berberis vulgaris and Mahonia Aquifolium ; uredo- and teleutospores on Agropyron caninum, A. repens, Agrostis alba, A. canina, A. vulgaris, Aira caespitosa, A. flexuosa, Alopecurus pratensis, Arrhenatherum elatius, Avena fatua, A. saliva, Brack y podium pinnatum, Bronius arvensis, B. asper, B. erectus, B. giganteus, B. mollis, B. secalinus, Dac- tylis glomerata, Festuca Myurus, F. ovina, F. rubra, Holcus lanatus, H. mollis, Hordeum murinum, H. pratense, B. vulgare, Lolium perenne, Milium effusum, Molinia caerulea, Phalaris canariensis, Poa compressa, P. nemoralis, P. pratensis, P. ■2'rl PUCCINIA trivialis, Secale Cereale, Trisetum flavescens, Triticum vulgare. These ;ire t li> • British grasses among those recorded by Sydow : the teleutospores have not been found, and probably do nol occur, on all these in this country. (Figs, lit:;. 1!)4.) This is the famous " rust " (uredo-) or "mildew" (teleutospores of corn about which so much has been written. But all the earlier observers (.•ont'usoil together the various rusts of Cereals of which a number are now distinguished. /'. graminis is known as the "Black Rust," on account of the dark colour of the teleuto-sori : these are very distinctive, forming narrow black lines, | — 1 cm. long, chiefly on the sheaths and culms. However abundant this species may have been in the past, it is much less common in many parts of England now than some of the following species. Whether this is due to the general extirpation of wild Barberry hushes or not, is not certain ; at any rate they arc very uncommon, and the avidium on the cultivated species of Berberis and Mahonia is rarely met with in England. The test by which alone the Black Rust can be absolutely distinguished is the power possessed by its basidiospores of producing the characteristic secidia on the Barberry. There is a European but possibly non-British species of Puccinia (P. Arrhenatheri) which has also the Barberry for its alternate host, on which it produces peculiar " witches'-brooms," the mycelium living perennially in the twigs. This is JEeidium graveolens Shuttl., which was formerly wrongly identified with /E. magelhaenicum Berk, from Tierra del Fuego. The uredo-stage of P. graminis can generally be recognised in the field by its sori, which may reach a length of 10— -15 mm. and are of a rusty- orange or brownish-ochre colour ; they do not become general till the beginning of June. Forms of P. disperse/, are often mistaken for it. Microscopically, the uredospores are seen to be longer compared with their breadth (more ellipsoid) than is the case with the other cereal species ; the teleutospores, which germinate only after a winter's rest, are longer and have longer pedicels ; their sori form much more conspicuous lines and do not remain for long covered by the epidermis. It is the uredo-stage which dees the greatest harm to the crops ; it is reported to cause much less in the United States, South Africa, Australia and Tasmania, but not much in India. This species has been divided by Eriksson into six biological races, but they are of a very indefinite character and later researches (see Carleton, '99, p. 52) throw grave doubt upon their reality. At any rate, they are not the same in America as in Europe, though this may he explained by supposing that, since these forms are undergoing evolution at the present moment, the course of this evolution is different in America from what it is in Europe. The existence of these races is, however, important; they show that the wheat cannot necessarily be infected by ON GRAMINE2E 25^ the forms which grow upon wild grasses ; according to Eriksson this is absolutely true of his " f. sp. Tritici," but Carleton found that the uredo from wheat would infect certain wild grasses and that the uredo from some of them, in turn, would infect the wheat. Most of these races can equally infect the Barberry ; yet Wheat-Rust abounds in South Africa, Australia and in parts of India, where no species of Berberis are indigenous. McAlpine in particular finds P. graminis on 27 species of Graminea?, yet he was unable to infect the Barberry-plants imported from England, even though wheat planted closely around them was covered with the Puccinia. He comes to the conclusion that the Wheat-Rust of Australia may be a biological race which has lost the power of producing recidia owing to the absence of its secidial host. According to the evidence at present available, this seems also to be the case with other species, e.g. the aacidial host is not known in Australia for P. Agrostidis, P. bromina, P. Festucae, P. Lolii, and especially P. Poarum. The latter case is the most striking, since the Coltsfoot does not exist in Australia, and the uredospores on Poa have been found there all the year round. The only practicable remedy for this disease is to plant seeds of varieties which have been shown to be immune : a certain progress has already been made by Professor Biffen and others in the production of these, and McAlpine mentions a variety, " Rerraf," which has been found to be rust-resistant in many of the Australian States, though it lost that power when transferred to other countries. In the year 1889, which had a wet and "muggy" spring, the loss due to rust for the whole of Australia was estimated to be between two and three million pounds sterling. Distribution : In every country of the world. IIS. Puccinia coronata Corcla. sEcidium crassum Pers. Syn. Fung. p. 208 p.p. Cooke, Handb. ]). 538; Micr. Fung. p. 196 p.p. jE. Frangulae Schum. PI. Sail. ii. 225. Puccinia coronata Corda, Ic. Fung. i. 6, pi. 2, f. 90. Cooke, Handb. p. 494; .Micr. Fung. p. 203, pi. 4, f. 60—2. Plowr. Ured. [». 163 p.p. Sacc. Syll. vii. 623 p.p. Sydow, Monogr. i. 699. Fischer, Ured. Schweiz, p. 373, f. 270. Spermngones. Epiphyllous or amongst the secidia. sEcidios pores. yEeidia hypophy lions or on the petioles, in roundish groups or irregularly scattered,, on yellow or purplish spots, producing distortion especially of the petioles, cylindrical, with a white torn revolute margin ; spores very delicately verruculose, orange, 16 — 25 x 15 — 20 p,. Uredospores. Sori amphigenous, but mostly epiphyllous, 2-")4- PUCCINI A scattered or arranged in rows, rarely confluent, minute, more or less oblong, pulverulent, orange; spores globose to ovate, shortly eehinulate, yellow, 10 — 2o x 14 — '20 /x, with three or four germ-pores (about ten, Fischer, but ?), and mingled with a tew paraphyses. Teleutospores. Sori hypophyllous, irregularly scattered. rarely confluent, oblong or linear, covered by the epidermis, s naked, black; spores cuneate, flat at the summit and crowned with about 5 — 7 obtuse (digital iform) darker-coloured teeth, hardly or not at all constricted, gradually tapering towards the base, smooth, brown, 35 — 60 x 12 — 22 li; pedicels short, rather thick. /Ecidia on Rhamnus Frangula, May and June; uredo- and teleutospores on Agropyron repens, Agrostis alba, A. stolonifera , A. vulgaris, Calamagrostis lanceolata, Dactylis glomerata, Festuca sylvatica, Holcus lanatus, H. mollis, PJudaris arun- dinacea (but not yet recorded on all these grasses in Britain), August — October. Common. On account of the processes at the summit of the teleutospore this species is called Crown Bust. It was surmised by Plowright that there are two Crown Rusts; these have since been called P. coronata and /'. Lolii ( = P. coronifera). They are equally widely distributed, but are said to occur on different grasses, with the exception that they are both found on the two species of Holcus. In accordance with custom, they are here kei>t separate, but aside from the distinction of the hosts they can be separated only by minute differences. When they occur on Holcus, therefore, the only test that could absolutely decide the matter would be to await the maturation of the teleutospores, and then try if they would infect R. Frangula. The lighter-orange colour of the uredo-pustules, and t lie character of the germ- pores will, however, distinguish either of them from P. graminis, when occurring on the same hosts ; in the Crown Rusts, moreover, both kinds of sori are confined almost entirely to the leaves. The teleuto-sori of P. coronata have a less decided tendency to group themselves round the uredo-sori than in P. Lolii, and do not remain so long covered by the epidermis, becoming naked early in the autumn. In the ufedo-sori, Eriksson says that paraphyses occur in P. coronata and hardly at all in P. Lolii ; but the evidence seems to favour the conclusion, in generd, that the presence or absence of paraphyses is a note of little importance. Plowright mentions an interesting fact, confirmed by Pole- Evans, that the Crown Rust, when it occurs on Dactylis, is an early summer OX GRAMINE.K 255 species and is accompanied by few uredospores, while that on Lolium perenne (P. Lolii) occurs only in the autumn with a profuse simultaneous development of uredospores. This agrees with my experience. Never- theless, I consider that P. coronata and P. Lolii are merely biological forms of one species. The experiments of Carleton, in the United State:-, have shown that the distinctions of hosts are quite insufficient to dis- criminate the two forms ; among others he succeeded in infecting a host, stated by Eriksson to be confined to P. coronata, with spores from a host belonging to P. Lolii. P. Festucae Plowr. belongs to the same group, but has its secidium on Honeysuckle. Crown Rusts have been found in Europe on many other species and genera of grasses (Sydow, i. 705), but they cannot, in the absence of cultures, be even temporarily arranged under the two heads. Barclay described a form from Simla (P. coronata var. himalensis) on Brachy podium silvaticum with its secidium on Rhamnus dahurica ; this has since been raised by Dietel to a species (P. himalensis). It is a remarkable fact that the only other species of Puccinia known, provided with the same processes, are two (P. Mesnieriana = digitata, and P. ScJiweinfurthii) whose telcutospores occur on Rhamnus. A somewhat similar form, but with much longer apical processes, is found on Lonicera in Turkestan (P. longirostris Kom.). These instances are similar to those of P. fusca and P. Pruni-spinosae already mentioned and can be explained in the same way ; see Grove, New Phytologist, 1913, p. 89. 119. Puccinia Lolii Nielsen. JEcidium crassum Pers. ; Cooke, Handb. p. 538 ; Micr. Fung. p. 196 p.p. jE. Cathartici Schum. PL Sail. ii. 2:25. Puccinia coronata Corda ; Cooke, Handb. p. 494 ; Micr. Fung. p. 203 p.p. Plowr. Ured. p. 163 p.p. Sacc. Syll. \ ii. 623 p.p. P. Lolii Niels. Ug. for Landrnaend. i. 549 (1875). Sydow, Monogr. i. 704. McAlpine, Rusts of Australia, p. 123 and many figures. /'. coronifera Kleb. Zeitsch. f. Pflanzenkr. iv. 132 (1894). Fischer, Ured. Schweiz, p. 375, f. 271. jEcidiospores. yEcidia hypophyllous or on the petioles, seated on yellow or purplish spots, scattered or arranged in groups, producing distortion of the parts, cylindrical, with a whitish torn revolute margin ; spores very delicately verrucu- lose, orange, 1G — 25 x 12 — 20 ft. Uredospores. Sori amphigenous, scattered or in large patches, forming blister-like swellings, minute, sometimes con- fluent, lanceolate or more or less oblong, pulverulent, orange ; spores globose to obovate, echinulate, yellow, 18 — 27 x 16 — 24/a, 256 I'll < I XI A with bhre ■ four inconspicuous germ-pores; paraphyses very few or want ing. Teleutospores. Sori hypophyllous, sometimes arranged in circles round the uredo-sorij rarely scattered, occasionally con- Fig. lfl-5. P. Lolii. .Ecidia on leaves of /.'. catharticus : Teleutospores, a, on Lolium, b, on Arrhenatherum. fluent, oblong or linear, ^ — 1 mm. long, nearly always covered by the epidermis, black ; spores as in P. coronata, but very irregular; mesospores also occur. .Ecidia on Rim minis catharticus, May and June: uredo- and teleutospores on Alopecurus pratensis, Arrhenatherum elatius, Avena fatua, A. pratensis, A. sativa, Festuca elatior, Glyceria aquatica, Holcus lonatus, H. mollis, Lolium perenne; not yet recorded on all these species of grass in Britain. (Fig. 195.) The Crown Rust of the Oat is most commonly found on Rye-grass, frequently also on Arrhenatherum and Holcus, and also on cultivated Oat which alone of the cereals it attacks, doing considerable damage. The teleutospores can be found on Arrkenatlierwn from the middle of August onwards and, as Plowright remarked, are "accompanied by a profuse development of uredospores "— so profuse, indeed, as to attract the notice of even non-botanical eyes. The uredo-sori form more blister- like swellings and the teleuto-sori remain longer covered by the epidermis than is the case in P. coronata. The uredospores are much brighter in colour than those of P. graminis. This species has been divided by Eriksson into a varying number of biological races, of which P. Lolii .1 venae is the most important ; see p. 68. It is found in Australia on Oat and Rye-grass (introduced with seed ?) although no species of Rhamnus is indigenous there (McAlpine). DISTRIBUTION: Europe) Asia. North America. Australia. ON GRAMINE.E 257 120. Puccinia Festucae Plowr. Mcidivm Periclgmeni Schum. Enum. PI. Sail. ii. -2-2'). Cooke, Micr. Fung. p. 196. Plowr. Ured. p. 264. JE. crassum var. Pen'cli/meni Cooke, Handb. p. 539. Puccinia Festucae Plowr. Gard. Chron. 1890, ii. 42, 139, and 1891, i. 460; Grevillea, xxi. 109. Sacc. Syll. xi. 194. Sydow, Monogr. i. 752. Fischer, Ured. Schweiz, p. 377, f. 272.- McAlpine, Rusts of Australia, p. 119, f. 13. Spermogones. In .small clusters, honey-coloured. JEcidiospores. ^Ecidia hypophyllous, on round yellow or brownish spots, in roundish clusters 2 — 5 mm. diam., shortly cylindrical, whitish-yellow, with recurved irregularly torn margin; spores delicately verruculose, orange, 16 — 27 /x. Uredospores. Sori epiphyllous, scattered, minute, oblong, yellow ; spores globose to ellipsoid, echinulate, yellow-brown. 22 — 30^, without paraphyses. Fig. 196. P. Festucae. Teleutospores and mesospore. Teleutospores. Sori hypophyllous, minute, scattered, oblong or sublinear, black-brown; spores clavate-oblong, crowned at the summit, with four to six curved and sometimes bifid processes, gently constricted, attenuated downwards, smooth, pale-brown, 40 — 60 x 15 — 23 ft; pedicels brown, persistent, 15 — 25 //, long: a few mesospores intermixed. iEcidia on leaves of Lonicera Periclymenum, June — August,, not uncommon; uredo-and teleutospores on Festuca duriuscula, F. ovina, August — October, not common or at least rarely observed. (Fig. 196.) It was Plowright who first, in 1890 (after twenty-eight unsuccessful trials), proved that the well-known secidium on Lonicera was connected. G. U. 17 258 PUCCINLA with a Crown Rust on Festuea ; Kleh.dm and Fischer have since repeatedly confirmed his results. Fischer records /'. Festucae for Festuea rubra and its \ar. fallax, and he connects it with an secidium on Lonicera coerulea and A. nigra. Be states that the uredospores have about six germ-pores ; that the teleuto-sori are situated in the groove of the upper side of the leaf, at first covered by the epidermis which at Length opens by a slit ; also that the teleutospores have a thick-walled pedicel and are much thickened and occasionally conical at the vertex : mesospores were sometimes found intermixed. McAlpine found the uredospores with five scattered germ-pores on one face (i.e. probably six or more altogether). His specimens were on Festuea ovina and F. rigida, hut in Australia no secidium on Lonicera is known. He found very commonly, mixed with the teleutospores, mesospores provided with similar processes at the apex. Distribution: Central and Northern Europe, North America, Australia. 121. Puccinia glumarum Er. et Henn. Uredo glumarum Schmidt, Allg. okon. Fl. i. 27 (1827 . Trichobasis glumarum Lev. ; Cooke, Handh. p. o2!J ; Micr. Fung. p. 2-2?,. l'urrinii linbiyo-cera Plowr. Ured. p. 167 p.p. P. straminis Cooke, Micr. Fung. p. 202 p.p. P. gliuiitinni, Erikss. et Henning, Getreideroste, p. 141. Trans. Brit. Myc. Soc. i. 5!J. Sydow, Monogr. i. 706. Fischer, Ured. Schweiz, p. 366, f. 265. Uredospores. Sori amphigenous and on the inflorescence, minute, oblong, as much as 1 mm. long, arranged in long lines, on yellow discoloured spots, not often confluent, lemon-yellow : spores globose to broadly ellipsoid, echinulate, yellow, 25 — 30 x 18 — 26 ix ; membrane always distinctly colourless. Teleutospores. Sori hypophyllous or culmicolous, arranged in long fine lines, a few scattered on the inflorescence, oblong, dark-brown or black, covered by the epidermis; spores clavate, rounded, truncate or obliquely conical above, where the exo- spore may be as much as 4 — 6 (or even 10) /x thick, slightly constricted, attenuated below, smooth, brown, SO — 70 x 12 — 24m; pedicels very short or almost none; paraphyses brown, numerous, curved, surrounding each little group of teleutospores. ON GRAMINE.E 259 On leaves, culms and glumes of Agropyron caninum, A. re pens, Brachypodium sylvaticum, Bromus mollis and other species, Elymus arenarius, Hordeum vulgare, Secede Cereule, I). ;.<•• ..*' / : a b1 Fig. 197. P. glumarum. a, uredo-sori and b, teleuto-sori, on leaves, nat. size ; c, teleuto-sori, on glume ; d, uredospore ; e, teleutospore. Triticum vulgare. It is one of the few species that attack the ears, to which it does serious damage. (Fig. 197.) This is one of two forms originally included under the collective name P. Rubigo-vera DC. ; they can best be distinguished in the uredo-stage. In the one, P. glumarum, the uredo-sori are abundant, clear lemon- or sometimes orange-yellow, and stand in long lines, often occupying half the leaf- blade ; in the other, P. dispersa and its subordinate forms, they tend towards brownish-orange or even chocolate-brown, and are scattered, rather thinly and without order, over the whole leaf-surface. Eriksson and Henning first proved in 1896 what had been long surmised, that they are quite distinct. P. glumarum has no known secidial form, and has been divided into five biological races (see p. 67). It is common in certain districts of England and is called the Spring Rust on account of its early appearance, and Yellow Rust on account of its bright colour, which varies from sulphur- to pale cadmium-yellow. The lines of the uredo-sori may be as much as 7 cm. long, chiefly on the upper face of the leaf ; they can be found all the year round on suitable leaves, and are frequently abundant on Wheat as early as the beginning of May. The teleutospores germinate as soon as they are mature ; the basidium is yellow until the basidiospores are formed, not colourless as in P. dispersa. On Hordeum murinum there is recorded a form, P. Hordei Fckl., which has smaller yellow sori, arranged less evidently in lines ; this may belong to P. glumarum but has not yet been found in Britain. It is probably not identical with P. simplex (q. v.). Distribution : Europe, Egypt, North America, Japan. 17—2 260 puccixia 122. Puccinia dispersa (sens, lat.) Er. ct Henn. Trichobasis Rubigo-vera L6v. ; Cooke, Micr. Fung. p. 222, pi. 7, f. 140- -2 , Puccinia straminis Cooke, Micr. Fung. p. 202 p.p. P. Rubigo-vera Plowr. Qred. p. Ui7 p.p. Sure. Syll. vii. 624 p.p. /'. dispersa Erikss. et Eenning, Getreideroste, p. 210 L896 . Trans. Brit. Myc. Soc. i. 58. Uredospores. Sori generally epiphyllous or a few hypo- phyllous, 1 — 2 mm. long, scattered without order, rarely con- fluent, oblong or punctiform, rust-coloured or dirty-ochre, be- coming paler; spores more or less globose, shortly echinulate, dirty-yellow or dull-orange, 16 — 28 /x ; membrane distinctly brownish (pale chocolate-umber) when mature; germ-pores 7 — 10, scattered over the whole surface. Teleutospores. Sori hypophyllous or less often on the sheaths, scattered or slightly and irregularly aggregated, rarely in distinct lines, small, oblong, covered by the epidermis, black : spores oblong to clavate, truncate, rounded, or obtusely and obliquely pointed above, slightly thickened, gently constricted, narrowed downwards, smooth, brown, 35 — 56 x 12 — 23 fx: pedi- cels short ; paraphyses numerous, brownish, more or less curved, surrounding the spores. This is a general description of the forms included under the name Brown Rust, to which the title P. dispersa was originally given. The dirty-orange colour of the uredospores, which distinguishes them at a glance from P. glumarum, is due to the fact that the membrane of the spores is brownish, not hyaline ; the spore contents are orange in colour. The germ -pores are scarcely perceptible in the immature or untreated spore, but they can be seen easily if a spore is squeezed strongly between the cover-glass and the slide, or by choosing a mature and empty spore. In the teleutospores only the upper slightly thickened wall is dark- chestnut, the rest being thin-walled and pale ; there is usually also a chestnut-brown band at the apex of the pedicel. The structures called paraphyses here in the teleuto-sori are quite different from those called by the same name in Melampsora, etc. ; they are erect, coherent, thick- walled, prismatic cells, which surround the teleuto-sori, or in the case of the larger ones divide them into loculi. There are also paraphyses of the ordinary shape, with a brownish membrane, mingled with the uredospores in certain cases, but the occurrence of these seems, so far as is known at. present, to be somewhat fortuitous. ON GRAMINE.E 261 Eriksson has divided this species into a number of forms which show- certain differences, chiefly biological : they are given in what follows, but it must be understood that they are distinguished almost entirely by their host-plants. They all show the same scattered, brownish-orange uredo- sori. Some, it is true, have recidia, others are not known to have them, but this is a difference which time may remove ; also Pole-Evans (1907) has shown that they present minute differences in the mode of germination of their uredospores. Many authors prefer to consider the biological races which follow as distinct species, but if that is done it is a mistake, which entails continual confusion, to retain the name P. dispersa for one of them. Distribution : Europe, Asia Minor, North America, and in Australia (probably only introduced). (1) PUCCINIA SECALINA 110V. nom. Puccinia dispersa (sens, strict.) Erikss. Ann. Sci. Nat. ser. 8, ix. 268, pi. xi, f. 1—6 (1899). Sydow, Monogr. i. 709. Fischer, Ured. Schweiz, p. 357, f. 261. Klebahn, "Wirtswechs. Rostp. p. 237. JEcidvwm Anchusae Erikss. et Pfenning, Getreideroste, p. 210. 2E. Asperifolii Pers. ; Cooke, Handb. p. 541 ; Micr. Fung. p. 197 p.p. JEcidiospores. iEcidia hypophyllous or often on the calyx and fruit, seated on rounded yel- lowish or reddish-yellow spots, cup-shaped, with an incised revo- lute margin ; spores verruculose, orange, 20 — 26 p.. ^Ecidia on Anckusa (Lycupsis) arvensis, very rare, Shere, Folkes- tone, Eltham, etc., August; uredo- and teleutospores on Secede Cereale, May— October. (Fig. 198.) Fig. 198. P. secalina. yEcidia on leaf of Lycopsis ; teleuto- spores, ou neighbouring stems of Rye. The Brown Rust of Rye in the uredo- and teleuto-stages seems to be confined to that cereal. Eriksson and Klebahn have proved that it can be transferred from species of Anchvsa to the Rye and vice-versd, but not, they say, to other plants. Plowright's observation (Ured. p. 168) that the secidium was produced on Anchusa by infection from a rusted bundle of wheat straw is discredited by them, but possibly without sufficient reason. The teleutospores are capable of germination as soon as they mature ; hence the lecidium is usually met with in August and September- 262 PUCCINIA el) Puccinia bromina Erikss. Puccinia bromina Erikss. I.e. p. 271, pi. xii, f. 1:2 — 17. Sydow, Monogr. i. TIi.'. McAlpine, Rusts of Australia, p. L16, pi. ( ', and fig. 28. /'. Symphyti-Bromorum F. Mull. Beihefte Bot. Centralbl. 1901, x. 201. Klebahn, I.e. p. 239. Fischer, Qred. Schweiz, p. 3">!», f. 262. /'. dispersa Erikss.; Marshall Ward, Ann. Bot. L902, xvi. -I?,?,-, Annal. Mycol. 1903, p. 132. Freeman, Ann. Bot. ibid. p. 4N7. Mcidium Symphyti and M. Pulmonariae Thiim. Spermogones. Horn \ -coloured. .Eridtusporrs. .Ecidia hypophyllous or on the petioles, sometimes even on the calyx, seated on large round or irregular -pots which are purplish-brown and surrounded by a yellow zone, cup-shaped, with an incised revolute margin ; spores verruculose, orange, 18 — '21 fx. 199. P. bromina. a, teleutospores ; b, a mesospore ; c, paraphyses surrounding the teleuto-sori ; . Fischer, Ured. Schweiz, p. 366. McAlpine, Rusts of Australia, p. 132, pi. i, f. 3, 6, 10; pi. xl, f. 302; pi. A, f. 1, 2. ON GRAMIXE.K 263 On Triticum vulgare. June — August. No secidium is known to belong to it. (Fig. 200.) The Brown Rust of Wheat has been frequently so abundant in this country in its uredo-stage as to cause great loss. The uredospores can be distinguished from those of P. graminis, when both occur upon wheat, by being subglobose, not elongate-ellipsoid, and by the more numerous germ- pores which are scattered instead of forming an equatorial band ; also they appear early in spring, before those of P. graminis. Sometimes the teleuto-sori occur on the culms, and are then arranged more or less in lines, but they are most common on the underside of the " flag " ; their spores germinate only after a winter's rest. Mesospores are not frequent in this species. Klebahn tested the basidiospores of this Rust on forty-two likely species of plants in the hope of discovering an secidium in its life-cycle, but without any result. The uredospores were found to be capable of surviving the winter by McAlpine in Australia (where it is an introduced species), and by Carleton in the United States south of lat. 40° N. Fig. 200. P. triticina. Teleutospores. (4) Puccinia holcina Erikss. Puccinia holcina Erikss. I.e. p. 274, pi. xiii, f. 22 — 5. Sydow, Monogr. i. 715. Klebahn, I.e. p. 249. Fischer, Ured. Schweiz, p. 365. On Holcus lanatus, H. mollis. June — October. Common. The uredo-stage must be carefully distinguished from that of P. coronata, which occurs on the same hosts ; the number of germ-pores at once decides the question. The uredo-sori are of a brighter colour than in the other forms of P. dispersa and stand upon conspicuous pale spots. The teleuto- spores are more rarely produced and require to be looked for closely ; they resemble those of P. triticina, but are mingled with a few mesospores. (5) Puccinia agropyrina Erikss. Puccinia agropyrina Erikss. I.e. [>. 273, pi. xii, f. 18 — 21. Klebahn, I.e. p. 249. Sydow, Monogr. i. 712. Fischer, Ured. Schweiz, p. 365. On Agropyron caninum, A. repens. August — October. (Fig. 201.) This is one of the commonest of the Rusts on wild grasses in the autumn, and is easily recognised by the following points : The small scattered dull-orange uredo-sori on the upper leaf-surface ; the round faintly echinulate uredospores, which when empty show a pale-chocolate 264 PUCCINIA membrane marked with about nine (7—10) germ-pores, which are each surrounded by ;i little thickening of the cell-wall, so that they look some- what Like a " bordered pit " ; the teleuto-sori mostly on the lower leaf-surface or sheath, black, covered by the epidermis ; the teleutospores obconical or with nearly ] .arallt'l sides, truncate, rounded, or pointed (obtusely and often obliquely) at the apex; the slightly thickened apical wall and a broad band at the base chestnut-brown, but the remainder rather pale ; each little group of teleutospores surrounded by a dense wall of brown closely coherent "paraphyses." Oftentimes there are numerous meso- spores, especially on the margin of the sorus. The clavate paraphyses which are frequently present in the uredo-sori of other forms of /'. dispersa Fig. 201. P. agropyrina. a, teleutospores; b, paraphyses with the name ; c, plan of paraphyses ; d, paraphysis with uredospores ; e, uredospore. are mostly absent in this ; when present they have a thin brownish membrane exactly of the same colour as that of the uredospores. This fungus is as distinctively an autumn parasite as P. glumarum is a spring < nic on the same hosts. (6) Puccinia Triseti Erikss. Puccinia Triseti Erikss. Ann. Sci. Nat. ser. S, ix. 277, pi. xiii, f. 26 — 9 (1899). Sydow, Monogr. i. 710. Fischer, Ured. Schweiz, p. 364. Klebahn, I.e. p. 249. On Trisetum fiavescens. Uncommon; Alvechnrch, Hereford, etc. June — October. P. graminis occurs on the same host. (7) Puccinia simplex Er. et Henn. Puccinia simplex E. et H. (ietreicleroste, p. 258 (189G). Sydovr, Monogr. i. 756. Klebahn, I.e. p. 24S. Fischer, Ured. Schweiz, p. 368, f. 266. McAlpine, Rusts of Australia, p. 130, pi. i, f. 1, 4, 9 and pi. B, f. 9, 10 (introduced in 1902). P. Rubigo-vera var. simplex Kern. ; Plowr. Ured. p. 168. ON GRAMINE.*: 265 On leaves and culms of Hordeum vulgare and other species of Hordeum. Teleutospores, August, September. (Fig. 202.) The Dwarf Brown Rust of Barley, distinguished by the fact that it hears few two-celled teleutospores, but very numerous mesospores, which are variable and asymmetrical, slightly thickened at the apex (4 — 6/x), measuring 25 — 45 x 16 — lip. It is to be found in the uredo-stage all the year round. The teleutospores germinate in spring ; Klebahn tried to infect, with their basidiospores, the same forty-two species which he tested with P. tritivina, but equally in vain. P. simplex. Teleutospores. P. simplex may be worthy of being regarded as a distinct species ; it presents a little more difference from the other forms of P. dispersa than they do from one another. The sori of both kinds are amphigenous and more minute and punctiform (except on the culms), and the uredospores are of a brighter yellow. On Hordeum distichum I have found sori of P. simplex on the leaves, and with them those of P. graminis on the culms. It will be noticed that all the last five races are without any known secidium ; it follows, apparently, that they must maintain themselves by their uredospores, but one might venture to suggest that future, unexpected, discoveries will throw light upon this obscure matter. Eriksson and Klebahn have both proved, by numerous infection experiments, that these races or subspecies of P. dispersa are .all biologically distinct; with few, and doubtful, exceptions none of them can be transferred from its own to the other hosts. There are other forms of Brown Rust of which little is known. The aecidium on Echium vtdgare mentioned by Plow- right (Ured. p. 168) may belong to one of these ; this plant was one of the forty-two previously mentioned, tested by Klebahn. I have also some specimens of uredospores on Aira fiexuosa and A. caespitosa sent by Mr T. B. Roe from Scarborough: those 266 PU( I l\l \ on A. fleccuosa have no paraphyses, as those od A. caespitosa have, but otherwise they are almosl identical and arc undoubt- edly P. dispersa (sens. lat.). I tiixl similar spores on A. caes- pitosa round Birmingham, and have a specimen on the same host sent by the late H. T. Soppitt from Saltaire; this latter is the plant referred to in a note to P. I'm mi (I'lowr. CJred. ]>. 192) — both these have abundanl paraphyses with the uredo- spores. Herr H. Sydow informs me that he considers the presence or absence of these paraphyses to be a character of little importance in P. dispersa. It must be remembered that P. graminis also grows on ^1. <-itosu, but the uredospores cm be easily distinguished by their elliptical shape and three or tour subequatorial germ-pores; their membrane is brownish, but without the chocolate tinge of P. dispersa. I "2-'l. Puccinia sessilis Schneid. Puccinia sessilis Schneid. in Schrot. Brandpilze Schles. p. 19. P. linearis Rob. ; Cooke, Micr. Fung. p. 203. ./„'<■ i'/iospores. See the descriptions given for the four specialised biological races. Uredospores. Sori amphigenous, scattered, minute, puncti- form or shortly linear, yellow: spores globose to ellipsoid, echinulate, brownish-yellow, 20 —28 x 18 — 24 /J,: epispore thin, with about seven germ-pores. Teleutospores. Sori similar, sometimes confluent, long cov- ered by the epidermis, pul- vinate, black ; spores oblong or oblong-clavate, rounded or truncate above where they are darker and slightly thickened (2 — 5 ft), hardly constricted, somewhat narrowed below, smooth, brown, 35 — 52 x 15 — 22 /j.; pedicels very short or absent ; an occasional mesospore is found. JEcidia on various species of Monocotyledons ; uredo- and teleutospores on Pkalaris (Digraphis) arundinacea, uncommon, July— May. (Fig. 203.) Fig. 203. P. sessilis. Teleutospores and mesospore. ON GKAMINE.K 267 The four following biological races agree exactly in the teleutospores, and these can only be distinguished by their successful use to infect the alternate host ; though sometimes the question may presumably be decided by finding one or more of those hosts, in the immediate neigh- bourhood, affected by the secidium. It is evident from the disagreement between various authors that it is impossible to decide to which of the four biological races the name P. sessilis Schneid. should be applied : it will be better, therefore, to use it as a collective title, which can be employed in cases where the oecidial host cannot be determined. A fifth race, P. Schmidtiana Dietel, having its recidia on Leucojum, has not yet been found in Britain. Distribution: Europe and North America. (1) Puccini a Digraphidis Soppitt. JEcidium Convallariae Schum. Enum. PI. Sail. ii. 224. Plowr. Ured. p. 26-1. Soppitt in Gard. Chron. ser. 3, vii. 643. Puccinia Digraphidis Soppitt in Journ. Bot. 1890, p. 213. P. intermixta Friend in Gard. Chron. ser. 3, viii. 270 p.p. P. Paridis Plowr. in Gard. Chron. 1892, p. 137 ; Journ. Linn. Soc. 1893, p. 43. /'. Sinilacearum-Digraphidis Kleb. Zeitschr. f. Pflanzenk. 1896, p. 261. Fischer, Ured. Schweiz, p. 340, f. 251. P. sessilis Sydow, Monogr. i. 781. Spermogones. Epiphyllous or in the midst of the oecidia. uEcidiospores. /Ecidia hypophyllous, loosely clustered on roundish or irregular yellow spots, cup-shaped, with a cut white revolute margin; spores verruculose, yellowish, 19 — 27 fx. /Ecidia on Convallaria majalis, Paris quadrifolia. Not common. May and June. Attempts have been made to subdivide still further the fungi included under this head. P. Digraphidis Soppitt, on Convallaria, and P. Paridis Plowr., on Paris, are two of these forms which to their authors appeared under cultivation to be confined to their respective secidial hosts. But, on the other hand, Klebahn lias been able to infect, from one and the same Puccinia, both Convallaria, Maianthenmm, Paris, and Polygonatum ; nevertheless his attempts to induce specialisation, by cultivating the fungus year after year on Polygonatum alone, had the result that towards the end (while it still grew freely on that genus) it could be transferred only with difficulty or not at all to the other genera. Evidently we have here a case where specialisation is naturally in progress, but has not yet proceeded far enough to effect complete separation. 268 puccinia (2) I'rci'ixiA Orchidearum-Phalaridis Kleb. JEcidium Orckidearum Desm. Cat. Plant, omis. p. 26. Cooke, Baudb. p. 545; Micr. Fung. p. 200. Plowr. [Jred. p. L79. Puccinia Orchidearum-Phalaridis Kleb. Zeitschr. f. Pflanzenkr. 1899, p. L55. Sydow, Monogr. i. 7S2. Fischer, Ured. Schweiz, p. 343. Speriiioiiuites. Epiphyllous or in the midst of the group- <>f secidia. ^h'risj)ares. Sori numerous, similar, bul larger and thicker, very convex and compact, black : spores oblong, rounded ,-it both ends, thickened (4 — 9/a) above, constricted, smooth, ( |. -i| ) vclluw ish-brown, 45 — 65x16 — 25 /u, sometimes 75 /j, lung; pedicels hyaline or yellowish, thick, persistent, 100 — 200 /x long. .Kcidia on Rumex acutus, R. crispus, R. conglomeratus, R. Hydrolapathum, Ji. obtusifolins. Rheum officinale, -May ami .lane: uredo- and tclcutospores on Phragmites communis. July — May, often in the same sori. Xot comniun, except locally. (Fig. 208.) It was Plowright who first showed that P. Phragmitis has its secidium, not on Ranunculus, but on species of Rumex and lllo-um (not, however, on Rumex Acetosa). Klebahn and Fischer have confirmed his results, and Arthur lias done the same for the Xorth-American forms. It is a remarkable fact, however, that the reeidium had not been previously found in North America until Arthur obtained it artificially by infection of Rumex crispus and R. obtusifolius with the teleutospores of P. Phrag- mitis. Afterwards it was found in Nebraska on various species of Rumex and Rheum. It is suggested by Sydow that this species is dispensing with the seeidium, in which case it must winter by means of its uredo- s] lores. Even in England the secidium seems relatively scarce, but it is very conspicuous, and can be found on Rumex growing amidst Reeds. Distribution: Europe, South Africa, Japan, North America, Chili. 128. Puccinia Trailii Plowr. Mcidium rubellum Gmel. ; Cooke, Handb. p. .">44 ; Micr. Fung, p. 199 p.p. • Puccini" Trailii Plowr. Ured. p. 176. Sacc. Syll. ix. 312. Sydow, Monogr. i. 790. Fischer, Ured. Schweiz, p. -I'rl, f. 193. jEcidiospores. Similar to those of the last species ; but the spots are purple, surrounded by a yellow margin, the a?cidia are wider and flatter, and the spores are on the average somewhat larger. Uredospores. Sori amphigenous, scattered, rather large, ON GRAMINE.*: 275 elliptical or linear, reddish-brown, pulverulent, without para- physes; spores subglobose or ovate, echinulate, brownish, 25 — 35 x 20—25^. Teleutospores. Sori similar, but larger (2 — 4 mm.), compact, pulvinate, black: spores oblong, rounded at both ends, with a cap- like thickening (5 — 10 //,) above, plainly constricted, brown, 50 — 60 x 20 — 23 fx; pedicels brownish, thick, persistent, 75 — 100 p long. ^Ecidia on Rumex Acetosa, May and June; uredo- and teleu- tospores on Phragmites communis, from July. Rare. (Fig. 209.) The results of Plowright's cultures have been confirmed by Klebahn, but at any rate this species is very closely allied to the previous one, and should rather be considered as merely a biological race of it. Distribution : Holland, Germany. 129. Puccinia Agrostidis Plowr. JEcidium Aquilegiae Pers. ; Plowr. Ured. p. 263. jE. Ranunculacearum var. Aquilegiae DC. ; Cooke, Handb. p. 539. P. Agrostidis Plowr. Card. Chronicle, 1890, ii. 139 ; Grevillea, xxi. 110. Sacc. Syll. xi. 202. Sydow, Monogr. i. 717. McAlpine, Busts of Australia, p. 114, f. 27(0- Fischer, Ured. Schweiz, p. 353, f. 257. Spermogones. Honey-coloured, on round spots. sEcidiospores. iEcidia on rather large roundish yellow spots which are often thickened and margined with brown, crowded, hypophyllous, shortly cylindrical, with a torn white margin; spores faintly verruculose, orange, 16 — 30 x 14 — 20 /x. Uredospores. Sori amphigenous, on yellow spots, elongated or linear, about 1 mm. long, bright-orange ; spores globose to IS— 2 Fig. 209. P. Trailii. Teleuto- spores, and sori on leaf of Phrag- mites. 276 puccinia ovate, yelloWj faintly echinulate, 20 — 25x16 — 22 fi: membrane colourless. Teleutospores. Sori hypophyllous, minute, covered by the epidermis, oblong or linear, rarely confluent, black; spore9 somewhal clavate, rounded, truncate or gently nan -owed above, slightly thickened (about 5 /x), faintly constricted, narrowed below, smooth, brown, darker upwards, 38 — 48x12 — 20/j,: pedicels very short. . lv-idia on Aquilegia vulgaris, Lewes, Sussex ; Lake Win- dermere; Wyre Forest, etc., May, June; uredo- and teleuto- spores on leaves and sheaths of Agrostis alba, A. vulgaris, August. Uncommon. The Puccinia should be looked for on the grass near the place where the secidium was seen. The connection between the aecidium <>n Aquilegia and the Puccinia on Agrostis was first demonstrated by Plowright (see Gard. Chron. 1890, ii. 139, and 1891, i. 683) ; the fact has since been confirmed ami extended to Aquilegia alpina. McAlpine's species seems to be rather different ; he records numerous mesospores, and uredospores with as many as nine -•in i -pi ires, circularly arranged, on one face. Distribution: Central and Western Europe, Siberia, India, ainl ( '.) Australia. 180. Puccinia Ivlcliniae Tul. dEcidium Melampyri K. et S. exsicc. no. 165. Puccinia Moliniae Tul. Ann. Sri. Nat. ser. 4, ii. 141, pi. ix, f. 9 — 11. Cooke, Grevillea, v. 57 ; Micr. Fung. p. 203. Plowr. Ured. p. 179. San-. Syll. vii. 631. Sydow, Monogr. i. 762. Fischer, Ured. Schweiz, p. 256, f. 195. P. nemoralis Juel, Ofvers. K. Vet.-Akad. Forh. 1894, p. 506, f. a—f. P. .E'idi't-MiLi uipyri Liro, Act. Soc. Faun, et Flor. Fenn. xxix. 55. [JEcidiospores. iEcidia hypophyllous, clustered on roundish red or purple spots 3 — 5 mm. diam., cup-shaped, with a cut white revolute margin ; spores very minutely verruculose, yellowish (?), 15—18/*.] Uredospores. Sori amphigenous, generally hypophyllous, often on brownish or purplish spots, scattered or arranged in lines and confluent, oblong or linear, brown; spores more or less ON GRAM1NE.E 277 globose, aculeate, yellow-brown, 20 — 28x20 — 24 /j, ; epispore 3 — 6 fi thick, with three germ-pores. Teleutospores. Sori similar, often confluent and as much as 8 mm. long, conspicuous, pulvi- nate, black ; spores ellipsoid, rounded at both ends, slightly thickened (up to 5 p) above, hardly constricted, smooth, brown, 32—46x20—30^; pedicels hya- line or yellowish, curved, per- sistent, rather thin, very long (as much as 120/i); a few mesospores sometimes intermixed. [JEcidia on Alelampyrum spp.J: uredo- and teleutospores on Molinia coerulea, July — Oc- tober, Perthshire (Dr B. White). This secidium is not recorded for Britain, and appears to be very rare everywhere. (Fig. 210.) Plowright, relying upon the experi- ments of Rostrup, connected the secidium on Orchis latifolia with this Puccinia, though he himself could not succeed in the infection. Others have similarly failed, and there seems to be little doubt that Rostrup's conclusions were inaccurate. Juel has since then succeeded in showing that an Becidium on Alelampyrum pratense is part of the life-cycle of a, Puccinia on Molinia, which he named P. nemoralis, but of which there is no proof that it is different from P. Moliniae Tul. (Juel, I.e.). Liro confirms this result and names his species P. JEcidii-Melampyri (I.e.). The recidiospores are described by Sydow as "yellowish," but Juel describes them as colourless, like those of the allied species P. Phragmitis and P. Trail ii. Since the eecidium on Alelampyrum has not been found in this country. the British species may turn out, on investigation, to be different from these. For there is a closely allied species or biological race, named by Cruchet (Centralbl. f. Bakter. 2. xiii. 9(5) P. Brune/larum-Aloli/tiae, which has teleutospores very like those of P. Moliniae Tul. but its secidium on Fig. 210. spores, White). P. Moliniae, Rannoch (Dr Teleuto- Buchanan 278 PUCCINIA Brunella = /Ectdium Prunellae Wint. . This was found by Fischer in Switzerland, and the connection of the two hosts was proved by Cruchet. To this, doubtless, belongs the secidium found on Prunella vulgaris by Dr Keith at Forres Plowr. Ored. p. 264), and Dr Buchanan White's Puccinia on Molinia may belong there likewise. The apes of the teleuto- spores in his specimens is, however, much less strongly thickened than in the figures given by Fischer. DISTRIBUTION : Throughout Europe, exccpl the south. 131. Puccinia Poarum Niels. .K< -iii 'i'i in Tussilaginis Gmel. in Linn. Syst. Nat. ii. 1473. .K. Compositarum var. Tussilaginis Cooke, Handb. p. 542; Micr. Fung. p. 198. Pucemia Poarum Niels, in Botan. Tidsskr. ii. 26. Phill. et Plow. Grevillea, xiii. 54. Plowr. [Jred. p. 168; Grevillea, xi. 52. S Syll. vii. 625. Sydow, Monogr. i. 795. Fischer, Ural. Schweiz, pp. 361, 556, f. 263. McAlpine, Rusts of Australia, p. 128, f. 22. Spermogones. Epiphyllous, pale-yellow, often very numer- ous. .Ecidiospores. JEcidia hypophyllous, usually in dense clusters on circular yellowish or reddish thickened spots 1 — 2 cm. diam., seldom scattered, cup-shaped, with a dentate white revolute margin ; spores verruculose, orange, 18 — 25 x 16—20/*. Uredospores. Sori on the leaves, sometimes on the culms, minute, roundish or elliptic, soon naked, yellow ; spores globose to ellipsoid, densely and minutely verruculose, yellow, 17 — 28 x 17 — 25 /a. with about five scarcely perceptible scattered germ- pores, and intermixed with numerous, hyaline, capitate para- physes. Teleutospores. Sori similar, oblong or linear, more or less in short rows, long covered by the epidermis, surrounded by a small pale area, black : spores oblong-clavate, cylindrical, or obconical, variable, rounded, truncate, or rarely conically at- tenuated above where they are slightly thickened (4 — 8 /n), hardly or not at all constricted, more or less tapering below, smooth, chestnut-brown, becoming gradually paler downwards ON GRAMINE.E 279 30—45 x 16 — 22 /z ; pedicels short, brownish, persistent ; an occasional mesospore is found. Fig. 211. P. Poarum. Teleutospores ; a, uredospore on P. nemoralis; h, para- physes with same; c, mesospore; rf, teleuto-sori on P. pratensis; e, typical teleuto-sori of Uromyces Poae, on the same. /Ecidia on Tussilago Farfara, about May, June, and August, September, very common ; uredo- and teleutospores on Poa annua, P. nemoralis, P. pratensis, P. trivialis, about July, August and October — December, common but easily overlooked unless searched for. (Fig. 211.) First stated by Nielsen, the connection of the two hosts was demon- strated by Plowright and Klebahn. This hetercecious Puccinia differs from all others in having two generations in one year. The earlier crop of secidia begins to appear in May, and is followed by the uredo- and teleutospores on the surrounding leaves of Poa ; these germinate quickly and the second crop of secidia is produced about August, and the second generation of teleutospores may be found on Poa from October. The latter germinate in the following spring, but according to Lagerheim the uredospores also, in a favourable climate, can survive the winter. This is certainly true in Australia, where (though the Puccinia is an introduced one) the uredospores have been found the whole year round. In that country the Coltsfoot does not exist, and the fungus is carried through the winter by the uredo-stage ; in fact, according to McAlpine, it is most common in the winter there. Arthur and Carleton say that the fungus does the same as far north as Nebraska in North America, where the Coltsfoot is only a naturalised plant. Uredospores were found alive in every month of the year at Washington, D.C. In the Scottish Naturalist ('84, p. 270) this species is recorded for Poa tin it, i i3:> ; Micr. Fung. p. 222. I'ni'cinia Baryi Wint. Pilz. Deutsch. p. 178. Plowr. Ured. p. 191. Sacc. Syll. vii. 660. Sydow, Monogr. i. 737. Fischer, Ured. Schweiz, p. 369, f. 267. Uredospores. Sori mostly epiphyllous, on linear brown spots. scattered or in groups, often disposed in long linear series, minute, elongated, reddish-brown; spores globose to obovate, delicately verruculose, yellow, 18 — 25 /x ; paraphyses numerous, clavate to capitate. Fig. 212. P. Baryi. Teleutospores; a, abnormal teleutospore ; b, paraphysis; c, uredospore ; all on />'. silvaticum. Teleutospores. Sori similar, but long covered by the epi- dermis, blackish-brown: spores very irregular, ellipsoid or sub- clavate or pyriform, obtuse or truncate and slightly thicken'' I and undulated above, hardly constricted, somewhat attenuated below, smooth, clear-brown, darker above, 25 — 40 x 15 — 25 /x; pedicels none or very short, brownish, darker at base of spore where the transverse wall is much thickened. ON GRAMINE.E 281 On leaves (living or fading) of Bracliypodium pinnatum, B. silvaticum. Not uncommon. July — November, teleuto- spores not before September. (Fig. 212.) Both Plowright and Fischer mention that, mixed with the uredospores, are numerous hyaline capitate paraphyses ; Sydows' Monographia omits all mention of these. The specimens which I have seen show them always in great numbers. Often the pedicel of the teleutospores is almost non-existent, and the basal cell-wall is strongly thickeued. The upper margin of the teleuto- spore is often undulated ; occasionally one is met with having three cells. An secidium, though often suggested, has not yet been discovered for this species. Distribution : Central and North-Western Europe. 133. Puccinia Agropyri Ell. et Ev. jEcidium Clematidis DC. Flor. fr. ii. 243. Plowr. Ured. p. 265. JE. Ranunculacearum var. Clematidis Cooke, Handb. p. 539. Puccinia Agropyri Ell. et Ev. in Journ. Mycol. vii. 131. Sacc. Syll. xi. 201. Sydow, Monogr. i. 823. Fischer, Ured. Schweiz, pp. 350, 555, f. 255. McAlpine, Rusts of Australia, p. 113. Spermogones. Amphigenous. jEcidiospores. iEcidia hypophyllous and on the petioles and stems, usually on brownish spots, causing considerable distortion, scattered or in clusters of very varied size, shortly cylindrical, with white torn broadly re volute margin; spores verruculose, orange, 18 — 27 fi. [Uredospores. Sori amphi- genous but generally hypophyl- lous, on irregular yellow spots, scattered, oblong or linear, 1 — 1^ mm. long, cinnamon ; spores more or less globose, delicately echinulate, pallid-yellow, 19 — 27 /j, ; epispore rather thin, with three or four germ -pores. Teleutospores. Sori epiphyllous, scattered, sometimes con- fluent, oblong or linear, as much as 3 mm. long, covered always by the lead-coloured epidermis, black ; spores cylindric-clavate, Fig. 213. P. Ayropyri. Teleuto- spores and luesospore, on .4. repens, Pomerania (ex herb. Sydow). 282 puccinia usually truncate above and thickened (up to 6fi), gently con- stricted, attenuated downwards, smooth, pale-brown, darker at the apex, 40 so, ll -22/*; pedicels very short, persistent, somewhat hyaline, with a dark-brown band at the apex.] .Kridi.-, on Clematis Vitalba, May— July: [uredo- and teleuto- spores on species of Triticum (Agropyron), but not recorded as ye1 for Britain]. (Fig. 213.) Dietel first showed, by culture experiments, that tlic secidium on Clematis Vitalba is connected with a Puccinia on Triticum glaucum. In Europe the Eecidium is frequently observed (probably because of the striking distortions it produces), but the Puccinia has been far more rarely noted, being by no means conspicuous in its appearance. According to Sydow, it occurs also on Triticum junceum and T. repens, on which it may presumably be found in this country, if looked for. The statement of Rathay, that JEcidium Clemati5f>, f. 2.">4. Spermogones. Epiphyllous, in little clusters, orange. jEcidiospores. ^Ecidia hypophyllous, in clusters on thick- ened spots which are purple-brown above and yellow with a brownish border below, urn -shaped or subcylindrical, yellow, with a torn white margin ; spores minutely verruculose, orange, U_28^. ON GRAMINE.E 283 Fig. 214. P. persistens. Teleutospores. Uredospores. Sori minute,, roundish or elongated, orange, on yellowish spots; spores more or less globose, minutely echinulate, yellow, 25—30 p,. Teleutospores. Sori hypophyl- lous, minute, ovate, oblong or linear, black, long covered by the epidermis ; spores clavate-oblong or irregular, rounded, truncate or obliquely attenuate above, slightly thickened (4 — 7 /jl), more or less constricted, rounded or often tapering below, smooth, brown, 50 — 60 x 15 — 20 //. ; pedicels short, hyaline, persistent. /Ecidia on Thalictrum flavum, T. minus, May — July; uredo- and teleutospores on Agropyron repens, from July onwards. (Fig. 214.) The connection of these spore-forms was ascertained by Plowright and confirmed by Rostrup. The secidia on various other species of Thalictrum may belong to the same Puccinia or to Puccinias on Agropyron caninum, Arrhenatlierum elatius or even on species of Poa. There is abundant scope here for further investigation. According to Fischer the teleuto-sori are subdivided into compartments by groups of paraphyses, in the same way as in P. Agropyri. In fact the three species, P. persistens, P. Agropyri, and P. Actaeae- Agropyri Fischer, form a natural group which should be regarded rather as biological races of one species, having their teleutospores on Triticum {Agropyron) and their pecidia on Thalictrwm, Clematis and Aetaea respectively ; the latter (P. Actaeae- Agropyri) has not yet been found in P>ritain. Distribution : Europe, Siberia, Japan, Himalaya, North America. The evidence for considering the three following species as British is not yet sufficient : — 135. Puccinia Phlei-pratensis Erikss. et Henn. Puccinia Phlei-pratensis Erikss. et Henn., in Zeitschr. fiir PHanzenkr. 1894, p. 140 ; Getreideroste, pi. 5, f. 55—6. Sacc. Syll. si. 204. Sydow, Monogr. i. 784. Fischer, Ured. Schweiz, p. 260. 284 puccinia / redospores. Sori on the leaves or most often on the culms, on the latter and on the sheaths continent in long lines (as much as 3 cm. long) and splitting the epidermis, on the leaves more often scattered, minute and oblong, 1 — 2 mm. long, pulverulent, dart yellowish-brown; spores ellipsoid to oblong or even somewhat pear-shaped, aculeate, dirty-yellow, 18 — 30 x 15 20 p. Teleutospores. Sori similar, but chiefly on the culms, black- brown ; spores clavate, rounded or conically attenuate above, thickened (up to 8/x), constricted, attenuated below, smooth. brown, 38 — 52 x 14 — 20 fi; pedicels brownish, persistent, rather thick, as much as 60 /x long. On leaves and culms of Phleum pratense (and P. nodosum)-, also on Festuca elatior. This was separated by the authors from P. graminis on the ground that it will not infect Berberis ; Eriksson has shewn that it will feebly infect Phleum Michelii, Secale cereale, and Avena sativa, but not Triticum vulgare, Hordeum vulgare or Poa pratensis. It chiefly occurs in the uredo- stage, and winters thereby or by its mycelium ; in many localities the teleutospores are not formed at all. This may be one of the cases where the secidium has been dropped completely from the life-cycle. I have not seen a British specimen, but it is extremely likely to occur here. Distribution: Germany, Austria, Denmark. Sweden. 130. Puccinia Arrhenatheri Erikss. jEcidium graveolens Shuttl. apud Cooke, in Bull. Sec. Bot. Fr. 1877, xxiv. 315. Sacc. Syll. vii. 77s. Puccinia perplexans Plowr. f. Arrhenatheri Kleb. in Abhandl. Natur. Ver. Bremen, 1893, p. 366. P. Arrhenatheri Erikss. in (Vim's Beitrage zurBiologie d. Pflanz. 1898, viii. 1. Sydow, Monogr. i. 739. Fischer, Ured. Schweiz, p. 345. [Spermogones. Very numerous, minute, covering a great part of the leaf or the whole leaf uniformly. JEcidiospores. zEcidia hypophyllous, sometimes even on the flowers, deforming the affected branches, generally distri- buted densely and evenly over the whole leaf, between cylin- drical and cup-shaped, with a torn whitish revolute margin ; ON GRAMIXE.K 285 spores subglobose or ellipsoid, verruculose, yellowish, 19 — 32 x 16—24^.] Uredospores. Sori epiphyllous, very rarely hypophyllous, on minute yellow spots, elliptical or oblong, minute, pale rust- coloured; spores globose to ovate, echinulate, yellowish, 19 — 30 x 1!) — 2(3 ft, with numerous germ-pores, and mixed with nume- rous capitate paraphyses which are 10 — 14 fi broad and as much as 80 fx long. Fig. 215. P. Arrhenatheri. a, teleutosporea on Arrhenatherum, from Hampton- in-Arden ; b, another, and c, two abnormal ones, from Lichfield ; d, a paraphysis and uredospore from the latter. Teleatospores. Sori hypophyllous, scattered, minute form or shortly linear, covered by the epidermis, black; spores ellipsoid -oblong or oblong- clavate, rounded, truncate or rarely gently attenuated a- bove where they are thickened (5 — lOyu.) and darker, hardly or not at all constricted, gen- erally attenuated downwards, smooth, pale-brown, 30 — 45 x 18 — 24/x, mixed with brown- ish paraphyses; pedicels short, brownish. puncti- Fig. 216. P. Arrhenatheri. spores, paraphysis and ur on Arrhenatherum, Sydow). Silesia Teleuto- edospore,. (ex herb. [/Ecidia on leaves and flowers of Berberis vulgaris ;] uredo- and teleutospores on Arrhenatherum elatius. (Figs. 215, 216.) This .species produces on the Berberis its characteristic witches' -brooms, composed of many upright branches ; it was at first mistakenly identified with jE. Magelkaenicum Berk. The secidium has not been seen in Britain. 286 PUCCINIA but 1 have found on Arrhenatkerum on many occasions teleutospores and uredospores which seem to be identical with those of this species, though the former are, in my specimens, often irregularly three- or four-celled. Distribution: Central and Northern Europe, Turkestan. loT. Puccinia paliformis Fckl. Puccinia paliformis FckL Symb. Myc. p. 59, pi. ii. f. 17. Plowr. [Jred. p. 203. Sacc. Syll. vii. 731. Sydow, Monogr. i. 759, f. 534. Fischer, Lied. Schweiz, p. 264, f. 200. Teleutospores. Sori on the leaves and culms, scattered, minute, roundish or oblong, up t<> I mm. long, pulvinate, surrounded by the cleft epidermis^ black- brown; spores clavate, usually t runcate above, more rarely round- ed or conically attenuate, much thickened (10 --16 ft), hardly constricted, tapering downwards, smooth, pale-brown. 40 — 56 x 10— 22 fi: pedicels hyaline, persistent, about as long as the spore. ery rare. September and October, (Fig. 217.) There is much doubt about this fungus ; it was suspected by Winter, on account of the likeness of its teleutospores to those of P. Caricis, that the original specimens on which the species was founded grew not on Koeleria, but on Care.r. It seems to have been recorded only twice, once by Morthier in the Jura, in spring on old leaves of the previous year, and once as above. The three figures quoted in the synonymy agree fairly well, but appear to have been all taken from the same material, viz. that gathered by Morthier, which according to Sydow may well be Koeleria and _jaot Carex. I have examined the specimens, preserved in Herb. Kew, gathered by Prof. Trail ; they are undoubtedly Koeleria, they have split .sheaths and, though not in flower, agree perfectly with specimens of Koeleria cristata from the Highlands, in the same herbarium. But they bear uredospores in slightly swollen epiphyllous yellow patches, sunken in the leaf (these are mentioned by Prof. Trail in the Scottish Naturalist, 1883, p. 85) ; therefore the fungus cannot be P. paliformis as described by Fuckel, but is probably P. longissima, Schrot. I found the uredospores oval or obovate, pale brownish-yellow, sparsely echinulate, 23 — 24x17 — 18 /x. Fig. 217. P. paliformis $). Uredo- spore and teleutospores, from the specimens collected by Trail. On Koeleria cristata. V near Aberdeen (Prof. Trail). TRIPHRAGMIUM 287 Tt should be noted that there is an secidium belonging to P. longissima, on species of Sedum, formerly called by mistake Endophyllum Sedi. This has been found on S. acre, and according to Mayor (Annal. Mycol 1911, ix. 341) on S. reflexum, but not, so far as I know, as yet in Britain. Further search in the Highlands will no doubt readily decide which of the two species occurs there. TRIPHRAGMIUM Link Autcecious. Spermogones subcuticular, flattish, without ostiolar fila- ments. Cseomata indefinite, large, without paraphyses ; uredo- sori small, definite, encircled by paraphyses ; spores in both borne singly on pedicels: pores not evident. Teleuto-sori more or less definite ; spores coloured, radiately 3-celled, more or less verrucose, with one pore in each cell, equidistant from the septa, i.e. apical. It is most likely that the foreign species, usually classed with ours on account of the form of the teleutospores, are not closely allied. Spores of this particular shape are met with, abnormally, in other genera, as in several species of Puccinia, even in Puccinia graminis. Neither is this genus closely allied to Phragmidium ; the teleutospores have no gelatinous outer coat, and the germ-pores are differently placed. Triphragmium Ulmariae Wint. Uredo Ulmariae Schum. Enuni. PL Sail. ii. 227. Uromyces Ulmariae Lev. ; Cooke, Micr. Fung. p. 212, pi. 7, f. 147 — 8. Triphragmium Ulmariae Wint. Pilze, p. 225. Cooke, Handb. \>. 492 ; Micr. Fung. p. 202, pi. 3, f. 47—9. Plowr. Ured. p. 218, pi. 4, f. (i. Sacc. Syll. vii. 768. Fischer, Ured. Schweiz, p. 423, f. 293. Arthur, N. Amer. Fl. vii. 178. Sydow, Monogr. iii. 171. T. Filipendulae Pass. Nuov. Giorn. Bot. Ital. vii. 255. Cooke, Grevillea, xi. 15. Plowr. Ured. p. 219. Massee, in Grcvillea, xxi. 115. Fischer, Ured. Schweiz, p. 425, f. 294. Sacc. Syll vii. 769. Sydow, Monogr. iii. 174. 288 TRIPHRAGMIl M Spermogones. On the leaves and petioles, circulate, flat, yellowish. Uredospores. Sori of two kinds 'primary, i.e. caeomata, amphigrimiis, large, expanded, bright-orange, mostly on the veins and petioles where they cause distortion, without para- physes; secondary, hypophyllous, small, round, scattered, orange, surrounded by paraphyses; spores brilliant-orangt . ellipsoid to obovate, verrucose, 25 — 28x18 — 21 fi, without evident germ- pores. Fig. 218. T. Ulmariae. Normal telentospore ; a and b, two abnormal ones ; on S. Ulmaria. Teleutospores. Sori hypophyllous, small, round, brownish- black, persistent, but pulverulent, sometimes arising in the primary uredo-sori; spores subglobose, flattened, chestnut- brown, more or less rough with obtuse warts, 35 — 49 \i ; each cell has, at a point opposite to the inner corner, a germ-pore round which the warts are often crowded; pedicels colourless, persistent., variable in length: abnormal spores may have two or four to five cells. ( )n Spiraea Ulmaria, S. Filipendula. Very common on the ft inner host. Primary uredospores, May — July ; teleutospores, August — November. (Fig. 218.) The primary uredo-sorus may be looked upon as a cseoma, i.e. an recidium, and in any case it corresponds to that developmental stage ; Klebahn proved (Zeitschr. f. Pflanzenkr. 1907) that it is produced by infection by the basidiospores. Dietel observed that, in elevated situations, the secondary uredo-spore generation on S. Ulmaria was almost absent, and the teleutospores arose in connection with the primary uredo-sori ; this is in agreement with the usual shortening of the life-history that PHRAGMIDIUM 289 takes place in arctic and alpine Uredinales. As Plowright remarks, the teleutospores can be found in spring on last year's leaves, and germinate readily when placed in water. The form on S. Filipenchda, which is very uncommon in this country, differs from that on S. Ulmaria in having the teleutospores usually smooth, and only occasionally provided with a few warts round the germ-pores; hut similar spores may be found on both hosts. In both, some of the spores have three super-imposed cells, as in Phragmidium, also two cells placed as in Puccinia or laterally as in Diorchidium. The uredospores on S. Filipendula are often pyriform and as much as 35 n long ; it may be a distinct species. Arthur, who records T. Uhiutriae in the North American Flora on Filipendida rubra, says that it is confined to one locality " of less than a half hectare extent," at Lafayette, Indiana. Distribution : Europe, Siberia, Japan, Indiana. PHRAGMIDIUM Link. Autcecious. Spermogones subcuticular, conical or flattened, without ostiolar filaments. Caeomata indefinite, usually encircled by- incurved paraphyses ; spores in chains, each with numerous germ-pores. Uredo-sori definite, usually encircled by para- physes; uredospores borne singly on pedicels, often with paraphyses intermixed, pores numerous, scattered, indistinct. Teleutospores two- to several-celled by transverse septa ; wall thick, laminate, usually coarsely verrucose, the middle layer dark and rigid; pores two or more in each cell, placed laterally; pedicels often swollen below ; basidiospores globose. This genus is confined entirely to the family Rosacese. There are many species in North America, but with the exception of P. Poteiitillae and those introduced on cultivated r< >ses they are all different from those of Europe. The warts often found on the outer surface of the teleutospores are due to the contraction of the external gelatinous layer, which can be caused to swell up enormously by boiling in lactic acid. This G. u. 19 290 I'llltACMIDIC.M character, as well as the rigid middle lamina and the position nf the germ-pores, distinguishes the genus from all the neigh- bouring <>nes. 1. Phragmidium Fragariastri Schrot. /'t/,;'i,,/'4. Lecythea gyrosa Berk. ; Cooke, Micr. Fung. p. 222, pi. 8, f. 162 — 4. rhniiiiiu'iliiim gracile Cooke, Handli. p. 4!H ; Micr. Fung. ]>. 201, pi. 3, f. 42, 43 ; Grevillea, iii. 171, pL 45, f. 9. /'. Rubi-Idaei Karst. Myc. Fenn. iv. f>2. I'lowr. Ured. p. 226. Fischer, Ored. Schweiz, p. 420, f. 291. Sacc. Syll. vii. 748. Sydow, Monogr. iii. 146. Spermogones. Epiphyllous, in little groups, yellowish. Fig. 226. Ph. Rubi-Idaei. a, part of leaf of Raspberry, showing the circular groups of caeomata, nat. size; b, a single group, magnified, showing the spermogones in the centre, surrounded by a pulvinate ring of caaomata, still partly covered by the epidermis ; c, ascidiospore and rf, paraphysis, x 600 ; e, teleutospore x 300 ; /, teleutospore after boiling in lactic acid ; g, single cell of same, showing the three coats, (1) hyaline and gelatinous, (2) dark and rigid, (3) the thin endospore ; /(, one of the middle coats, detached by pressure, showing the three bordered germ-pores. sEcidiospores. Ca?omata epiphyllous, small, round, yellow, usually forming a circle round a group of spermogones, sur- rounded by a dense border of hyaline, clavate, inward-curved, KUEHNEOLA 299 thin-walled paraphyses ; spores ellipsoid or ovate, sparsely echirmlate, yellow, 21 — 24x18 /a. Uredospores. Sori hypophyllous, small, scattered, pale- orange, surrounded by thin-walled clavate paraphyses ; spores roundish or ellipsoid, sparsely echinulate, yellow, about 21 x 18 /a; pores obscure. Teleutospores. Sori similar, but black ; spores cylindrical, of 6 — 10 (mostly 7 or 8) cells, slightly tapering at the apex and papillate, not constricted, rounded below, verrucose, brown,- 80 — 135x28 — 35//,; three germ-pores to each cell; pedicels very long, thick, colourless, persistent, swollen towards the base. On Rubus Idaeus, and possibly on other species of the genus. Not uncommon. June — October. (Fig- 226.) Lecythea gyrosa of Berkeley is the caeoma-stage, in which a pulvinate ring of spores is formed round the little group of spermogones, leaving the centre, at first sight, apparently unoccupied. This form is not confined to the early part of the season, but may be found as late as August or September. It is much rarer than the other spore-forms. - Although this Phragmidium is not found in America, there is a very similar one, P. imitans Arthur, on allied species of Rubus, inhabiting the United States and Canada, in which the cseomata are similarly formed.' On the cultivated Raspberry, when this disease is present, it can be prevented from spreading by spraying with potassium sulphide solution or dilute Bordeaux mixture. All diseased leaves should be burnt. Distribution : Europe, Siberia, Japan. KUEHNEOLA Magnus. Autoecious. Spermogones subcuticular, somewhat flattened, without ostiolar filaments. Uredo-sori of two kinds : primary, i.e. the equivalent of the caeomata, often surrounding the spermogones, without paraphyses ; secondary, similar, but scattered, some- times with paraphyses ; uredospores borne singly on pedicels, with few and inconspicuous equatorial pores. Teleutospores of :><)<> KUEHNEOLA several cells as in Phragmidium, but the wall is faintly coloured or colourless, and smooth : pores one in each cell, apical. This genus is qoI confined t<> Rosacea-, being recorded on Malvaceae in America, where also lioth the Rritish species are found. It is not closely allied to Phragmidium: the wall of the teleutospores and the germ-pores are quite different. But I am also of the opinion that the two species included here are not in reality congeneric. 1. Kuehneola albida Magnus. Uredo Mulleri Schrot. Flor. Schles. iii. 375. Plowr. Ured. p. 256. Chrysomyxa albida Kiihn, Bot. Centralbl. xvi. 154 (1883). See also Bull. Soc. Myc. Fr. xvii. 31. Phragmidium albidum Lagerh. Mitth. Bad. Bot. Ver. 1888, p. 44. Fischer, Ured. Schweiz, p. 415. Kuehneola albida Magn. Bot. Centralbl. lxxiv. 169 (1898). K. Uredinis Arthur, N. Airier. Fl. vii. 186 (1912). Spermogones. Epiphyllous, clustered on small reddish spots. Fig. 227. A", albida. Uredospore and teleutospores. Uredospores. Sori, primary epiphyllous, yellow, often in rings surrounding the spermogones, secondary hypophyllous, scat- tered, occasionally on the calyx and stems, smaller, punctiform KUEHNEOLA 301 and paler, when old whitish; spores globose, obovate or irregu- larly polygonal, closely verruculose-echinulate, yellow, about 20 — 26 fi; pores indistinct (? three or four). Teleutospores. Sori hypophyllous, singly or in little round- ish groups, but never confluent, \ — £ mm., pulvinate, yellowish- white ; spores cylindrical-clavate, flattened and irregularly coronate at the summit, of 2 — 13 (mostly 5 or 6) trapezoidal cells which are smooth and colourless, each measuring 17 — 30 x 15 — 25 fi; the wall of each cell becomes thicker from below upwards, and the upper edge is irregularly undulated ; the germ-pore is situated in one of the finger-like projections at the upper edge of each cell ; pedicels very short, sometimes wanting ; each cell is really a perfectly distinct spore. On Rub us fr.uticosus. Not rare; Taunton, Hereford, Worcestershire, Woolmer, New Forest, North Wootton, Ayrshire, Ireland, etc. Teleutospores, September — November. (Fig. 227.) The life-history of this species is imperfectly known. The uredospores may precede the teleutospores, but may also be found simultaneously with them and (presumably the secondary uredospores) even in the same sorus. The primary uredospores seem to occur in chains, represent secidiospores, and probably germinate at once. The teleutospores germinate in situ on the leaves as early as the beginning of September. Juel, who experi- mentally demonstrated the connection of the two spore-forms, suggested that some uredospores survive the winter and germinate in the spring ; J. Midler and S. Strelin state the same of the secondary uredospores. This species is said to be found on Rubus caesius ; it occurs on many species of Rubus in North America. The name given by Arthur (K. Uredinis) rests upon a (probably true) idea that the teleutospores constitute the fungus named by Link Oidium Uredinis and placed among the Hyphomycetes : see Sacc. Syll. iv. 16. Distribution : Europe, North America. 2. Kuehneola Tormentillae Arthur. I'h ray, iridium Tormentillae Fckl. Symb. Myc. p. 46. Plowr. Ured. p. 222. Fischer, Ured. Schweiz, p. 414, f. 288. Sacc. Syll. vii. 744. Sydow, Monogr. iii. 10"). P. Potentillae-canadeasis Dietel in Hedwig. Beibl. xlii. 179. Sydow, Monogr. iii. 106, f. 48. Kuehneola Tormentillae Arthur, Res. Sci. Congr. Vienn. p. 342 (1904). K. obtusa Arthur, N. Amer. Fl. vii. 185 Q912). 302 XEX<>nornrs Spermogones. Epiphyllous, in little groups. Uredospores. Sori of two kinds, primary epiphyllous, sur- rounding the spertnogones, secon- dary hypophyllous, scattered, small, round, punctiforin, orange, sur- rounded byafewclavato paraphyses; spores spherical to obovate, finely echinulate, reddish-orange, 20 — 23x14—20^. Teleutospores. Sori hypophyl- lous, similar, but bright-brown; spores cylindrical, fusiform or cla- vate, of 2 — 7 (mostly five) cells, ofte'n curved, thickened at the apex like many Pucciniae, slightly con- stricted, tapering below, smooth, sienna-brown, 52 — 140 x 18 — 24 fi; epispore thin, with one germ-pore in each cell; contents orange; pedicels varying in length, persistent, not much widened below. On Potentilla Tormentilla (= P. erecta) and possibly on other species of the genus. Very rare. September, October. (Fig. 228.) This species resembles a Puccinia in some respects, especially in the thickening of the apex of the teleutospores, and the position of the solitary germ-pore of each cell ; the wall of each cell becomes darker upwards, the lower cells being nearly colourless, and the uppermost a pale clear brown, all quite free from any warts or projections. They can germinate in autumn (September) like those of K. albida. Dietel says that the uredospores and their mycelium can survive through the winter. Distribution : Europe, North America. Fig. 228. K. Tormentillae. Teleutospores. XENODOCHUS Schlecht. Autoecious. Caeomata large, indefinite, without paraphyses or peridium. Uredospores absent, represented by the secondary smaller casomata. Teleuto-sori similar, often on the same mycelium ; teleutospores of very long chains of cells, not verrucose. XENODOCHUS 303 Arthur records and names another species, X. minor, on Sanguisorba latifolia, from Alaska. Xenodochus carbonarius Schlecht. Xenodochus carhonariv& Schlecht. in Linnsea, i. 237. pi. 3, f. 3. Cooke, Handb. p. 489 ; Micr. Fung. p. 201, pi. 3, f. 29. Plowr. Ured. p. 227. Sacc. Syll. vii. 751. Phraymidium carbonarium Wint. Pilze, p. 227. Fischer, Ured. Schweiz, p. 406, f. 284. Sydow, Monogr. iii. 157, f. 67. ^Ecidiospofes. Cseomata hypophyllous, on coloured spots, elongated, large, erumpent and pul- verulent on the nerves, petioles and stems, roundish and scattered on the leaves, bright orange-red or vermilion; spores roundish to oblong, verruculose, orange, 18 — 24 /x, in short chains. Teleutos'pores. Sori amphigenous, often confluent with the cseomata, roundish, soon naked, thick, large, pulvinate, black ; spores elongated- cylindrical, often curved, of 4 — 22 cells, rounded at both ends, strongly constricted, smooth, dark-brown, 200 — 300 x 24—28 fi ; each cell with two opposite germ-pores at the upper margin, except the uppermost which has one apical germ-pore ; pedicels very short, persistent. The whole chain is surrounded by a distinct sub- hyaline membrane, which swells up considerably in lactic acid. On Sanguisorba officinalis. Not uncommon. June — Oc- tober. (Fig. 229.) The teleutospore-cells of the Burnet Chain Rust mature from above downwards, the upper ones being darkest and the lower ones often colourless. There are no uredospores ; perhaps they are represented by secondary secidiospoies. The distinctions of this species from the other Phragmidieaj are quite sufficient to justify its generic separation. Winter assigns to the cseomata clavate paraphyses which I cannot find. Distribution : Europe, Asia. Fig. 229. A', carbonarius. Teleutospore x 360 ; upper part of same, x 600. 304 GYMNOSPORANGIUM GYMNOSPORANGIUM Hedw. fil. Eeteroecious. All fche sori subepidermal. Spermogones Bpherical, with ostiolar filaments. iEcidia more or less elongated or tubular; peridium membranous, rupturing by lateral slits; spores brown, with numerous evident germ-pores. Teleuto-sori erumpent, naked, variously shaped, gelatinous, expanding when moist; spores with long pedicels, usually two-celled, generally with two pores (1 — 4) in each cell, mostly near the septa: the cell-wall of the pedicels becomes gelatinised. Germination by an ordinary basidiu m and roundish basidiospores : these are thrown off with a jerk, at maturity, like the spores of the Hymenomycetes (Coons, 1912). There are many species of Gymnosporangium in America (Kern in X. American Flora gives :V1), of which only three occur in Europe, and one or two in Japan. Several of these have three, four, or five-celled teleutospores. The aecidia of the Gymnosporangia are on Rosacese (except one on Hydrangeae) ; the teleutospores are all on Cupressineae (Juniperus, Cupressus, Chamaecyparis, Libocedrus), on which they often form swellings, i.e. galls. But we get also one remarkable exception to this rule in the autcecious species (the only one known) G. bermu- dianum, which produces both its geeidia and its teleuto-sori on Junipers (J. bermudiana and other species). 1. Gymnosporangium clavariaeforme DC. Tremella clavariaeformis Jucq. Collect, ii. 174. Podioiuia Junqjirri Fr. ; Cooke, Handb. p. 510; Micr. Fung. p. 214. Roestelia lacerata Tul. ; Cooke, Handb. p. ">34 ; Micr. Fung. p. 193, pi. 2, f. 22—6. /.'. carpophila Bagnis, Flofa^Jsiii. 317. Gymnosporangium clavariaeforme DC. Flor. fr. ii. 217. Plowr. Ured. p. 233. Fischer, Ured. Schweiz, p. 383, f. 27"). Sacc. Svll. vii. 737 p.p. Kern, in North Amer. Fl. vii. 202. Sydow, Monogr. iii. 59, f. 29. >'//- rmogones. Numerous, amphigenous, but chiefly epi- phyllous, in little clusters on red spots, yellow, then dark-brown. GYMNOSPORANGIUM 305 /Ecidiospores. yEcidia clustered on yellow or orange thick- ened spots on the leaves, fruits, and stems, cylindrical, up to Fig. 230. G. clavariae forme. /Ecidia (Rnestelia) on leaf, fruit and branch of Hawthorn (reduced) ; a, peridium x 16. The fruit and gall on branch are shown as they appear when the peridia are old and the mass looks some- what like a honey-comb. •2h mm. high, fimbriate above, at length lacerate to base, pale- brown ; spores verruculose, brownish, about 28 — 30 jx ; pores 8 — 10, scattered. Teleutospores. Spores collect- ed in long, cylindrical, conical, ribbon-like or tongue-shaped masses about 1 cm. long, which are at first firm, then gelatinous, finally horny when dry, pale-orange, protruding from fusiform swellings m of the branches ; spores fusiform, varied much in length and breadth, from thick- walled, 50—60 x 15—21 /*, to thin-walled, 100—120 x 10—12^, rounded or attenuated above, hardly constricted, tapering gradu- ally below, smooth, dark yellowish or pale-brown ; germ-pores two in Fig. 231. G. clavariaeforme. each cell; ped.cels hyaline, very *££ J* SftSSSS (Z long. duced) ; teleutospores x G00. G. u. 20 306 GYMNOSPORANGIUM . Keidia <>ii Crat'iiyiis Oxyoanttha, C. rnonogyna, Pyrus communis. July — Augusl : teleutospores on Juniperus com- munis, April and May. Rather common. (Figs. 230, 231.) The teleutospore-mycelium is perennial in tin- branches of the Juniper ; the sporea are produced in spring and germinate immediately. The ;rciili;i cm We produced <>n the Hawthorn in two to four weeks or more after infection. Their spores in turn infeel young shoots of Juniper, in which the mycelium gives rise to teleutospores in one year (von Tubeuf), ortwo years Plowrighi , and in each succeeding" year so long as the branch survives. The mycelium in Crataegus is purely local and annual. The experimental culture of this species has been performed unnumbered times by many mycologists ; the secidia have been produced on several specie-, of Crataegus and possibly on the Apple and the Medlar ; also (spermogones only in som n Cydonia, Sorbus, and Avielanchier and the teleutospores on Juniperus nana and ./. oxycedrus, as well as on •/. count} "nix. This culture is one of the easiest to perform of all that are known among the Uredinales. The only means of checking this disease on the pomaceous hosts is by the removal of the affected Junipers, if thej can be found. Distribution : Europe, Algeria, North America. 2. G-ymnosporangium confusum Plowr. JEcidium Mespili DC. Flor. Fr. vi. 98. Roestelia Cydoniae Thiim. in Sacc. Syll. vii. 834. Gymnosporangium confusum Plowr. Ured. p. 232, pi. 4, f. 13, 14. Fischer, Ured. Schweiz, p. 385, f. 276 ; Zeitschr. f. Pflanzenkr. 1891, i. 193, 260. Sydow, Monogr. iii. 56, f. 27. Spermogones. Orange-coloured. jEcidiospores. /Ecidia on thickened spots, obconical or cylindrical, opening at the summit which is at length fimbriate, orange above, often surrounded by a reddish or purple line ; spores verruculose, pale-brown, 21 — 24 /x. Teleutospores. Spore-masses at first pulvinate, dark choco- late-brown, almost black, then irregularly conical, 5 — 8 mm. long, chestnut-brown, swelling when moist ; spores oval to fusiform, smooth, of two kinds, some with hyaline spore-walls and orange contents, tapering above, others with dark-brown thick walls, rounded above, about 30 — 50 x 20 — 25 /x ; germ- pores 2 — 4 ; pedicels hyaline, rather long. GYMNOSPORANGIUM 307 iEcidia on leaves, branches and fruits of Crataegus Oxy- acantha, C. monogyna, Mespilus ger- manica, Cydonia vulgaris, June — August; teleutospores on Juniperus Sabina, April and May; uncommon. (Fig. 232.) The course of the life history of this species is identical with that of the pre- ceding. It is distinguished by its shorter and broader teleutospores, which resemble those of G. Sabinae, but the rccidia are quite different from those of that species and resemble those of O. clavariaeforme. The fficidia of G. confusum are reported also doubtfully on Pyrus communis and Cotoneaster integerrimus. The cells of the peridium have their side-walls marked with elongated obliquely placed ridges, while those of G. clavariaeforme have them coarsely warted. This description is taken from Fischer and Plowright, both of whom give long accounts of its peculiarities. In all Plowright's specimens of G. confusum (produced artificially on Hawthorn, Quince, Medlar) the peridia were smaller, shorter, less deeply torn, and rather more inflated forme, i.e. of the typical R. lacerata. Fig. '2rS2. G. confusum. a, cluster of ascidia on Haw- thorn twig, nat. size; b, two peridia of same x 10 ; c, clus- ter of aBoidia on Quince leaf, x 5 ; d, two teleutospores on J. Sabina x 600. All from cultures by Plowright. than those of G. clavariae- Distpjbution : Europe, Northern Persia. 3. Gymnosporangium Juniperi Link. Tremella juniperina L. Sp. PL p. 162"). Roestelia cornuta Tul. ; Cooke, Haudb. p. 534, f. 218 ; Micr. Fung. p. 193, pi. 2, f. 18, 19. Gymnosporangium Juniperi Link, Obs. i. 7. Cooke, Handb. p. 509, f. 203 ; Micr. Fung. p. 214. Sydow, Monogr. iii. 27, f. 11. G. juniper inum Fr. Syst. Myc. iii. 506. Plowr. Ured. p. 235. Fischer, Ured. Schweiz, p. 391, f. 278. Saccx Syll. vii. 738 p.p. G. corn "turn Arthur, Mycologia, i. 240. Kern, N. Amer. Fl. vii. 198. Spermogones. Epiphyllous, subepidermal, in roundish crowded groups, yellowish, at length black. 20—2 3<>X QYMNOSPOEANGIUM JEcidiospores. zEcidia hypophyllous, in irregular or circular groups, horn-shaped, conical, curved, .1 nun. wide, 2 nun. long, at length open and fimbriate above, yellowish- brown, on round spots which are brownish below, and bright-orange or 'xli red on the upper side; spores finely verruculose, brown, 21 — 28x19 — 24/^; Pig. 233. G. Juniperi. germ-pores 8— 10, scattered. Groups of reculia on leaflet ° J of Mountain Ash xih; Teleutospores. Spore-masses on "•:m anopened peridium young twigs and occasionally on leaves, more or less globose, 1 — 3 mm. across, at first chocolate-brown, then orange, soft, gelatinous; spores obtusely fusiform, of two kinds, first thick-walled and brown, second thin-walled and yellowish, 31 — 52x21 — 30 /z, (Dietel), 66 — 75 x 17 — 27 /x (Plowright); germ-pores one or two in each cell ; pedicels rather long. .Ecidia on Pyrus Aucuparia, July — October; teleutospores on Juniperus communis. May and June. Not common: Surrey, etc. iEcidia very abundant at Blair Athol, August, 1905 (D. Prain). (Fig. 233.) The teleutospores differ from those of the other European species in the possession of a broad colourless papilla over each germ-pore. Their mycelium causes fusiform swellings of the smaller branches. Brcbner infected a leaf of Mountain Ash from the teleutospores and obtained spermogones in eleven days. Distribution : Europe, North America. 4. Gymnosporangium Sabinae Wint. Tremella Sabinae Dicks. PI. Crypt. Brit. i. 14. Podisoma Sabinae Fr. : < 'ooke, Handb. p. 510 ; Micr. Fung. p. 214. Roeslelia cancellata Reb. Fl. Neom. p. 350. Cooke, Handb. p. 533 ; Micr. Fung. p. 193, pi. 2, f. 20, 21. Gymnospora n Empetri Pens. ; Cooke, Handb. p. 527 ; Micr. Fung. p. 216. Chrysomyxa Empetri Schrot. Krypt. Flor. Schles. iii. 372. Plowr. Ured. p. 253. Sacc. Syll. vii. 762. Fischer, Ured. Schweiz, p. 557. Melampsoropsis Empetri Arthur, N. Arner. Fl. vii. 118. Uredospores. Sori hypophyllous, occasionally (according to Magnus) epiphyllous, small, roundish or elongated, some- times arranged in lines parallel to the midrib, covered by the raised epidermis, orange; spores in short chains, ellipsoid or polygonal, densely verrucose, 2(3—35 x 18—25 fi, 30—35 x 21 — 28//, (according to Fischer), 25—30 x 17—25 fi (Plowright), 26—37 x 18—26 n (Arthur); wall rather thick, colourless ; contents orange. Fig. 235. G. Empetri. a, leaf of Empetrum nigrum, showing three uredo-sori, x 10 ; b, uredospores x600. On Empetrum nigrum. Uncommon ; North Wales, etc. May — October. Often in small quantity, and very inconspicuous except when fresh. (Fig. 235.) This fungus is considered to be a Chri/somyxa because the uredospores are produced in chains. The teleutospores seem doubtfully to have been observed by Rostrup and Lagerheim, and might possibly be discovered in this country if looked for. This parasite also occurs on the same host in the northern half of North America, but there also no teleutospores have ever been seen. Distribution : Europe, North America. 312 CHRYSOMYXi 2. Chrysomyxa Pyrolae Rostr. Y. vi. 99. 1 !.'•. Handb. i>. 529; Micr. Fun-. . /.'• idiwm Pyrolae l>< '. Flor. TVichobasis Pyrolae Berk. p. 223 p.p. Chrysomyxa Pyrolae Rostr. Bot. Centr. iii. I ^G (1881). Plowr. Ured. p. 253. Fischer, [Jred. Schweiz, p. 429. Sacc. Syll. vii. 761. Melampsoropm Pyrohn- Arthur, X. Amer. Fl. vii. I Is. Uredospores. Sori hypophyllous, often covering the whole surface uniformly, roundish, h — 1 nun. diam., soon naked, surrounded by the torn epidermis and a very delicate evanescent peridium, yellow: spores in chains, roundish or polygonal, verrucose, orange, 21 — 28x18 — 21 /x. Teleutospores. Sori hypophyllous, covering the whole leaf-surface uni- formly, up to I mm. wide, roundish or oblong, flat, waxy, yellowish- then blood-red, when dry brown ; spores ellipsoid, about 8 fx wide, in rows as much as 100 — 120^ long. On Pyrola motor. P. rotundifolia and its var. P. maritima. Uncom- mon ; Edinburgh, Kew Gardens, Lancashire, etc. April — August. (Fig. 236.) Fig.236. ('.Pyrolae. a.uredo- sori, on underside of nat. size ; b, chains of young uredospores, showing inter- calary cells; c, mature ure- dospore. On I', maritima (Formby, ex herb. H. J. Wheldon). This parasite may possibly be heteroe- cious ; Fraser (Mycologia, 1911, iii. 67) suggests that Peridermium conorum-Piceae is its secidium ; Rdstrup, Arthur, and Kern had already expressed the same idea. The teleutospores are rarely formed, and the fungus probably maintains itself by its uredospores, which can he distinguished from those of Pucciniasirum Pyrolae by their sori being scattered (not in groups), and by the absence of a distinct peridium. Distribution : Europe, North America. CRONARTIUM 313 As the remarkable suggested secidium-stage may possibly be found in this country, it will be convenient to add here the description given of it, so as to aid in its identification : PERIDERMIUM CONORUM-PlCEjE (JEcidiinii conorum-Piceae Reess.) ^Ecidia on the outer surface of the scales, large, ±, cm. or more in diam., oblong or irregular in shape, one or two (or in any case few) on each scale, immersed, white ; spores ellipsoidal, orange, 28 — 35 x 18 — 24 jx ; epispore with numerous, large, crowded, prismatic warts, each about 3 4 „ wide Fig. 237. P. conorum-Piceae. ^ JEcidiospore, from scales of On cones of Picea eoccelsa. An- ^goo ""'" ^ FiSCher)' gust, September. In some years rather common in the Alps and Jura Mountains. (Fig. 237.) CRONARTIUM Fries. Spermogones hemispherical. iEcidia with a broad, inflated, irregularly torn peridium ; recidiospores with a coarsely verru- cose membrane, smooth on one side, without germ-pores, separated by well-marked intercalary cells. Uredo-sori enclosed in a hemispherical peridium which opens at the summit by a narrow pore ; uredospores produced singly, on pedicels, echinu- late, without germ-pores. Teleutospores abstricted in long chains, and remaining united into cylindrical columns which are horny when dry, germinating as soon as mature. Basidio- spores round, very minute. 1. Cronartium asclepiadeum Fr. Cronartium asclepiadeum Fr. Ob.s. Myc. i. 220 (1815). Fischer, Ured. Schweiz, p. 431, f. 295. Sacc. Syll. vii. 597. C. Paeoniae, Cast. Catal. PI. Marseill. p. 217 (1845). Cooke, Micr. Fung. p. 215. 314 CRONARTIUM C. flaccidum Wint. Pilze, i. 236 1884). Plowr. Ured. p. 254. Sacc. Syll. \ii. 598. Peridemiium Cornui Kleb. Zeitschr. f. Pflauzenkr. L892, ii. 269, pi. 5, f. 2. /Ecidiospores. /Ecidia (P. Cornui) erumpenl from the bark, forming large reddish-yelkm Madders, generally occupying a portion of a branch in large numbers : spores ellipsoid, 'I'l — 26 x 16 — 20 /a; epispore 3 — 4 \x thick, verrucose, thinner od part of its surface and there smooth or somewhat reticulate. Uredospores. Sori small, pustular, surrounded bya peridium which opens at the summit with a pore; spores ellipsoid or ovate spai Immlate. 21 — 24x17 — 21 fi. Fig. 238. C. asclepiadeum. n. Peridermium Cornui on branch of Pine; h, teleuto-sori on leaf of Peony (reduced); c, uredospore x 600 ; d, part of a column of teleutospores x 300. Teleutospores. Sori cylindrical, often curved, arranged in large clusters, over 1 mm. high, brown, at length horny, compact; spores ellipsoid or cylindrical-oblong, reaching 5G /u. long and 14 /j, broad; epispore thin, slightly thicker above. .Keidia on the branches of Pinus silvestris, May, June; uredo- and teleutospores on Paeonia officinalis in gardens, July — October. Very uncommon. (Fig. 238.) It has been proved by the researches of Cornu, Klebahn, Fischer and many others, not only that the remarkable eecidia on Pine are genetically connected with the other spore-forms on Peony, but also that they can CRONARTIUM 315 produce the same on Vincetoxicum, Nemesia, Cynanchum, and Verbena, as well as on many species of Paeonia. There is reason for believing that the parasitism on Nemesia, at least, has arisen at a very recent date. This species is therefore plurivorous in its teleuto-stage, but not in its secidial stage. The mycelium is perennial, according to Fischer, in the pine- branches; it produces eecidiospores in May and infects the alternate hosts, on which uredo- and teleutospores are borne during the summer — the latter can germinate at once. 2. Cronartium Quercuum Miyabe. Uredo Quercus Brondeau in Duby, Bot. Gall. ii. 893 (1830). Cooke, Handb. p. 526 ; Micr. Fung. p. 216. Plowr. Ured. p. 257. Fischer, Ured. Schweiz, p. 539. /'< i- ill milium defafiiKiHS and P. giganteum Tubeuf, Pflanzenkr. p. 429. Melampsora (?) Quercus Schrot. ; see Sacc. Michel, ii. 308 ; Syll. vii. 594. Cronartium asclepiadeum var. Quercuum Berk. Grevill. iii. 59 (from the U.S.). Sacc. Syll. vii. 597. C. Quercuum Miyabe, Bot. Mag. Tokyo, xiii. 74 (1899). C. Quercus Arthur, North Americ. Fl. vii. 122 (1907). \_yEcidiospores. ^Ecidia caulicolous, forming' subglobose swellings 5 — 25 cm. across, arranged in tortuous lines or cerebroid, at first /r^\ ^^\ orange-yellow, bladdery; peridium col- ourless, circumscissile, soon falling away in flakes or sheets, about 2 cells thick, outer surface smooth, inner verrucose ; Fi"- '239- c- Quercuum. . Uredospores, on Quercus cells roundish or irregularly compressed, (ex herb. Thos. Brittain). walls very thick, lumen small ; spores obovate, 25 — 32 x 17 — 23 p. ; wall colourless, uniformly thick (2| — 31 /x), coarsely verrucose, usually with smooth spot at base and extending up one side, tubercles somewhat deciduous.] Uredospores. Sori hypophyllous, thickly scattered, round, small (^ mm.), hemispherical, dehiscing by an apical pore, at length surrounded by the torn epidermis, yellow ; peridium delicate or wanting; spores obovate to broadly ellipsoid, orange- yellow, 15 — 25x10 — 17 /x ; wall colourless, 3 /x thick, evenly echinulate with short strong points. [Teleutospores. Sori hypoj^hyllous, columnar, filiform, 2 — 3 mm. long, 150 — 175 p, thick; spores fusiform to oblong, 30 — 40 316 I RON \i:Tll'.M x 1") — 20^; wall nearly colourless, smooth, 2 — 3// thick; basidiospores oval. ] [iEcidia on branches of Pinus ;] uredo- and beleutospores on leaves of Quercus pedunculata. Rare; Hastings, St Leonards, Sin iv. Eursl aceaux, Salisbury. October. (Fig. 28!).) (>nlv the uredosporea have been found in this country, and usually so als the continent. The aecidia have been found in the United States on five species of Pinus, .is well aa on others in Japan ; and the other spore-forms on at least 20 species of Quercus in various localities. The description (except for the uredospores is taken from Arthur (I.e.). The uredospores seem to be most common on the leaves of the shoots thai spring up from the stools of felled oaks. They occur also in France and Switzerland in the same way. Their dimensions are smaller in this country than those given by Arthur. DISTRIBUTION: Europe. United States, Guatemala, Japan. 3. Cronartium ribicola F. de Waldh. Cronartium ribicola Fisch. de Waldh. in Rab. Fung. Eur. no. 1595 ; Hedwig. xi. 182. Klebahn, Wirtsw. Rostp. p. 382. Sacc. Syll. vii. 598. Fischer, Ured. Schweiz, p. 433, f. 296. Gard. Chron. ser. 3, xi. 736; xii. 44, 133, 137, 501 ; xiii. 425; xxiii. 202, 211: xxvi. 72-3, 94. Arthur, N. Amor. Fl. vii. 122. Peridermium Strobi Kleb. Abhandl. Nat. Ver. Bremen, x. 153, pi. 1, f. .-,_io, 13, 14: Zeitschr. f. Pflanzenkr. ii. 269, pi. 5, f. 1. Spermogones. Irregular elevations of the bark, 2 — 3 mm. broad, yellow, with a minute opening through which exudes a sweet fluid in which the spermatia are included; spermatia ovoid to elliptical, 2 — 4^yu. JEcidiospores. iEciclia erumpent from the bark in the form of a bladder, with an inflated peridium, about 1 cm. high, yellowish-white ; spores roundish or polygonal, coarsely verru- cose, except on an elongated smooth patch, orange, 22 — 29 x 18 — 20 /m; epispore 2 — 1\ \x thick, 3 — 3^/a at the smooth spot. Uredospores. Sori hypophyllous, small, pustular, forming orbicular groups 1 — 5 mm. across, surrounded by a delicate peridium which opens at the summit with a. pore; spores ellipsoid to obovoid, distantly and sharply echinulate, orange, 21 — 24x14 — 18 /J.; epispore colourless, 2 — 3^ thick. CRONAKTIUM 317 ' Teleutospores. Columns hypophyllous, cylindrical, curved, up to 2 mm. long, crowded especially along the veins of the leaf, sometimes covering the whole leaf, orange to brownish- yellow ; spores oblong, smooth, reaching 70 fi long x 21 fi broad. vEcidia on stems and branches of Pinus Cembra, P. monti- cola, P. Strobus, March— June ; uredo- and teleutospores on Uibes nii/rum, R. rubrum, etc., July — October. Uncommon except when imported; Surrey, King's Lynn, Exeter, West- burv, Woburn, Windsor Forest, Perth, etc. (Fig. 240.) Fig. 240. G. ribicola. a, spore of Peridermium Strobi; b, the teleutospore- columns on leaf of Eed Currant (reduced); c, uredospore; d, top of a column of teleutospores, x 600. This dangerous parasite, sometimes called the Weymouth Pine Rust, is confined in its secidial stage to the five-leaved Pines ; it is reported on the continent also on P. excelsa and P. Lambertiana. The aecidiospore is distinguished from that of Peridermium Pini and P. Cornui by the fact that a great part of the surface is smooth. The attacked trees are stunted, the tops show a bushy growth that can easily be recognised; the part where the mycelium is growing is swollen. In the Currant, the attacked leaves become thicker in texture, and different in colour. It was first discovered, in this country, at King's Lynn by Dr Plow- right, who found the Gronartium on leaves of black, white, and red currant on July 3, 1892. On August 13 he exhibited in London a branch of P. Strobus affected by the mycelium of the alternate stage; the tree grew about fifteen yards away from the currant bushes. On March 19, 1893, he found the Peridermium in full perfection at the same place. Though the secidium occurs in Europe chiefly on the Weymouth Pine (/'. Strobus), yet that cannot be its original host, since neither Cronart'm m ribicola nor Peridermium Strobi was found in America (the home of P. Strobus) until the fungus was imported on it from Germany. P. Cembra, 318 COLEOSPOKIACE.K the Swiss Stone Pine, however, will equally serve as host, as ba shown by Tranzschel, and it 1ms been found "ii that tree in Russia and in Switzerland. One ■>! these countries was probably the original home of the parasite, from which it is spreading wherever its hosts will grow. This is one of the species with which attempts have been made to produce infection by the use of tin- abundant spermatia (Klebahn, Wirtsw. Rost. p. :i^7 , bul numerous trial- on healthy Weymouth Pines well- .lit irely without result. Distribution: Europe (north of the Alps), Siberia, Japan, North America. COLEOSPOPJACE.K. Jvidium furnished with a peridium. Spermogones sub- epidermal, flattish, linear, without ostiolar filaments, dehiscing by a slit. Teleutospores in one (more rarely two) subepidermal layers, dividing as they mature into four superimposed cells, each of which germinates by a sterigma bearing one basidio- spoiv. ( 'ui.KeM'oiUK.i:. Hetercecious. . Kcidia (Peridermiwn) more or less cylindrical, with inflated peridium, irregularly torn at summit. Uredospores abstricted in chains. Teleutospores with a strongly thickened gelatinous wall above. Basidiospores ovate. Coleosporium. Ochropsore.e. Heteruecious. . Ecidia with cup-shaped peridium. Uredospores abstricted singly. Teleutospores thin-walled. Basidiospores fusiform. (Jr/n-nj ,.<,,/•> i. Zaghouanisle. Autcecious. For the present, the abnormal genus Zaghouania may lie arranged as a subfamily of the Coleosporiacese ; it is distinguished especially by the fact that the four-celled basidium is formed internally, but emerges from the teleutospore before the formation of the round basidiospores. See p. 331. Zaghouania. The internal "basidium'" which has been considered as a character of this family is not confined to it, being found also in CJirysopsuro, which belongs to the Pucciniacese. It is remarkable that no species of the family has up till now been discovered in Australia, while only one of the Eurasian species COLEOSPORIUM 319 is indigenous to North America. This family retains probably, in the mode of germination of its teleutospore, a very primitive character, but has nevertheless undergone a large amount of recent evolution, and is no doubt worthy of subdivision. There is a North American species, belonging to the Coleosporiacea1, which is deserving of great attention. It is Oallowaya Pini Arthur (formerly Coleosporium Pini Galloway). which has teleutospores only, and on leaves of Pinus inops, i.e. on trees of the same order on which Coleosporium has its secidia. Similarly, among the Melampsoracese, there is a like case in Necium Farlowii Arthur, which has its teleutospores on Abies canadensis, while various hetercecious Melampsoracea^, with similar teleutospores on other (non-coniferous) plants, have their aecidia on Conifers. Again in the Cronartiaceae, Chrysomyxa Ledi and C. Rhododendri are hetercecious species having their secidia on Picea excelsa ; but there is also C. Abietis having its teleutospores on the same host (P. exceUu) and no other spore form. Gymnosporangium bermudianum, already mentioned (p. 304), furnishes a somewhat similar instance. The evolutionary significance of these facts has not yet been elucidated. COLEOSPORIUM Lev. iEcidia with a more or less cylindrical inflated peridium, which opens by a cleft and becomes irregularly torn; secidiospores with colourless membrane, without germ-pores, superficially tuberculate, the tubercles somewhat deciduous. Uredospores not enclosed in a peridium, abstricted in short chains, resembling the secidiospores. Teleutospores in flat, waxy, indehiscent sori, with a colourless gelatinous membrane, which is thin and wavy at the sides but strongly thickened above, at first filled with a rich orange-red oily mass ; at length each spore divides into four superimposed cells, which in autumn can germinate in situ as soon as mature, with a long sterigma. The species of Coleosporium are morphologically very much alike, and are distinguished chiefly by their hosts. Moreover. 320 COLEOSPOMUM since the secidia of all of them live on needles of two-leaved Pines, it is impossible to say where any given a'cidium of this class belongs, vvithoul culture-experiments, [f a Goleosporiwm can !»• found in the immediate neighbourhood, tin' aecidium may be conjecturally assigned in that. But though the gecidia are very much alike, there are certain differences to be seen, although nothing is yet known about their constancy as specific characters. In certain cases nearly all the peridia will be found on one <>n/;/ of the two leaves of Pine in each fascicle; this is, according to my experience, almost always true of G. Senecionis. In others, the peridia occur on botJi leaves of the fascicle; this seems to be the case usually with C. Tussilaginis. Some peridia are very bladdery and inflated, others are more Hat or cylindrical. It may be possible in the future to distinguish them by these means, especially since the Goleosporia lend themselves easily to artificial cultures; see under G. Rhin- anthacearum and C. Tussilaginis. 1. Coleosporium Senecionis Fr. Credo farinosa var. Senecionis Pers. Syn. p. 218. Peridermium Pini Chew ; Cooke, Handl). p. 535 p.p., f. 219; Micr. Fung. t)l. 2, f. 27, 28 (not the description p. 194). /'. acicolum Link ; Cooke, Micr. Fung. p. 194. Puccinia glomerata uredospores) Cooke, Handb. p. 500. Coleosporium Senecionis Fr. Summ. Veg. Scand. p. 512. Cooke, Micr. Fung. p. 218, pi. 7, f. 145, 146. Plowr. Ured. p. 248 p.p. Sacc. Syll. vii. 751. Fischer, Ured. Schweiz, p. 451. Arthur, X. Amer. Fl. vii. 94. Spermogones. Amphigenous, scattered, conspicuous. sEcidiospores. vEciclia (Peridermium acicolum) oblong or shortly cylindrical, fragile, rupturing irregularly, whitish; spores almost all oblong, seldom roundish, densely verrucose, orange, 25—35x15—25/*. U 'redospures. Sori hypophyllous, roundish or on the stems elongated, soon naked and pulverulent, orange ; spores mostly oblong, verruculose, 20 — 31 x 14 — l7,u; epispore rather thick. Teleutospores. Sori hypophyllous, forming little red crusts ; spores prismatic, length up to 100/z, breadth 18 — 24p ; epispore at summit up to 22/x thick. COLEOSPORIUM 321 /Ecidia on (?one of the two) leaves of Pinus austriaca, P. silvestris, May, June ; uredo- and teleutospores on Senecio Jacobaea, S. palustris, 8. silvaticus, S. viscosus, S. vulgaris, all through the year; also recorded for S. pulcher, S. sarracenicus, etc., in Botanic gardens, and on cultivated Cinerarias (i.e. Senecio) at Sydenham; see Journ. Roy. Hort. Soc. 1908, xxxiii. 511. Very common. (Figs. 241, 242.) Fig. 241. C. Senecionis. a, chain of uredospores ; b, two fascicles of leaves of P. silvestris, bearing peridia on one leaf only of each fascicle (reduced). Fig. 242. C. Senecionis. Teleuto- spore germinating. This is the species of Coleosporium whose life-history has been longest known ; Wolff first experimentally demonstrated it in 1872, and he was followed by Plowright in 1882, as well as by Cornu, Hartig, Rathay, Von Thiimen, Rostrup, Klebahn and Fischer. The secidium had previously been called Peridermium. Since Senecio vulgaris continues to live through the winter in our climate, and Magnus and others have found the uredo- spores throughout the year, the intervention of the alternate host is not in this case necessary. Yet it is generally quite easy to find the Peridermium on the needles of P. silvestris if one searches in June any trees that may be growing in the neighbourhood of Groundsel infected with the parasite ; since the Peridermium is not conspicuous, it may easily be overlooked unless special search is made. It is probable that there are several biological races of this fungus, on different species of Senecio, and it has been proved by Fischer that it cannot be transferred to Cacalia or Sonchus. G. U. 21 322 COLEOSPORIUM Since there are other so-called " " of Peridermium on the leaves of /'. silvestris, which are morphologically not distinguishable from thai belonging to this species, it is always advisable, when such a one is found, to look on the possible hosts in the neighbourhood for the corresponding Coleosporium. Bui the Peridermium found on the bark of Scots Pine is lly distinct, both morphologically and biologically, although Wolff and Plowrighl recorded them as identical. Plowright, however, failed to infect S. vulgaris by spores from "a specimen of ./.'-•. Pini on the bark of a young fir-branch" i.e. pine-branch) — naturally enough ; and he also puts on record I.e. p. 250 his frequenl failures to infect the Groundsel with spores from secidia ./-.'■. Pini var. acicola which seemed to him to be like those with which he succeeded. Ili> consequent suspicion, thai "there must be more than one species included under this name.'" is now abundantly continued. He was experimenting, in these latter cases, with secidia belonging to some of the species of Coleosporium mentioned in the following pages. In North America, C. Senecionis has been found on S. vulgaris, apparently in one locality only (Rhode Island . probably introduce! from Europe ; the Peridermium was not observed. In that quarter of the globe there are many indigenous species, biologically resembling ours, but mostly LEOSPORIUW Uredospores. Sori scattered, orange, at first covered by the epidermis, soon pulverulent : spores ovate or ellipsoid, densely and evenly verruculose, 21 — 34 x 14 — 21 p. Teleutospores. Sori forming little red crusts; spores pris- matic, 1 « ■ i i ■_; 1 1 i up to 100 (x, breadth IS — 24^: epispore up to 14^ thick at the summit. Jv-idia on loaves of l'i ims sih-,-st ,-is\ uredo- and teleutospores on Peta&ites officinalis, August — November, not uncommon. The life-cycle has been demonstrated by Fischer and Wagner for P. officinalis : the parasite may also extend to other species of Petasites. Distribution : Europe. 4. Coleosporium Sonchi Lev. Credo Sonchi Pers. Syn. Fung. p. 217. Coleosporium Sonchi Lev. Ann. Sci. Nat. 1847, p. 373. Cooke, Handb. p. 521 ; Micr. Fung. p. 218, pi. 8, f. 178, 179. Plowr. TJred. p. 250 p.p. Sacc. Syll. vii. 752. Fischer, Ured. Schweiz, p. 453. Peridermium Fischeri Kleb. Zeitschr. f. Pflanzenkr. v. 13. JLcidia (P. Fischeri) and spermogones like S per i nog ones uEcidiospores those of neighbouring species : spores ellipsoid or polygonal- roundish, verrucose, 25 — 32 x 18 — 25 /x. Fig. 244. C. Sonchi. On S. arvensis. Two chains of young uredospores ; teleutospore, showing fusion-nucleus ; aud uredospore. COLEOSPORIUM 325 Uredospores. Sori small, round or oblong, girt by the epidermis, in irregular groups, especially along the veins, yellowish-orange ; spores roundish-oval, densely and finely verruculose, 18 — 25 x 15 — 21 /z. Teleidospores. Sori forming small, flat, roundish, waxy-red crusts ; spores prismatic, length reaching 100 fi, breadth 18 — 30 fi ; epispore at summit reaching 18 p, thick. ^Ecidia on leaves of Piiuts silvestris; uredo- and teleuto- spores on Sonchus arvensis, 8. asper, S. oleraceus, August — November, not uncommon. (Fig. 244.) Klebahn, Fischer, and AVagner have demonstrated the life-cycle of this parasite; the two former have also proved that it cannot be transferred to Cacalia, Campanula, Senecio, or Tussilago. Fischer records this species on Sonchus palustris as well as on the hosts given above. Distribution: Europe. 5. Coleosporium Cacaliae Fckl. Uredo Cacaliae DC. Flor. fr. vi. 65. Coleosporium Cacaliae Fckl. Synib. Myc. p. 43. Cooke, Micr. Fung. p. 218. Fischer, Ured. Schweiz, p. 446. C. Sonchi Plowr. Ured. p. 250 p.p. Peridermium Magnusianum Fisch. Contrib. etude du genre Coleo- sporium, in Bull. Soc. Bot. Fr. xli. 171. P. Magnusii Wagner, Zeitschr. f. Pflanzenkr. vi. 9 and viii. 257. L J . . ** [ Resembling those of other Coleosporia.] AjCidwspores ) Uredospores. Sori numerous, roundish, at first covered by the epidermis, at length pulverulent, orange ; spores ellipsoid; verruculose, deep-yellow, 24 — 35 x 21 — 24 /jl. Teleidospores. Sori hypophyllous, forming flat, red, waxy crusts; spores prismatic, up to 140 jx long, 18 — 25 /x broad, thickened (up to 28 //,) at the summit. [iEcidia on leaves of Pinus montana, perhaps also of P. silvestris;] uredo- and teleutospores on Cacalia hastata, G. suaveolens. The spermogones and gecidia have not been observed in Britain ; the other spore-forms at Bath (Rev. J. E. Vize), 326 COLEOSPORIUM Batheaston (C. \i. I>ni<»me), Oxford Botanic (J aniens (Herb. Bloxam). October. This is, of course, an introduced species. Distribution : Europe. 6. Coleosporium Rhinanthacearum LeV. Coleosporium Khincmthacearum LeV. Ann. Sci. Nat. 1847, p. 373. Cooke, Eandb. p. 521; Micr. Fung. p. 218, pi. 8, f. 17'".. 177. C. Euphrasiae Plowr. CTred. p. 252 S:\n-. Syll. vii. 754. This species is now divided into two, but on purely bio- logical grounds : no morphological distinctions worthy of the name can be discerned, and since the hosts are all closely allied it is perhaps best to retain the collective name; at any rate for a time. (1) Coleosporium Euphrasia: Wint. Credo Euphrasiae Solium. PI. Sail. ii. 230. Coleosporium EuphrasiaeWinter, Pilze, p. 246. Fischer, Ured. Schweiz, p. 442. Peridermium Stahlii, Kleb. Zeitschr. f. Pflanzenkr; ii. 269, pi. 5, f. 5. S 1 ic r liiononcs ) ' ' y Like those of the allied species. zEcidio- spores oval or roundish, 20—30 x 15 — 24 p. Uredospores. Sori small, scattered, roundish, flat, }Tellowish- red ; spores irregularly polygonal, densely verruculose, 20 — 24 x 14 — 17 fju] epispore thin, colourless. Teleutospores. Sori small, flat, roundish, red; spores pris- matic, orange, up to 105 /u. long, 18 — 24^ wide; epispore at summit about 14 p thick. xtEcidia on leaves of Pinus silvestris; uredo- and teleutospores on Euphrasia officinalis, Bartsia Odontites, Rhinanihus Grista- galli, July — September, very common. It is not certain that Bartsia can be infected from Euphrasia or Rhinanthus ; no experiments on that point are recorded. Klebahn proved abundantly that the parasite can be transferred from Rhinanthus to Euphrasia, but not to Senecio, Sonckus, or Tussilago. Wagner 'Zeitschr. f. Pflanzenkr. viii. 261) infected Euphrasia with secidiospores from Pinus montana. COLEOSPORIUM 327 (2) Coleosporium Melampyri Karst. Uredo Melampyri Rebentisch, Flor. Neomarch. p. 355. Coleosporium Melampyri Karst. Myc. Fenn. iv. 62. Fischer, Ured. Schweiz, p. 440, f. 269. Peridermium Soraueri Kleb. Zeitschr. f. Pflanzenkr. iv. 194. The only apparent differences from C. Euphrasiae are in the size of the spores : uredospores 24 — 35 x 21 — 28 //, ; teleutospores as much as 115 fi long, 21 — 28 p wide; epispore very thick (up to 28 /x) at the summit. iEcidia on leaves of Pinus silvestris ; uredo- and teleutospores on Melampy- r a in arvense, M. pratense and its var. montanum, July — September, not uncommon. (Fig. 245.) Wagner records the secidium also on P. montana. Klebahn has demonstrated that the spores of this species will not infect Euphrasia, Rhinanthus, or Campanula. I have not seen the thickening on the summit of the teleutospores so pronounced in ours as in the continental specimens, possibly because they were not so mature. Since the uredo-hostsof C. Rhinanthacearum are all annual and die at the approach of winter, it would seem probable that fresh infections must occur each year from the ajcidium, but as this is, at any rate, not commonly found, the secidiospores must be widely distributed by the wind; it is very possible, however, that the fungus winters in some other manner as yet unsuspected, or that the secidia are mi >re abundant than is thought to be the case. They should be searched for in May and June. There is here great scope for experimental research, especially since young pot-plants of Pinus can be used for infection. Klebahn placed such a Pine amongst a clump of Melampyrum, strongly infested with Col. Melampyri, and left it from July to September (the pot sunk in the earth); in the latter month spermogones appeared and the Becidium {Peridermium Soraueri) in the following spring. Fig. 245. C. Melampyri. a, teleutospore on M. pratense, Randan Woods; b, teleutospore on the same, gathered at Bonn, Germany. Distribution : Europe. 32s < Ol KOSI'OHU'M 7. Coleosporium Campanulas L6v. Uredo Campanvlae Pers. Syn. p. ^17. Coleosporium Campanulae, LeV. Ann. Sci. Nat. ser. 3, viii. 373. ke, II 'IhIIi. p. 521 ; Micr. Fung. p. 218j Plowr. Ored. p. 251. Syll. vii. 753. Fischer, Ured. Schweiz, p. 44:i Arthur, X. Anii-r. Flor. vii. 88. /' ridermium oblongisporium Fckl. Symb. Myc. p. 42. Rostr. Bot. Tidsskrift, xix. pp. 40, II, 19. /'. Ro8trupii Fischer in Bull. Soc. Bot. Fr. xli. p. clx.xii. Spemnogones. Amphigenous, scattered, conspicuous. J'k'ldiuspure.s. .Ecidia (Peridermiwm oblongisporium) like those of allied species, but spores distinctly oblong, 34 — 40 x 18— 22/z(?). OOQ Fiji. 240. C. Campanulae. 0. On C. rotundifolia. Uredospores. Sori hypophyllous, scattered, often confluent, covering the whole leaf-surface, also on the stems, roundish or irregular, covered by the epidermis for some time, then naked, yellow or yellowish-red, fading to whitish: spores ellipsoid to oblong, subangular, strongly and rather densely verruculose, orange, 21 — 35x14 — 21 /*. Teleutospores. Sori hypophyllous, subepidermal, scattered, often confluent, forming small, flat, red, then pale-brownish crusts; spores prismatic, up to 100 fi long, 21 — 28 p, wide; epispore hyaline, reaching 35 fi at the summit, but thin elsewhere. OCHROPSORA 329 iEcidia on leaves of Pinus silvestris; uredo- and teleutospores on Campanula glomerata, C. latifolia, C. rotundifolia, C. Tra- chelium, C. turbinata, end of July — October, not uncommon ; sometimes abundant on cultivated species of Campanula, in both uredu- and teleuto-stages. (Figs. 246, 247.) Some of the forms on the various species are separated by Klebahn as biologically distinct, but the specialisation is not in any case sharply fixed (Wirtswechs. Rostp. pp. 365 — 9). C. Campanxdae is also recorded, in other countries, on many other species of Campanula and on Phyteuma and Specularia. In North America the secidia are found on Pinus rigida but, as in this country, are much less common, or at least less frequently observed, than the spore-forms on Campanulaceae. On C. rotundifolia, this species and Puccinia Campanulae may be found in company. Distribution : Europe, North America, China, Japan, East Indies. OCHROPSORA Dietel. yEcidia with cup-shaped peridium. Uredospores solitary, on pedicels. Teleutospores united loosely into waxy crusts, club-shaped or cylindrical, not thickened above, dividing as they mature into four superimposed cells. This genus is not closely allied to Goleosporium. Tt is, indeed, doubtful whether the character upon which the Coleo- sporiacese are united into one group, viz. the internal basidium, is really an indication of close affinity. There can be little doubt that Chrysopsora, which also has an internal basidium, belongs to the Pucciniacese, and not to the Coleosporiaceae. Ochropsora Sorbi Diet. jEcidium leucospermum DC. Flor. fr. ii. 239. Cooke, Handb. p. 536; Micr. Fung. p. 194, pi. 1, f. 4—6. Plowr. Ured. p. 269. Ochropsora Sorbi Dietel, Ber. Deutsch. Bot. Gesell. 1895, xiii. 401. Fischer, Ured. Schweiz, p. 455. Spermogones. On the foliage leaves (loosely spread over the whole upper surface) and even on the sepals, whitish, then brownish. 330 (K'HKOI'SOKA JUcidiospores. .Kcidi.i scattered pretty regularly over the lower surface of the leaves, not very crowded, shortly cylindrical, white, with born revolute margin; spores irregularly oblong, colourless, thin- walled, very delicately verruculose, 18— 30 x 15— 21/*. [Uredospores. Sori hypophyllous, small, roundish, scattered, not more than \ mm. diam. ; spore-mass grey- ish or yellowish-white, surrounded by a circle of paraphyses. which form a kind of peridium, but their upper ends, when mature, are free and sub- clavate; spores subglobose to ovate, pale-brownish, distantly verrucose, 25— 28x18— 25 /u; epispore 1 — l^yu, thick, with no perceptible germ-pores. I i. . 248. 0. Sorbi. Meid- iiim leucospermum. a.secidia on Leaf of .1. nemorosa, nat. size : b, the same, x2 ; c, secidiospores, x 600. Fig. 249. 0. Sorbi. Section of teleuto-sorus, before the division of the spores * into four cells (one spore is shaded) ; a, basidiospores. (After Fischer.) Teleutospores. Sori hypophyllous, ^— \ mm. diam., at first covered by the epidermis, pustulate, pale flesh-colour, roundish or oblong, clustered in groups; spores cylindrical, rounded above, crowded, grey, granular, subopaque, 70x10 — 18 fi, at length divided into four cells; basidiospores fusiform, 22- 25 x 8 p.] iEcidia on Anemone nemorosa, April — June, not common, Oxford, Cambridge, Devon, North Wales, Yorkshire, Scotland, etc. [Uredo- and teleutospores on Pyrus Aucuparia, August and September, not yet observed in Britain.] (Figs. 248, 249.) ZAGHOUANIA 331 The description of the uredo- and teleutospores is after Fischer ; the latter mature in autumn and germinate at once. According to him, the mycelium of the recidial stage is perennial in the rhizome, but Klebahn proved (Zeitschr. f. Pflanzenkr. 1907, p. 144) that the teleutospores infect the growing points of the rhizome in autumn, and produce the tecidia in the following spring. He could also infect other species of Pyrus {Aria, torminalis, scandica, Malus) from the Anemone ; Fischer did the same for P. fennica and P. com run nix. In the Anemone nearly every leaf of the affected plant will be attacked, as well as the flower-shoots. The leaves become longer, narrower and of a paler green, and are borne on longer petioles. They are often divided into more segments than the normal leaves. Fischer remarks that, wdien the fungus appears on the sepals, the cells in the neighbourhood develop chloroplasts. The discovery of this hetercecism was due to Dietel and was confirmed by Klebahn ; previously the gecidium has been mistakenly attributed, by Soppitt, to Endophyllum as E. leucospermum. The mode of germination of the spores will, of course, easily distinguish them : about this there seems to have been some misapprehension — Soppitt (Journ. Bot. 1893, p. 274) distinctly stated that the spores did not germinate with "promy- celial " spores, but in the Trans. Brit. Myc. Soc. i. 84, 98, this is altered into the statement that the spores " germinate as do those of the (other) Endophytic/,." Puccinia fusca lives upon the same plant (Anemone nemorosa) but affects its host in a diffei'ent way (see p. 215). uEcidimn punctatiim, a stage of Puccinia Pmni-spinosae, is found on A. nemorosa as well as on garden Anemones (A. coronaria), but differs in the character of the broader peridium and in having faintly coloured spores. Moreover, its spermogones are dark-coloured and are found on both sides of the leaf (see p. 207). ZAGHOUANIA Patouillartl. Sori erumpent, subpulverulent. /Ecidia (Peridermium) with a peridium which is irregularly lacerate above ; margin slightly involute ; spores in short chains, soon seceding. Sper- mogones flask-shaped, with ostiolar filaments. Uredospores pedicellate, solitary. Teleutospores one-celled, ovoid, pedicel- late, with a slightly thickened hyaline and verruculose epispore, germinating as soon as mature ; basidium four-celled, semi- internal ; basidiospores nearly sessile. The description is founded upon that given by Dumee and Maire in Bull. Soc. Myc. Fi\, 1902. The semi-internal basidium is characteristic. Dumee and Maire make a separate family, •1.) J ZAGHOUANIA Zaghouaniacese, for its reception and compare it bo Septobasi- dium among the Auriculariese; it seems more likely that Zaghouania will be proved to belong to the PucciniacesB. Zaghouania Phillyreae Pat. Jlridrinn Phillyrecu DC. Flor. fr. vi. 96. Sacc. Syll. vii. 807. Massee, Journ. Hot. 1908, xlvi. L53. Trans. Brit. Myc. Soc. iii. 123. ./-.'. crass nm var. Phillyreae Cooke. Handb. p. 539. Uredo Phillyreae Cooke, Exsicc. i. ~>U2. Plowright, Ured. p. 258. Sacc. Syll. vii. 856. Zaghouania Phillyreae Pat. Bull. Soc. Myc. Fr. 1901, xvii. 187. Duinee et Maire, Bull. Soc. Myc. Fr. 1902, xviii. 23. Spermoc/ones. Flask-shaped, with well-developed ostiolar filaments; spermatia ovoid, hyaline, 4 — 5 x 2 — 3 /x. J-j: id \i i is pores. zEcidia nu- merous, densely crowded, semi- immersed, with a more or less involute, nearly entire, whitish margin : spores variable, glo- bose, ovoid or oblong- pyriform, alveolate-reticulate, deep or- ange-yellow, 20—30 x 14— 18 p. Uredospores. Sori hypo- phyllous, usually on yellowish spots, solitary or aggregated, subangular, at length naked, orange-yellow ; spores globose to ovoid, echinulate, orange-yellow, 24 — 32 x 12 — 16 /la. [Teleutospores. Mixed with the uredospores, oblong-ovoid, 45 — 65 x 15 — 18 /m, with a rather thick, hyaline, verruculose epi- spore ; basidiospores sessile, smooth, subglobose, 12 — 14//. diam.] . Kcidia on leaves and young shoots of Phillyrea latifolia, Pevensey Churchyard, August, 1907 (G. Massee); on Phillyrea media, near Chichester, ascidia, 1869, uredo, April, 1874 (F. Paxton). The uredospores have been gathered in Italy as early as February. (Fig. 250.) Fig. 250. Z. Phillyreae. a, uredb-sori on leaf of P. media, Chichester, nat. size; 6, uredospore from the same; c, teleutospore beginning to germi- nate (after lhimee and Maire). ENDOPHYLLACE^E 333 The fungus in the cecidium-stage forms rounded swollen pustules on the leaves or extensive patches on the stems. Every shoot of the year is usually attacked and contorted, and in August is covered and made conspicuous by a copious development of the orange spores. The teleuto- spores have not been detected in this country. They are remarkable for their mode of germination — the basidium is formed internally as in Colcosporium ; this then emerges through a rupture at the base of the spore and produces its large nearly sessile basidiospores externally, one from each cell. Distribution : France, Germany, Italy, Corsica, Tunisia, Algeria. ENDOPHYLLACE^E Teleutospores in long chains, surrounded by a peridium, which is formed like that of a typical secidium of P actinia from the peripheral cell-rows, but is sometimes less strongly devel- oped ; spores separated by intercalary cells, produced from a fusion-cell, but germinating as soon as mature by a typical basidium and basidiospores ; germ-pores not perceptible ; spore- wall coloured, verruculose. Spermogones present ; both kinds of sori subepidermal. ENDOPHYLLUM Leveille. Characters of the genus the same as of the family. 1. Endophyllum Euphorbiae-silvaticae Wint. JEcidium Euphorbiae-silvaticae DC. Flor. fr. ii. 241. 2E. Euphorbiae Pers. ; Cooke, Handb. p. 537 ; Micr. Fung. p. 195 p.p. Endophyllum Euphorbiae-silvaticae Winter, Pilze, i. 251. Sacc. Syll. vii. 767. Fischer, Ured. Schweiz, p. 437, f. 298. E. Euphorbiae Plowr. Ured. p. 228. Spermogones. Epiphyllous, or a few amongst the secidia, yellowish, then brown. 334 i:\lHiNIVLLUM Teleutospores. Sori hypophyllous or occasionally epiphyllous, more or less covering the whole surface, crowded, aecidium-like, sunk in tlif leaf-tissue which is slightly swollen, surrounded by a thin peridium in the shape of a shallow cup, with a short, Fig. 251. E. Euphorbiae-silvaticae. a, affected leaf of E. amygdaloid.es, nat. size; b, aecidio-teleutospore ; c, germinating spore; d, another, not so advanced, x 540 ; e, inner, and /, outer wall of peridium-cell, x GOO. torn, slightly revolute margin; spores in evident chains, bluntly polygonal or subglobose, densely but finely verruculose, orange, 18 — 23/t; epispore about 1/x thick; basidiosp' »res i >bovate, yellow. On Euphorbia amygdaloides (= E. silvatica). Rather un- common. April — June. (Fig. 251.) The mycelium usually perennates in the plant. The affected shoots are taller than the normal ones, with shorter, wider and paler leaves; they seldom flower. This species externally closely resembles an tecidiuru, and can be distinguished from one only by the different mode of germina- tion of the spores. The secidia which occur on other species of Euphorbia belong to different life-cycles, Uromya s Pisi etc. The peridium of this species is more strongly developed than that of the following one ; its cells are densely warted, and arranged in distinct rows. A long account of the behaviour of the parasite is given by Plowright (I.e.). The spores germinate readily in the summer as soon as produced. Distribution : Europe. ENDOPHYLLUM 335 2. Endophyllum Sempervivi De Bary. Credo Sempervivi A. et S. Consp. p. 126. Endophyllum Sempervivi De Bary, Morphol. p. 304 (sec. Saccardo). Cooke, Handb. p. ">4(i ; Micr. Fung. p. 200. Plowr. Ured. p. 229. Sacc. Syll. vii. 767. Fischer, Ured. Schweiz, p. 436, f. 297. Hoffmann, Centralbl. f. Bakter. xxxii. 137, f. 1 — 14 and pi. i, ii. Fig. 252. E. Sem- pervivi. Sori on Sempervivum mon- tanum (reduced). Spermogones. Scattered amongst the secidia, roundish. Teleutospores. Sori rather large, amphi- genous, sunken in the leaf, secidium-like, sur- rounded by many layers of hyphse and by a peridium, which opens by a pore at the summit and finally becomes cup-shaped ; spores bluntly polygonal or roundish, }^ellowish-brown, densely reticulate-ver rucose, 24 — 35 x 21 — 28//.; epispore 3 — 4^ thick. On Sempervivum tectorum ; also found on S. calcareum, S. globiferam, S. montanum (Plowright), S. arachnoideum (Fischer) and others (Saccardo). Not common, Warwick- shire, Forclen, Kew Gardens, etc. April — August. (Figs. 252, 253.) It has been proved by De Bary, Hoffmann and others that the basidiospores produced by the teleutospores infect the leaves, and from them arises a mycelium which perennates in the stem. It produces spermogones and teleutospores in the following spring. The affected leaves are more erect than normal ones, twice as long, narrower and yellowish at the base : infested plants should be burnt, so that they may not infect others. See the fuller account given on pp. 53-5. Distribution : Europe. Fig. 253 E. Semper- vivi. ^Ecidio-teleu- tospore germinating (after Hoffmann). :;:;.; MELAMPSORA( I.I MKLA.MPSORACK.K Teleutospores not pedicellate, but seated on a dilated hyphal cell, produced singly in the tissues of the host or com- pacted side by side into flat crusts, one-celled or divided longitudinally into 2 — 4 cells. Germination by an external basidium, with minute round basidiospores (about 10 ^). Uredo- spores abstricted singly. Uredo-sori and secidia with or without a peridium. / MelniiipSoru.. Melampsoridium. M ii a.mi'surk.k. Teleutospores brown or brownish, Mrl±l ; Micr. Fung. ]>. 219. M. farinosa Schrot. Flor. Settles, iv. 360. Plowr. CTred. p. 238 p.p. see oote . M. Larici-Caprearum Kleb. in ForstL-naturw. Zeitschr. lw:»7. \>. !';'•». Fischer, CTred. Schweiz, p. 183, f. 312. /Ecidiospores. Caeomata minute, pale-orange; spores round- ish, oblong, or polygonal, 15 — 25x12 — 17/^: epispore up to 2 fjb thick, finely verruculose, with many thin places (?germ- pores). Fig. 254. M. Larici-Caprearum. Para- pbysis and uredospores (one showing the thin places in the epispore) ; b, teleutospores. On .S'. Gaprea. Fig. 255. M. Larici-Caprenrum. Teleuto-sori, on upper face of leaf of .S'. Gaprea, nat. size ; teleuto- spore, germinating, x 360. Credos-pores. Son hypophyllous, showing as yellow spots on the upper side, variable in size and arrangement, 1 — 3 mm. wide; spores roundish, oval, or polygonal, 14 — 21 x 13 — 15^; epispore 2 — 2^ /x thick, firm, distantly echinulate, with thin places (? germ-pores) ; paraphyses capitate, 50 — 60 x 18 — 26 fi, thickened (up to 5/x) above. Teleutospores. Sori epiphyllous, 1 mm. or more wide, dark reddish-brown, frequently confluent in extensive crusts, lying between the cuticle and the epidermis ; spores prismatic, rounded below, 30 — 45x7 — 14 ll, rather unequal in length: epispore clear-brown, thin (1 ll), but thickened (up to 10/a) ON WILLOW 339 above where it is pierced by an evident laterally placed germ- pore. Caeomata on Larix europaea ; uredo- and teleutospores on Sulix Caprea, more rarely on S. aurita and other Salices. The commonest species. (Figs. 254, 255.) The teleutospores germinate the following spring. They are distin- guished from those of all the allied species by being thickened above. Plowright remarks that the ca^oma is not uncommon early in the year on Larch foliage, but is very inconspicuous and easily overlooked ; he found the oecidiospores, in company with the uredospores on S. Caprea, at West Malvern, June, 1900. But it is almost impossible to say, without experi- ment, to which form of Melam psora any given caeoma on Larch is to be assigned. Plowright's M. farinosa seems to be chiefly this species, but several of the allied forms are continually recorded under the same name. The reddish teleuto-sori on the upper side of the leaf, are distinctive and are easily found by looking for them from September onwards. It may be mentioned here that both uredo- and teleutospores of the Melampsoras on "Willow and Poplar germinate readily : if the germinating uredospores are placed upon healthy leaves and kept in a damp chamber, infection usually follows in 7 — 10 days. 2. Melampsora Euonymi-Caprearum Kleb. Uredo confluens var. Euonymi Mart. Flor. Mosq. p. 230. Cooke, Handb. p. 527. U. Euonymi Cooke, Micr. Fung. p. 216. Caeoma Euonymi Plowr. Ured. p. 260. Melampsora Euonymi-Caprearum Kleb. in Pringsh. Jahrb. f. Wissensch. Bot. 1900, xxxiv. 358. Fischer, Ured. Schweiz, p. 489. Sperniogones. Flatly pulvinate. ^Ecidiospores. Ca3omata mostly hypophyllous, bright-orange, in elongated clusters on orange spots, \ — 1 mm. diam.; spores oval, rarely oblong, 18 — 23 x 14 — 19//,; epispore thick (up to 5 p.), finely and densely verrucose. Uredospores. Sori hypophyllous, on discoloured spots which show distinctly on the upper side, small, \ mm. wide, pulvinate, single or in groups; spores mostly roundish, orange, 14 — 19 x 14 — 17/z, distantly echinulate without smooth spots; epispore thin, or at times thickened (up to 4 jx), with several germ- 22—2 340 MELAMPSORA pores paraphyses capitate with a slender pedicel, thickened (up to 8 p) above, 50—70 x 18—25 p. Tr/cutusjxires. Sori hypophyllous, covered liy the epidermis, small, about h mm, diam., but united into groups bounded by the veins, brown with a tinge of bluish-grey ; spots brown on the upper surface; spores irregularly pris- matic, rounded at both ends, 25 — 40 x 7— 13 p, ; epispore thin, clear-brown, scarcely thickened above, with a barely perceptible apical germ-pore. Cseomata on Euonymus europaeus, August, September, rather rare; uredo- and teleutospores on Salve aurita, S. Caprea, S. cinerea. (Fig. 256.) Fig. 256. M. Euonymi- < 'aprearum. Old dead leaf of ,S'. Caprea, show- ing teleuto-sori on lower face (slightly reduced). Fischer records the cseoma in Switzerland in May and June. The distinction of this species from the preceding one (apart from the secidial host) lies in the teleuto-sori ; these are hypophyllous and subepidermal, while those of M. Larici-Caprearurn are epiphyllous and subcuticular. It is not certain, however, that this distinction is absolute ; Fischer in a culture on S. cinerea obtained a few teleuto-sori showing also the latter characters. There is a further difference in the form of the teleutospores. 3. Melampsora Larici-epitea Kleb. (emend. Fischer). Uredo epitea K. et S. Mycol. Heft. i. 68. Lecythea epitea Lev. ; Cooke, Micr. Fung. p. 221. Melampsora epitea Thiim. ; Plowr. Cred. p. 239. M. Larici-epitea Fischer, Ured. Schweiz, p. 485, f. 313. Kleb. in Zeitschr. f. Pflanzenkr. 1899, ix. 88. jEcidiospores. Cseomata hypophyllous, scattered or in rows, with yellow spots on the upper surface, roundish or oblong, | — 1^ mm. long, pale orange-yellow ; spores roundish or some- what polygonal, finely warted, 15 — 25x10 — 21 p,: epispore H — 3,11 thick, with no recognisable germ-pores. Uredospores. Sori amphigenous, seated on yellow spots, orange-yellow, | — l|mm. diam.; spores mostly oval, sometimes oblong, roundish, or angular, echinulate, orange, 12 — 25 x 9 — ON WILLOW 341 19 fi; epispore lt|- — 3^/x thick, without perceptible germ-pores; paraphyses capitate, with a thin pedicel, occasionally clavate, thickened (up to 10 /x) above, hyaline, 35 — 80 x 15 — 24 //. Fig. 257. M. Larici-epitea. Caeoma on Larch leaf, x 2 (one of Plowright's cultures from S. tin erect). Fig. 258. M. Larici-epitea. Uredospores and para- physes, on S. viminalis. Teleutospores. Sori amphigenous, covered by the epidermis, at length dark-brown, sometimes tinged round with greyish- blue or violet, \ — 1 mm. diam., densely clustered or confluent ; spores prismatic or subclavate, rounded at both ends, occasion- ally narrowed above, 20 — 50 x 7 — 14 p, ; epispore clear-brown, uniformly thin, without an evident germ-pore. Cseomata on Larix europaea ; uredo- and teleutospores on many species of Saline. The teleutospores in October and November. (Figs. 257, 258.) This species is one of the, most complex of all the Melampsoracese ; it has been divided into about eight biological races, all of which have their cseoma on Larch, while the other spore-forms are specialised to certain species of Willow. It is recorded on the following British Salices for other countries, but it is not possible as yet to say on which of these it is found in this country : IS. aurita, Caprea, cinerea, fragilis, Smithiana, viminalis for the typical form ; nigricans for the form M. Larici-nigricantis Schneider ; purpurea for the form M. Larici-purpureae Schneider ; and reticulata for M. Larici-reticulatae Schneider. Morphological differences between these forms are not discoverable. The teleutospores germinate in the spring following after their forma- tion. The teleuto-sori are formed under the epidermis, and appear to start usually just below a stoma ; they often occupy entire areas bounded by the veins. This description is taken mainly from Fischei1. The teleutospores are distinguished from those of M. Larici-Caprearum by the total want of thickening at the apex. Plowright produced the cseoma on Larch from the teleutospores on Salix cinerea, May 24th, 1900. .'!42 MELAMPSOB A 4. Melampsora Ribesii-viminalis Kleb. Curm/Ki /u'/ifsii Link, Sp. PL ii. i!s p.p. Melampsora Ribesii-viminalis Kleb. in Pringsh. Jahrb. f. Wissenscn. Bot. L900, w\iv. 363. Fischer, lied. Schweiz, p. 194. Spermogunes. Pulvinate, with flat hymenium. JEcidiospores. Caeomata hypophyllous, on discoloured spots which show on both sides, arranged in groups, 15 una. wide, bright-orange; spores roundish, rarely oval, 18 — 23 x 14 — 17 p.; epispore 2 — 4 ll thick, with many thinner places (? germ-pores), finely and densely verruculose. Uredospores. Sori hypophyllous, minute, about \ mm., scattered or in groups, pale orange-yellow ; spores more or less round, 15 — 19x14 — 16 p,\ epispore 2/x thick, uniformly echinu- late ; paraphyses capitate or more often clavate, hardly thick- ened above, 50 — 70 x 18—25 fi. Teleutospores. Sori epiphyllous, developed between the cuticle and the epidermis, \ — £ mm., scattered or in groups over the whole leaf, dark-brown, shining ; spores prismatic, rounded at both ends, more or less irregular, 25 — 40 x 7 — 14 /x ; epispore thin, clear-brown, not thickened above, with no evident germ-pore. Cseomata on Ribes Grossidaria, R. nigrum, R. rubrum ; uredo- and teleutospores on Salix viminalis. Klebahn showed, by many trials, the genetic connection of the parasites on these species. The teleutospores can be distinguished from all others except those of M. Larici-Caprearum and J/. Allii-Galanthi-fragilis by their being subcuticular. M. Larici-epitea is also recorded for Salix viminalis, but that has its teleutospores subepidermal, and the paraphyses of the uredo-sori strongly thickened at the apex. The cseoma on Ribes does not seem to be recorded for this country and is apparently rare. 5. Melampsora Ribesii-purpureae Kleb. Melampsora Ribesii-purpureae Kleb. in Pringsh. Jahrb. f. Wissensch. Bot. 1901, xxxv. 667. Fischer, Ured. Schweiz, p. 4(.t± Spe7"mogones. Subcorneal, with flat hymenium. j-Ecidiospores. Cseomata mostly hypophyllous, on pale-yellow spots which show on both sides, scattered or in subcircinate groups, ^ — 1^ mm., sometimes confluent, surrounded by the ON WILLOW 343 torn epidermis, orange ; spores roundish or subpolygonal, 18— 20x15 — 18 /x\ epispore about 3 p. thick, finely and densely verruculose, with a few thinner places (? germ-pores). Uredospores. Sori mostly hypophyllous, on conspicuous bright-yellow spots which show on both sides, \ — 1^ mm., pulvinate, surrounded by the torn epidermis, bright orange-red ; spores roundish, uniformly but not densely echinulate, 15—23 x 14 — 19 /x; epispore up to 2%fi thick, with a few thinner places; paraphyses variable, capitate or clavate, 40 — 70 x 12— 21 fji, not thickened above. Teleutospores. Sori amphigenous, but mostly hypophyllous, scattered or in little groups over the whole surface, subepider- mal, small, I — \ mm. diam., brownish-black ; spores irregular, prismatic, rounded at both ends, 25 — 35x7 — 10/x; epispore thin, clear-brown, not thickened above, without evident germ- pore. Casomata on Ribes alpinum, R. Grossularia (not on R. nigrum, R. rubrum) ; uredo- and teleutospores on Salix pur- purea. Description according to Fischer ; Klebahn's experiments have demon- strated the connection of the parasites on these hosts. It is probable that M. Ilibesii-auritae Kleb. is only a biological race of the same fungus, having spermogones and cseoma-spores (according to Klebahn) on Ribes nigrum as well as on the two mentioned above, and its other spore-forms on Salve aurita and possibly S. Caprea. They are scarcely distinguishable, if at all, morphologically ; both may be British. M. Larici-epitea is also recorded on >S'. purpurea, but can be distin- guished by its densely clustered teleuto-sori on spots often bounded by veins, and the paraphyses of the uredo-sori strongly thickened above. M. mixta (Plowr. Ured. p. 239) may belong to either, but his species ex- tended to the branches and inflorescence. 6. Melampsora Orchidi-repentis Kleb. Uredo confluens var. Orchidia A. et S. Consp. p. 122. Cooke, Handb. p. 527. U. Orchidis Mart. Flor. Mosq. p. 229. Cooke, Micr. Fung. p. 216. Caeoma Orchidis Wint. Pilze, p. 256. Plowr. XTred. p. 261 ; Card. Chron. 1890, viii. 41. Saec. Syll. vii. 868. Melampsora Orchidi-repentis Kleb. in Pringsh. Jahrb. f. Wissensch. Bot. 1900, xxxiv. 369. Fischer, Ured. Schweiz, p. 488. 344 MELA.MI'SOKA Spermogones. Hardly projecting, flat, mostly under the stomata. JEcidiospores. Cseoraata irregular in outline, clustered or circinate oe large pale-yellowish spots, often confluent, 1—2 mm. diam. ; bright orange- yellow: spores roundish-polygonal, 11 — 20 x 11 — 15 jx\ epispore very thin, delicately warted. Uredospores. Sori hypophyllous, minute, deep-orange, causing yellow spots on the upper side: spores more or less round, orange, echmulate, 13 — 17x12 — 14 /x; paraphyses mostly capitate, with thin pedicels, 40 — 70 xlG— 20 fi. Teleutospores. Sori hypophyllous or a few on the upper side, subepidermal, small, dark-brown ; spores prismatic, rounded at both ends, 16 — 48x7 — 14 /a: epispore clear-brown, uniformly thin (about 1 p), without any evident germ-pore. < 'n'omata on IAstera ovata, Orchis latifolia, 0. unirnlata, May, June: uredo- and teleutospores on Salix aurita, S. repens. Not common. (Fig. 259.) The cseorna is also recorded on Listera cordata, Goodyera repens, and Gymnadenia conopsea, but no experimental proof of the connection seems to be forthcoming in the case of these species, though both Plowright and Klebahn have demonstrated it for the other species. The teleutospores germinate, as usual, after the winters rest. Soppitt found the cajoma on 0. maculata and the teleutospores on S. repens growing together near Southport. — The secidium of Puecinia Orchidearum-Pkalaridis can be readily distinguished from the ca?oma of this species by its evident peridium and the regularity of its form. Fig.259. M.Orrhidi- repentis. Caeoma on leaf of Listera oratit. from one of Plowright's cul- tures (reduci d). Melampsora Allii-fragilis Kleb. Uredo Alliorum DC. ; Cooke, Micr. Fung. p. 217 p.p. Cm-oiiw Alliorum Link; Plowr. Fred. p. 261 p.p. Melampsora Vitellinae Plowr. I.e. p. 240 p.p. M. Allii-fragilis Kleb. in Pringsh. Jahrb. f. Wissensch. Bot. 1901, xxxv. 671. Fischer, Ured. Schweiz, p. 481. ON WILLOW 345 Spermogones. Scarcely projecting, with Hat hyrnenium. /Eddiospores. Cseomata on the leaves and stems, also on the bnlbils, clustered on discoloured spots, usually oblong, up to 2 mm. long, surrounded by the torn epidermis, bright orange- yellow : spores irregular, mostly polygonal, seldom round, 18 — 25 x 12 — 19 fi\ epispore 1 — H/a thick, finely verruculose. Uredospores. Sori hypophyllous or partly epiphyllous, minute, ^ mm., circular, surrounded by the torn epidermis, reddish-orange, causing reddish-yellow spots on the upper side ; spores distinctly oblong or obovate, 22 — 33x13 — 15/^; epispore up to 3 jx thick, with thinner places (? germ-pores), distantly echinulate, but smooth and somewhat thinner above; paraphyses capitate or clavate, 50 — 70x15 — 20^, with thin pedicel and uniformly thickened membrane (3—5 fx). Teleutospores. Sori chiefly epiphyllous, between the cuticle and the epidermis, scattered or in groups, pulvinate, \ — H mm. broad, dark-brown, shining ; spores prismatic, rounded at both ends, 30 — 48x7 — 14 //, ; epispore clear-brown, uniformly thick (about 1 fi), without evident germ-pore ; basicliospores orange. Cseomata on Allium Cepa, A. ursinum and others; uredo- and teleutospores on Salix fragilis, S. pentandra, and the hybrid between them. The teleutospores germinate after a winters rest. Klebahn's M. Galanthi-fragilis is morphologically identical, even occurring on the same species of Salix, but has its cseomata on Galanthus nivalis ; it can only be considered as a biological race. The same author's M. Larici-pentandrae infests the same species of Salix, but has its eteomata on Larix ; it is distinguished, however, by its teleuto-sori, which occur on both sides of the leaves, are more minute (though often confluent) and arise below the epidermis. I have a Melampsora on Salix fragilis, which has minute teleuto-sori abundantly on the upper surface and beneath the epidermis. This might belong to M. Lariei-pentandrae Kleb. 8. Melampsora Allii-Salicis-albae Kleb. Melampsora Allii-Salicis-albae Kleb. in Zeitschr. f. Pflanzenkr. 1902, xii. 19. Fischer, Ured. Schweiz, p. 480. Spermogones. Rather flat. sEcidiospores. C*omata on the leaves and stems, in groups 346 MELAMPSORA • Hi yellowish spots, about 1 mm. diam., surrounded by the epidermis, bright orange-yellow: spores mostly roundish, 17— 26 x 15 — 18 fx, densely \erruculose. Uredospores. Sori of two kinds: (1) in summer and autumn on the leaves, hypophyllous, h. mm- vvide, on inconspicuous discoloured spots, (2) in spring, erumpent from the bark of young twigs and as much as 5 mm. long, afterwards on the young leaves, as much as 2 mm. long and densely crowded; spores all similar, distinctly oblong, sometimes clavate or pyri- form, 20 — 36x11 — 17 yu,; epispore 2yu, thick, smooth above, distantly echinulate below : paraphyses capitate, 50 — 70 x 15— 20 fx, not thickened above, absent from the cortical sori. Teleutospores. Sori amphigenous, subepidermal, scattered thinly over the leaf-surface singly or in groups, dark-brown ; spores prismatic, rounded at both ends, clear-brown, 25 — 45 x 7 — 10 /x ; epispore scarcely 1 fx, thick, not thickened above, without evident germ-pore. iEcidia on Allium ursinum and other species; uredo- and teleutospores on Salix alba. The cseoma on Allium is indistinguishable from that of M, Allii- fragilis or M. Allii-populina. This species can winter by its teleutospores which produce the cseoma on Allium, or by the perennial mycelium in the cortex of the branches on which the uredospores appear in spring before the cseoma is produced ; these sori are without paraphyses. The whole of this account is due to Klobahn : I have a specimen from Yorkshire (C. Crossland) which is referred, doubtfully, to this species. 9. Melampsora arctica Rostr. M. arctica Rostr. Fung. Greenland. 1888, p. 535. Sacc. Syll. Fung, vii. 595. Annals of Scott, Nat. Hist. 1911, p. 37. Uredospores. Sori hypophyllous, gregarious, yellow ; spores spheroid to ovoid, echinulate, 18 — 23 //. ; paraphyses clavate. Teleutospores. Sori hypophyllous, scattered, very small, dark-brown ; spores prismatic, reddish-brown. On leaves of Salix herbacea, Scotland, Ben-an-Dothaidh, at 3100 ft. (J. A. Wheldon and A. Wilson). ON POPLAR 347 1 'hints of the Salix were brought from Scotland and cultivated at Walton, near Liverpool ; there were then uredospores only. When the leaves fell, they were left on the ground, and two teleuto-sori were developed on them, of a rufous black colour. I have not seen the specimens. 10. Melampsora populina Lev. Melampsora populina Lev. Ann. Sci. Nat. 1847, p. 375. Cooke, Handb. p. 523 ; Micr. Fung. p. 219. Plowr. Ured. p. 242. Sacc. Syll. vii. 590. (1) Melampsora Allii-populina Kleb. Caeoma Alliorum Link ; Plowr. Ured. p. 261 p.p. M. populina Cooke, Micr. Fung. pi. 9, f. 195, 196. Melampsora Allii-popxdina Kleb. in Zeitschr. f. Pflanzenkr. 1902, xii. 25. Fischer, Ured. Schweiz, p. 504. Spermogones. Pulvinate, projecting. sEcidiospores. Cseomata about 1 mm. wide, mostly in groups on yellowish-white spots on the leaves, surrounded by the epidermis, bright orange-red ; spores roundish or oval or polygonal, 17 — 23x14 — 19 pu; epispore about 2 p, thick, but sometimes thicker and then obviously thin at certain spots (? germ-pores), verruculose. Uredospores. Sori hypophyllous, or even rarely epiphyllous, round, scarcely 1 mm. wide, surrounded by the epidermis, causing yellowish spots, bright reddish-orange; spores distinctly Fig. 260. M. Allii-populina. a, teleutospores ; b, paraphysis ; c, uredospores. On Populw nigra. oblong or clavate, rarely oval, 24 — 38x11 — 18//,; epispore 2 — 4/i thick, with sunken places (? germ-pores), sometimes thicker at one end, but without any equatorial thickening, distantly echinulate, but smooth above ; paraphyses mostly capitate, 50 — 60 x 14 — 22 u, with a wall uniformly 2 — 3 p, thick. 348 MELAMPSORA Teleutospores. Sori hypophyllous, subepidermal, scattered over the leaf singly ;ni:! l>v looking in spring on fallen leaves of /'. tremula, /'. alba, at places where Mercurialis is found to be affected. The caeoma on the latter is very capricious in its occurrence ; in some years it may be found almost every- where, in other years hardly a specimen can be met with. The large yellow spots show conspicuously on the upper surface of the leaves, so that when it does occur it is impossible to overlook it. There are two places near Birmingham where all the spore-forms occur year after year in close proximity : it is from these specimens that the descriptions are taken. In one case the teleutospores are on P. alba, in the other on /'. trenvula. The morphological differences of the four (or five) species of Melam- psora which appear on P. alba and P. tremula are very slight. It is better tn include them all for the present under Melampsora tremulae, as biological races. The fourth species is M. Magnusiana Wagner, which has its ceeomata on Chelidonium, and which, according to Klebahn, is identical with M. Klebahai Bubak, on Gorydalis ; this has not yet been recorded for Britain, but Plowright mentions (I.e. p. 241) a Melampsora in) P. tremula from which he could not obtain cajomata on either Larix or Pimcs or Mercurialis — this might be M. Magnusiana. 12. Melampsora Euphorbiae Cast. Credo Helioscopiae Pers. Disp. Meth. p. 13. Lecythea Euphorbiae Lev. ; Cooke, Micr. Fung. p. 221. Melampsora Euphorbiae Cast. Obs. ii. 18, with plate (1843). Cooke, Handb. p. 523 ; Micr. Fung. p. 219, pi. 9, f. 193, 194. M. Helioscopiae Wint. Pilze, p. 240 (1884). Plowr. Ured. p. 236. Save. Syll. vii. 586. Fischer, Ured. Schweiz, p. 508, f. 318. | S/^rmogones. Flat, hemispherical. Jk-idiospores. Cseomata minute, \ — h, mm. diam. on the leaves, 1 — 4 mm. lung on the stems, yellowish-red; spores in i-'i.'. -'(it. G. U. M. Euphorbiae. a, teleutospores on E. exigua ; b, teleutos-porc, and c, uredospore and paraphysis, on E. Peplus. 23 354 MELAMPSORA short chains, without paraphyses, roundish to ellipsoid, densely verrucose, 21 — 28 x 19— 24/u,.] Uredospores. Son hypophyllous and on the stems, scattered roundish or oblong, soon naked, but surrounded by the epider- mis, golden-brown, mixed with numerous capitate paraphyses; spores roundish, orange. I ~> —17 x 11 — 14/a: epispore colourless, echinulate, without perceptible germ-pores; paraphyses hyaline, with strongly thickened heads, 16 — 23 (m diam. Teleutospores. Sori amphigenous, subepidermal, minute rmindish or oblong, reddish-brown, then black, more or less pulvinate, on the stems often confluent; spores prismatic, 50 — 60 x 10 — 14 fx: epispore brown, thin, at length thickened above (up to 5 /x) and darker. On Euphorbia eodgua, E. Helioscopia, E. Peplus. May- October. Very common. (Fig. 264.) Miiller, on the ground of cultures, considers the form on E. Helioscopia as a separate species from those on E. Peplus and E. exigua, the latter two also being biologically distinct from each other (Centralbl. f. Bakter. 2. xix. 441;. In his experiments he observed a caeoma-stage on E. exigua which the others did not have. This stage seems to be very rarely met with, and has probably not occurred in this country : the description of the spermogoues and cseomata is from Fischer. In the same species, as it occurs on E. Cyparissia.% Jacky states that he obtained the uredospores by infection with the basidiospores without the intervention of the cseoma-stage, so that M. Euphorbias may be in a transition state between a eM-form and a kemi-fovva. The secidium on E. exigua recorded by Plowright (I.e. p. 270) belongs to Uromyces tuberculatus ; see p. 102. Distribution : Europe, Siberia. 13. Melampsora Hypericorum Wint. Uredo Hypericorum DC. Flor. fr. vi. 81. Cooke, Handb. p. 526 : Micr. Fung. p. 215, pi. 8, f. 174. 17-">. Melampsora Hypericorum Winter, Pilze, p. 241. Plowr. Ured. p. 243. Sacc. Syll. vii. 591. Fischer, Ured. Schweiz, p. 506, f. 317. McAlpine, Rusts of Australia, p. 191. jEcidiospores. Cseomata hypophyllous, scattered, roundish or oblong, flatly pulvinate, often very small, subepidermal, long covered, at length erumpent, orange, showing as indefinite pale- yellow or orange spots on the upper face; spores in short MELAMPSORA 355 chains, ellipsoid to polygonal or subclavate, 18 — 28 x 10 epispore colourless, about 2 /x thick, rather densely verruculose, with no perceptible germ-pores; no para- physes. Teleutospores. Sori hypophyllous, subepidermal, small, roundish, red- dish-brown, then dark-brown ; spores prismatic, more or less rounded above, pale-brown, 28 — 40 x 10 — 17 //,; epi- spore thickened (up to 3 p,) above. -18^; Fig. 265. M. Hypericorum. a, teleutospores, under the epidermis ; b, ascidiospore, without paraphyses. On H. Androsaemum. On Hypericum Androsaemum, H. humifusum, H. perfora- tum, H. pulclirum. Not common. May — October. (Fig. 265.) What was described by Plowright as the uredo-stage of this fungus is stated by Fischer, Tranzschel, and others, to be the caeoma stage — the spores " being produced in short chains, with sterile intercalary cells, without paraphyses, but sometimes " (at least on Hypericum montanum) '• surrounded by a layer of swollen colourless cells which might almost be considered as an undeveloped peridium." Midler considers the form on H. montanum as a biological race, since he could not infect other species of Hypericum with spores from it. But Klebahn has proved (Zeitschr. f. Pflanzenkr. 1905, xv. 106) that a species of Hypericum can bear both the cseorna-forrn without paraphyses and the uredo-form with paraphyses. McAlpine (Rusts, p. 192) records that in Australia the uredo-sori have very abundant paraphyses, inter- mixed; he describes them as "hyaline, capitate, overtopping the spores, 50 — 68 x 18 — 24^i." His species was on leaves and stems of H. japonicum, ' and differs slightly from the British ones. His description of the uredo- sori is as follows : " Sori mostly hypophyllous, scattered or gregarious, at first bright-orange, becoming pale, pulverulent, up to \ mm. diam., erumpent and surrounded by the ruptured epidermis. Spores subglobose to ellipsoid, finely verrucose, orange-yellow, 14 — 21x11 — 17 (i, with two germ-pores on one face." On the British specimens which I have seen, there are no paraphyses and the spores are decidedly in chains. Distribution : Europe, Siberia, India. 14. Melampsora Lini Desm. Credo miniata var. Lini Pers. Syn. Fung. p. 216. U. Zi'm'Schum. PI. Sail. ii. 230. Arthur, N. Amer. Fl. vii. 101. 23—2 356 MELAMPSORA /. ythea Lini Berk.; Cooke, Eandb. p. 532; Micr. Fung. p. 222, pi. 8, f. l<;:, t. Melampsora Lini Desm. PI. Crypt, no. 2049. Plowr. [Jred. p. 237. 5 cc. Syll. vii. 588. Fischer, (Jred. Schweiz, p. ~>(>7. McAlpiue, Rusts of Australia, p. L92, f. 236 and pi. I. i. 36. Spermogones ) ^ ^^ mcidiospores J Uredosj i< u-r.--. Sori amphigenous and on the stems, small, scattered, roundish or oblong, flatly pulvinate, subepidermal, (?at first covered by a parenchymatous peridium, Fischer), orange; spores roundish to ellipsoid, echinulate, orange-yellow, 16 — 24 x 12 — 17 /Lt; paraphyses not numerous, hyaline, strongly capitate, much thickened above, 20 — 25 /x diani. Y'il'. 200. M. Lini. a. teleutospores ; b, plan of same ; c, paraphysis and two uredospores. On L. catharticum. Teleutospores. Sori similar, but confluent, chiefly on the st tins, subepidermal, reddish-brown, at length black and shining; spores prismatic, 35 — 60x7 — 10 fM) epispore thin, yellowish- brown, rather thickened above and darker. On Linum catharticum. (Fig. 266.) June — October. Not uncommon. Forms of Melampsora Lini occur widely on many species of Linum and have usually been regarded as identical. That which often acts as a very destructive parasite wherever the common Flax is cultivated has consider- ably wider teleutospores (17 — 20^, McAlpiue) and attempts to infect L. usitatissimum from L. catharticum have uniformly failed; it is therefore considered by some as a biological race or even species = Melampsora liniperda Kornicke (Centralbl. f. Bakter. 1911, 2. xxxii. 278). Teleuto- spores of this have been described as much as 80 /x long. Fromme (1912) has recently described spermogones and a-cidia to this form on cultivated MELAM PSORA 357 Flax ; the spermogones are flask-shaped, subepidermal, without ostiolar filaments, the secidia are small and apparently difficult to distinguish. X" other author seems to have met with these, unless the structures to which Fischer assigns a "parenchymatous peridium" were such secidia. It would keep the disease in check if infected Flax plants were pulled up and burnt as soon as seen, but such a remedy is impracticable on a large scale. No really immune varieties of Flax are known, but fortunately the parasite seems not to occur in the Irish flax-fields. Distribution: Europe, North and South America, Australia. 15. Melampsora vernalis Niessl. Uredo Saxifragarum DC. Flor. fr. vi. 87. Cooke, Handb. p. 525 ; Micr. Fung. p. 215. Caeoma Saxifragae Wint. Pilze, p. 258. Plowr. Ured. p. 259. Melampsora vernalis Niessl, in Wint. Pilze, p. 237 (1881). Plowr. Gard. Chron. 1890, viii. 41 ; Jour. Roy. Hort. Soc. xii. p. cxi ; Trans. Brit. Myc. Soc. i. 59. Sacc. Syll. vii. 592. M. Saxifragarum Schrot. in Cohn's Krypt. Flor. Schles. p. 375 (1887). Fischer, Ured. Schweiz, p. 511. Spermogones. Scattered, yellow. yEcidiospures. Cseomata small, elliptic, flat, solitary, golden- yellow; spores in chains, roundish, finely verruculose, 17 — SO p. [ Uredospores. Sori epiphyllous, very small, round ; spores ellipsoid, echinulate, golden-yellow, 16 — 20 x 15 /a.] Teleutospores. Sori densely clustered, subepidermal, small, irregular, chestnut-brown ; spores oblong to clavate, yellowish- brown, 40 — 50 x 14 ix. On Saxifraga granidata. Rare. June — September. Plowright states that, on the specimens collected by Mr James Taylor at Clark Farquhar, N.B., in June, 1890, the teleutospores were found on the lower leaves and stems, and there were no uredospores. The descrip- tion of the latter is after Voglino and Fischer, and may not belong to the British species. The connection of the cseoina with the teleutospores has been proved by Plowright and Dietel. Distribution : Germany, Switzerland, Italy. 358 MKI.AMI'SOKIDII M MELAMPSORIDIUM Klebahn. I lett'i-ii'cious. Teleutospnres one-celled, with brownish membrane, united into Hat waxy crusts, but each little group starts almost always directly beneath a stoma. Uredo-sori surrounded by a hemi- spherical peridium which opens by an apical pore, often begin- ning beneath a stoma; uredospores abstricted singly, more or less smooth at one end, with indistinct germ-pores, not mixed with capitate paraphyses. yEcidia with a well-developed inflated peridium. Sperrnogones subcuticular, other sori subepidermal. This genus is, in some respects, as closely allied to Puccini- (i strum as to Melampsora, or rather more to the former than to the latter. Melampsoridium betulinum Kleb. rmh) populina var. betulina Pers. Syn. p. 219. Melampsora betulina Desru. ; Cooke, Handb. p. 522, f. 212 (misnamed winter spores) ; Micr. Fung. p. 219, pi. 9, f. 189, 190. Plowr. Ured. p. 243. Sacc. Syll. vii. 592. M'iampsoridium betulinum Kleb. Zeitschr. f. Pflanzenkr. 1899, p. 21 (see also 1891, p. 130) ; Wirtswechs. Eostp. p. 401. Fischer, Ured. Schweiz, p. 512, f. 320. ^Ecidiospores. iEcidia hypophyllous, single or in longitu- Fig. 267. .V. betulinum. a, peridium of uredo-sorus, viewed from above, showing the point at winch it will dehisce, x 300 ; b, uredospores, x 600. dinal rows parallel to the mid-rib, small, oval or oblong, 1 — 1% mm. long, clear reddish-orange, with irregularly torn margin ; MELAMPSORID1UM 359 spores roundish, 14 — 21x11 — 16/a, echinulate; epispore thinner and smoother above. Uredospores. Sori hypophyllous, with yellow spots showing on the upper side, collected in groups and mostly limited by the veins, each sorus scarcely j\ mm. wide, surrounded by a dome-shaped peridium which at length opens at the summit (where its cells are drawn out into long sharp points, Fischer); spores decidedly oblong or subclavate, orange, 22 — 40 x 8 — 12 /u,; epispore colourless, with distant spines, often smooth above. Fig. 268. M. betulinum. Sorus of teleutospores. The fusion-nucleus is seen in four of them, x 600. Teleutospores. Sori hypophyllous, always covered by the epidermis, scarcely \ mm. wide, in dense clusters limited by the veins, often spread over the whole leaf, reddish, then brown ; spores prismatic, rounded at both ends, somewhat oblique, 30 — 50x7 — 15 fx\ epispore thin, scarcely thickened above, nearly colourless, without perceptible germ-pore. iEcidia on leaves of Larix europaea, May ; uredo- and teleu- tospores on Betula alba (both verrucosa and pubescens), August — November, lasting through the winter on the decaying leaves. (Figs. 267, 268; see also Fig. 37, p. 78.) It was Plowright who first, in 1890 (after many unsuccessful trials) discovered that the Melampsora on Birch was connected with an secidium- form on Larch (see Zeitschr. f. Pflanzenkr. i. 130; Gard. Chron. 1890, viii. 41). He performed the experiment in both directions, and his conclusions were confirmed, eight years later, by Klebahn. The secidium in this case does not belong to the caeoma-type, but to that of Peridermium (P. Laricis Kleb.), having a peridium and resembling in its spores also Peridermium Strobi. The forms on B. verrucosa and B. pubescens are, to a small extent, 360 MELAMPSORELLA biologically distinct. The secidium stage of this parasite is very rare everywhere, and seems not to be recorded in Britain, except by Plowright at King's Lynn; the other are exceedingly common all over the country, but do little damage. The teleutospores germinate after a winter's rest, but since it is supposed they cannot infed the Birch and the s^cidium-stage appears to 1"' so rare, it is doubtful by what means the fungus propagates itself from yen- to year. Liro states that in Northern Europe the secidium <>n the Larch does not occur, though the fungus is very common in the uredo-stage, especially on Birch seedlings; but as he also says that the uredospores d<> not survive the winter there, the question of its perennation is left in a very unsatisfactory state. On the Hornbeam (Carpinw Betidus) there is in Europe an allied species Melampsoridium Carpini to which an secidium has not yet been discovered. Distribution : Europe, Asia, North America. MELAMPSORELLA Schroter. Heteroecious. Teleutospores in the epidermal cells, with a thin, colourless membrane, usually one -celled, germinating at once on maturity. Uredo-sori furnished with a peridium, and without any para- physes amidst the spores ; uredospores yellow, sessile, produced in short chains or singly, without evident germ-pores. /Ecidia with a peridium. Uredo-sori and gecidia subepidermal : sper- mogones subcuticular, without ostiolar filaments. This genus bears a certain resemblance to Milesian but is distinguished from it by the absence of vertical septa in the teleutospores, as well as by its habitat on Phanerogams. 1. Melampsorella Caryophyllacearum Schrot. Mcidiwm elaiinum A. et S. Consp. p. 121. Plowr. Ured. p. 270. Peridermium elatinum Link; Cooke, Handb. p. 535; Micr. Fungi p. 194! Credo Caryophyllacearum Johnst. ; Cooke, Handb. p. ">26 ; Micr. Fung. p. 216. Melampsorella Garyophyllacearwm Schrot. in Hedwig. 1874, xiii. 85. Klebahn, Wirtswechs. Rostp. p. 396. Fischer, Ured. Schweiz, p. 516, f. 522—6. MELA.MPXORELLA 361 Fig. 269. M. Caryo- phyllacearum. Peri- dermium elatinum, on A. pectinata (slightly reduced) ; «, a leaf, x 10. Melampsora Cerastii Wint. Pilze, p. 242 (1881). Plowr. Ured. p. 217. Melampsorella Cerastii Schrot. Flor. Schles. p. 366 (1887). Sacc. Syll vii. 596. M. elatina Arthur, N. Amer. Fl. vii. 111. Spenuogones. Epiphyllous, scattered, conical, honey-coloured. fficidiospores. iEcidia hypophyllous, arranged in an irregu- lar row on each side of the mid-rib, erumpent, shortly cylindrical, roundish or compressed, pale orange-red, with torn white margin ; spores ellipsoid or polygonal, orange, 16 — 30x14 — 17,0.; epispore thin, densely verru- cosa. Uredospores. Sori generally hypophyl- lous, usually arising beneath a stoma, sur- rounded by a peridium which slowly opens by an apical pore, small, crowded, pustular, yellow; spores sometimes in short chains, ovoid-oblong or ellipsoid, yellowish, 20 — 30 x 16 — 21 fi ; epispore thin, beset with a small number of pointed warts which are only visible when dry, with an occasional glabrous strip (?), without perceptible germ-pores. Teleutospores. Hypophyllous, often covering the whole leaf, developed within the epidermal cells, whitish-yellow or pinkish, in little groups in each cell, roundish or flattened, one-celled, 14 — 21 fi ; epispore smooth, thin ; basidiospores globose, nearly colourless, 7 — 9 fi. yEcidia on leaves of Abies pectinata, June — September; uredo- and teleutospores on Cerastium arvense, C. triviale and its var. alpestre, C. viscosum, Stellaria graminea, S. media (more rarely) ; uredospores from May onwards. Not very common. (Figs. 269, 270.) In North America it occurs on other species of Abies, and on Alsine and other species of Cerastium ; also in Europe on numerous allied species of the subfamily Alsinea). The teleutospores are developed on those leaves of the second bost which live through the winter ; they germinate about May ;u id can infect the Silver Fir. There are comparatively few records of the secidium-stage in this country; it causes small erect 362 MELA.MI'SoKKLLA witches:-bi ms on the Fir, the infected buds producing dense clusters of small leaves which are not pseudo-distichous as in the norma] shoots) but spirally arranged, moreover they are not evergreen, but fall off in August. <»n these leaves the spermogones and secidia are seen in June; and the secidiospores infect the second host and produce uredo-sori in 10 — 14 day-. It will he seen that the teleutosj .ores found in spring do not belong to the same generation as the uredospores found in the following summer. The mycelium of hot h stages is perennial, so long as the host survives, this being the only known instance of such a state of things: for this reason the parasite can maintain itself on either the Fir or the other host quite independently. On the Fir, it produces in time large cankerous growths. There is some evidence that a specialisation into biolog : ces on different species of Alsineae has begun. In the latter the Fig. 270. 31. Caryophyllaeearum. a, cells of the peridium of a uredo-sorus, on Stellaria graminea, showing the opening pore bein^ formed beneath a stoma, seen through the epidermis, x 600 ; b, an epidermal cell of S. nemorum, containing teleutospores (after Fischer) x about 400 ; c, attacked leaf of .S'. graminea, slightly enlarged; d, one young and two mature uredospores, x (500. mycelium grows up with the young shoots and gradually spreads upwards into every leaf, making the leaves in many cases smaller, but not much different in colour. I have seen a few uredo-sori on the upper side of the leaves, and even on the sepals of Cerastium. The peridium of the uredospores is formed of a definite hemispherical pseudo-parenchymatous layer, one cell thick, most of the cells being polygonal by mutual pressure, nearly smooth, and containing at first a little of the same yellow oil as the uredospores. It bears a very close resemblance in texture to that round the uredo-sori of Pucciniastrum Circaeac, Mi'/'*;,,,, Blerhai, Credinopsis JUieina, and their allies. The hetercecism of this parasite was established by Fischer in 1901 (Zeitschr. f. Pflanzenkr. xi. 321) ; the result was attained only after a long MELAMPSORELLA 363 and vain scanh extending over many years; it has been confirmed by Tubeuf, Klebahn, and Bubak. The secidia of three other species, Melam- psorella Symphyti, Pucciniastrumpustulatum and Calyptospora Goeppertiana, which also live on the same host, do not cause any deformation of the shoots, and can therefore be easily distinguished. DISTRIBUTION : Europe, North America. 2. Melampsorella Symphyti Bubak. Uredo Symphyti DC. Encycl. viii. 232. Plowr. Ured. p. 255. Sacc. Syll. vii. 861. Trichobasis Symphyti Lev. ; Cooke, Handb. p. 529. Coleosporium Symphyti Fckl. Symb. Myc. p. 43. Cooke, Micr. Fung. p. 218. M.'lampxorelhi Symphyti Bubak, Centralbl. f. Bakter. 2. xii. 423. Fischer, Ured. Schweiz, p. 523. [S 'per ■nwg ones. Chiefly hypophyllous, crowded, often spread over the whole leaf, orange-yellow. JZcidiospores. iEcidia hypophyllous, in two rows parallel to the mid-i'ib. not crowded, shortly cylindrical | — f mm. high, opening at the summit by a cleft, at length torn to the base into 3 — 5 segments ; spores orange, verrucose, 20 — 40 x 18 — 29 p.] Uredospores. Sori hypophyllous, small, rounded, closely crowded, rich golden-yellow, often spread over the Fig. 271. M. Symphyti. whole leaf, at first covered' by the Jj^g*68' °n *" epidermis and a thin peridium, then pulverulent ; spores ovate or ellipsoid, finely verruculose, yellow, 28 — 35 x 21 — 28 /*, without evident germ-pores. [Teleutospores. Hypophyllous, developed within the epi- dermal cells, forming large whitish or pinkish patches, many crowded in each cell, pale-yellowish, smooth, 11 — 18x9 — 15 p.] [.Kcidia on leaves of Abies pectinata, May, June;] uredo- [and teleutospores] on Symphytum officinale, rather uncommon, May to September. (Fig. 271.) Only the uredospores have so far been recorded for Britain, but the teleutospores would in all probability occur on the same plants at a later date, while the rccidia could doubtless be found, if carefully looked for, in 36 I PUCCINI \sti:i'M w Is where Abies and Symphytum both grew together. Tins was the experience of Bubak see Ber. Deutsch. Bot. Gesell. L903, xxi. 356). The itospores appear to arise or at least to mature about May on a perennial mycelium from the previous year, and are capable' of immediate germination: their basidiospores infect the young Fir-needles, but the mycelium thus produced is not perennial ami causes no witches'-brooms. The secidiospores are distinguished from those of the other species which grow upon the same host by their much larger size. The whole of the foregoing particulars (excepl as regards the uredospores are due to Bul>;ik ami Fischer; the former found his uredospores chiefly upon Symphytum tuberosum. It will lie useful to tabulate here by what characters the various secidia on Abies pectinata can lie discriminated : that belonging to .)/■ /"uipsorella Cajyimphi/ti has large spores 20 — 40x18 — 29 /z. Puceiniastrum pustulatum has a smooth line down the spore. Calyptospora Goeppertiana has spores 21 — 30 x 14 — 18 p. Distribution : Europe. PUCCINIASTRUM Otth. HeteiXHcinus or recidia unknown. Teleutospores extracellular, in a single layer, subepidermal, with a brownish membrane, divided by vertical septa into 2 — 4 cells. Uredo-sori surrounded by a delicate hemispherical peridium, opening at the summit with a pore; uredospores yellow in mass, with indistinct or no germ-pores. ^Ecidia with a thin cylindrical peridium (so far as known) ; aecidiospores verrucose except on one side which is thinner and smooth (? always), not provided with germ-pores. 1. Puceiniastrum Agrimoniae Tranzschel. Uredo Potentillarum var. Agrimoniae DC. Flor. fr. vi. 81. U. Agrimoniae Plowr. Ured. p. 255. Sacc. Syll. vii. 839. Coleosporium ochraceum Fckl. ; Cooke, Micr. Fung. p. 218. Puceiniastrum Agrimoniae Tranz. Script. Bot. Hort. Petrop. iv. 301. Fischer, Ured. Schweiz, p. 465. Arthur, N. Arner. Fl. vii. 106. Uredospores. Sori chiefly hypophyllous, pulvinate, small, confluent, sometimes spread over the whole leaf, covered by the epidermis and surrounded by a thin peridium which opens at the summit with a pore, orange-yellow, fading to ochraceous ; PUCCINIASTRl'M 365 Fig. 272. P. Agrimoniae. Uredospores. spores shortly ellipsoid or obovate, echinulate, orange, 18 — 21 x 14- /x ; epispore rather thick, with indistinct germ-pores. [Teleutospores. Sori similar, but indefinite, clear-brown : spores subepidermal, extracellular, emirate, smooth, each divided into four cells by two longitudinal walls at right angles to one .-mother, 30 x 21 — 30 //,.] On Agrimonia Eupatoria. Uredospores common, July — September; teleutospores, very rare everywhere, not yet found in Britain. (Fig. 272.) We owe our knowledge of the teleutospores to Tranzschel and Dietel ; see Engler u. Prantl, Nattirl. Pflanzenfam. vol. i. pt. 1**, p. 24. Until they were discovered, the position of the fungus was cpiite uncertain. Klebahn (see Zeitschr. f. Pflanzenkr. 1907, xvii. 149) proved that the parasite could maintain itself by over-wintered uredospores. Distribution : Europe, Asia, North and South America. 2. Pucciniastrum Circaeae Speg. Uredo Circaeae Schum. PI. Sail. ii. 228. Cooke, Micr. Fung. p. 217, pi. 7, f. 135, 136. /' '//<:<_■/ '// i a Circaeae Vers. ; Cooke, Handb. p. 507 p.p. M/lampsora Circaeae Winter ; Plowr. Ured. p. 245. Pucciniastrum Circaeae Speg. Dec. Myc. 65. Sacc. Syll. vii. 763. Fischer, Ured. Schweiz, p. 461, f. 302. Fig. 273. P. Circaeae. a, half of a leaf of C. lutetiana, showing uredo- sori (slightly enlarged); b, uredospore x600; c, part of peridiuin x 180 ; d, teleutospores, be- neath the epidermis, x 300. Uredospores. Sori hypophyllous, on paler patches bounded by veins, minute, yellowish, crowded, slightly confluent, surrounded by the epidermis and by a peridium which opens at the summit with a pore; spores ovate, 21 — 24x12 — 14^; epispore thin, covered with minute distant warts, without evident germ-pores ; paraphyses wanting; peridium usually opening beneath a stoma. 366 PUCCINIASTRUM Teleutospores, I [ypophyllous, in little subepidermal groups, roundish or Battened al the sides, divided longitudinally into aboul 2 4 cells, 17 — 24x21 — 28/u.; epispore of uniform thickness (about 2 /x), clear-yellowish, smooth. On Gircaea intermedia, G. lutetiand. Rather common in the uredo-stage. June — September. (Fig. 273.) It is probable that this, like Pucciniastrum pustulatum, has an aicidium- stage mi a i iter, lmt untiling has yet been discovered to confirm the suspicion. Puccinia Circaeae bas sometimes been confounded with this species, though they have nothing in common but the host-plant. They may both be found on the same leaf. Distribution : Central Europe. 3. Pucciniastrum pustulatum Dietel. Vredo pustulata Pers. Syn. p. 219. Cooke, Handb. p. 526; Micr. Fung. p. 21"). M 'ebtm psora pustulata Schrot. Pilz. Schles. p. .'504. Plowr. LTred. p. 244. Pucciniastrum Epilobii Otth in Mittheil. Nat. Gesell. Bern, 1861, p. 72. Sacc. Syll. vii. 762. Fischer, Ured. Schweiz, p. 4.")!). P. Abieti-Chamaenerii Kleb. Wirtswechs. Rostpilze, p. 393. P. pustulatum Dietel, in Engler u. Prantl, Nat. Pflanzen. i. 1**, 47. Arthur, X. Atner. Fl. vii. 107. [Sperm ogu nes. Hypophyllous, abundant, subcuticular. ^Ecidiospores. iEcidia hypophyllous, whitish, mostly in two rows corresponding to the two white lines, J mm. wide, 1 mm. high, cylindrical, with an evident peridium, opening by longitu- dinal slits ; spores mostly oval, 13 — 22 x 10 — 14 p, (15 — 17/u, Dietel); epispore finely verrucose, but with a thinner and smooth area which often forms a longitudinal strip.] Uredospores. Sori hypophyllous, caus- ing reddish or yellowish spots on the upper surface, roundish, \ mm. wide, scattered or in little groups, often confluent, orange, surrounded by the peridium and the remains of the epidermis: spores mostly oval, orange- yellow, distantly echinulate, 15 — 22 x 11 — 14 /a: paraphyses wanting. Fig. 274 latum. V. pvstu- .lEcidia on leaf of A. pectinata, slightly enlarged; a, a peridium, x 10. From a foreign specimen. PUCCINIASTRUM 367 Teleutospores. Sori hypophyllous, clustered or widely con- fluent, {rnni. wide, covered by the epidermis, reddish, then blackish-brown; spores shortly cylindrical or prismatic, 17 — 35 x 7 — 14 /x, those in the middle of the sorus Melampsora-like, but at the periphery roundish or oval, and composed of 1 — 3 cells, i.e. divided by longitudinal walls; epispore clear-brown, thickened (up to 3/x.) above, but with a thinner spot (? germ- pore). [ . Ecidia on Abies pectinata, June, July ;] uredo- and teleuto- spores on Epilobium {Chamaenerion) angustifolium, E. palustre. August — October. Shere, Abinger, Surrey. (Fig. 274.) The recidiospores are distinguished from those of Calyptospora Goep- pertiana by their smooth area and generally shorter length. The ascidium of Mdampsorella Caryophyllacearum lives also on Abies pectinata, but produces thereon witches'-brooms, and finally swellings and canker of the branches, while P. pustulatum attacks the leaves only. Klebahn, Fischer, and Tubeuf have all demonstrated experimentally the genetic connection of the stages of this parasite, so far as concerns E. angustifolium ; but the form on E. palustre and its allies is probably, at least biologically, distinct. Klebahn's name, quoted above, includes only the form on E. angustifolium and the allied E. Dodonaei. He restricts the name P. Abieti-Chamaenerii to this, calling the form on E. palustre, E. montanum, etc. by the name P. Epilobii, and assigns to them also certain small morphological differ- ences. The collective species is common in North America, but the secidium has not yet been recognised there. Distribution: Europe, North America. 4. Pucciniastrum Pyrolae Dietel. JEcidium Pyrolae Gmel. in Linn. Syst. Nat. ii. 1473. Uredn Pyrolae Grew Flor. Edin. p. 440 p.p. Fischer, Ured. Schweiz, p. 539, f. 337. Trichobasis Pyrolae Berk. ; Cooke, Handb. p. 529 ; Micr. Fung. p. 223 p.p. Melampsora Pyrolae Schrot. ; Plowr. Ured. p. 247. Pucciniastrum Pyrolae Dietel, in Engler u. Prantl, Nat. Pflanzen. i. 1**, 47. Arthur, N. Amer. Fl. vii. 108. Thecopsora (?) Pyrolae Karst. ; Sacc. Syll. vii. 766. Uredospores. Sori hypophyllous, minute, hemispherical, orange, on yellowish or purplish discoloured spots, surrounded by the epidermis and by a peridium which both open at the 368 THECO PSORA [Teleutospores. suiniiiit with a pore; spores elongated-ellipsoid or clavate, provided with distant and pointed warts, yellowish, 26 — 35 x 14— hi/u,; epispore rather thick, without perceptible germ- pores. Sori hypophyllous, adjoining the uredo-sori, inconspicuous, flat, subepidermal, forming an even layer of laterally united cells: spores columnar or oblong, 24 — 28x10 — 12 /a : epispore colourless, uniformly thin (1//,) — (description after Dietel).] On Pyrola minor, P. rotundifolia. May — October, uredospores only; Scotland, Ludlow, etc. (Fig. 275.) Teleutospores have been met with by few observers; previous to their discovery, it was uncertain in what genus the fungus should be placed. It may possibly be hetercecious. The uredospores are often more coarsely warted at one end, though this is not invariably the case. Fischer figures the cells of the peridium round the pore as furnished above with pointed warts, of which one is distinctly longer than the others: those cells are enormously thickened on the lower wall. Distribution : Europe, North America. Pig. 275. /'. Pyrolae. Uredospores, on P. rotundifolia. THECOPSORA Magnus. Hetercecious or secidia unknown. Teleutospores intracellular, occupying and filling the epi- dermal cells of the leaves, united into a brown crust, other characters as in Pucciniastrum. Uredo as in Pucciniastrum. .r>idia hemispherical, with a thick brown peridium (so far as known); a>cidiospores verrucose, with a narrow, thin, smooth strip down one side (? always).* 1. Thecopsora Padi nov. comb. Licea strobilina A. et S. Cons}), p. 109, pi. 6, f. 5. Phdonitis strobilina Pers. ; Cooke, Handb. p. 409, f. 141. Perickaena strobilina Fr. ; Greville, Scot. Cr. Fl. pi. -1~,'>. J&cidium strobilinum Wint. Pilze, p. 260; Plowr. Ured. p. 266. THECOPSORA 369 Credo Padi K. et S. exsicc. 187. Cooke, Handb. p. 527. U. porphyrogenit.i Kze. ; Cooke, Micr. Fung. p. 21(5. Melampsora Padi Cooke, Handb. p. 523 (1871). Plowr. Ured. p. 246. Fung. Fl. Yorkshire, p. 184. Pucciniastrum Padi Dietel in Eng. u. Prantl, NatUrl. Pflanz. i. 1**, p. 47. Fischer, Ured. Schwciz, p. 463, f. 303 ; Centralbl. f. Bakter. 2. xv. 227. Thecopsora areolata Magn. in Hedwigia, 1875, p. 123. Sacc. Syll. vii. 764. Spermogones. Whitish, pustular, flat, open, exhaling a strong smell. sEcidiospores. yEcidia crowded, covering on the upper (sometimes the under) side the lower part of the scales of the fallen cones, hemispherical or polygonal; peridium thick, brown, woody, opening by a slit; spores oval, inequilateral, yellow, 21 — 28x17 — 20 /x ; epispore very thick (up to 6 /j.), echinulate, with a narrow, thinner, smooth stripe. Uredospures. Sori hypophyllous, clustered on spots 1 — 5 mm. wide which are brownish above, reddish or purplish below, and more or less bordered by the veins, covered by the epidermis and by a peridium which opens at the summit by a pore ; spores oblong-oval or irregular, echinulate, pale-yellowish, 15 — 21x10 — 14 //,; epispore about H/a thick. Teleutospores. Developed in the epidermal cells, several in each, epiphyllous or occasionally hypophyllous, forming dark- brown shining crusts which are bounded by the veins ; spores oval-cylindrical or prismatic, 22 — 30 x 8 — 14 /*, divided by thin longitudinal walls into 2 — 4 cells ; epispore thin, slightly thickened above, clear-brown, smooth, with a germ-pore in the upper and inner corner of each cell.. iEcidia on cones of Picea excelsa, Scotland, Yorkshire, August, November; uredo- and teleutospores on Prunus Padus,, August, September. Very rare. (Fig. 276.) g. u. 24 Fig. 276. Th. Padi. a, leaf of P. Padus, showing uredo- sori ; b, scale of cone of Picea excelsa, showing aecidia (both reduced). Some of thesecidia are broken and empty. 370 TllKcol'SOKA The connection of the spore-forms on these two hosts lias heen experi- mentally demonstrated by Klebahn, Tubeuf, and Fischer. The basidio- spores in spring (about the time of pollination of the Fir) infect the female tlowci's of the Spruce Fir, which are usually at the top of the high trees; occasionally also the young shoots are affected, hut they do not produce secidia. The secidia are developed in late summer, and mature on the fallen cones: their spores germinate in the following -May, and then infect the leaves of the Bird Cherry, on which they produce uredospores in the summer and teleutospores in the autumn (Fischer, I.e. and Centralbl. f. Bakter. 2, xv. 1906, p. 227). The description of the teleutospores is taken from Klebahn and Fischer. As will he seen from the synonymy, the secidium was originally classed among the Myxomycetes. The three stages appear in Cooke's Handbook (according to the knowledge then prevailing) on three different pages, the secidium from Appin, the uredo from some place in Scotland, and the "Melampsora" from Swanscombe, Kent (Cooke, 1865). It is also recorded on l\ Padus in Yorkshire Fung. Fl. p. 184, while the secidium is recorded on "pine-cone scales" on p. 369. The uredo has also been found at Braemar, Ahoyne, Perth, etc.; and the secidium in Dumfriesshire. Distribution : Europe. 2. Thecopsora Galii De Toni. Caeoma Galii Link, Sp. PL ii. 21. Melampsora Galii Wint. Pilze, p. 244. Plowr. Trans. Brit. Myc. Soc. i. 59. Pucciniastrum Galii Fischer, Ured. Schweiz, p. 471, f. 307. Thecopsora Galii De Toni in Sacc. Syll. vii. 765. Uredospores. Sori scattered or gregarious, small, round, pulvinate, reddish, covered by the epidermis and by a peridium which opens at the summit with a pore; spores shortly ellipsoid or ovate, sparsely echinulate, orange-yellow, 17 — 20x14 — 16/x; epispore colourless, without evident germ-pores. Teleutospores. Developed in the epidermal cells, forming little dark-brown crusts, crowded, roundish, longitudinally septate into 2 — 4 cells, 21 — 24x21 — 32 /a ; epispore rather thick, yellowish-brown, smooth, with an evident germ-pore at the upper and inner corner. On Galium verum (H. T. Soppitt), June — September, 1889. Very rare. It is reported, in continental Europe, as occurring on other species of Galium, also on Sherardia and Asperula. thecopsora 371 3. Thecopsora Vacciniorum Karst. Uredo Vacciniorum Link, Sp. PI. ii. 15. Cooke, Handb. p. 527; .Micr. Fung. p. 21. Uredospores. Sori hypophyllous, rather small, scattered or in groups on (sometimes orange) dis- coloured spots which are visible for some time before the sori appear, long covered by the epidermis, brownish, surrounded by a peridium which opens Fiur. '277. Th. Vaccini- ,i -,i arum. Uredosi)ores on at the summit with a pore; spores v Myrtillm (Scotland). ovate to ellipsoid, sparsely echinulate- vcrrucose,. orange-yellow, 15 — 16x12 — 13 fi, (21 — 28x14 — 18 fi, Fischer). [Teleutospores. Hypophyllous, developed in the epidermal cells which are filled by them and form little brown inclehiscent crusts, pale-brown, divided by longitudinal walls, 14 — 17 x 7 — 10 /x; epispore uniformly thin.] On leaves of Vaccinium Myrtillus, V. Vitis-idaea. Uredo- spores only, May — October, Shropshire, Yorkshire, North Wales, Scotland. * (Fig. 277.) The teleutospores are very rare, and can be found only on the dead or fallen leaves. Fischer suggests that this may be a hetenucious species as indeed is most likely the case, though no probable suggestion has yet been mi de as to the nature of the alternate host. He records it on V. uligino- xti tH also ; and Arthur on several other species of Vacciniaceae, in the United States. Distribution : Europe, North America. 24- 372 CALYPTOSPOB \ CALYPTOSPORA J. Kiilm. Eetercecious. Teloutospoivs intracellular, occupying the swollen epidermal cells all round the stem of the host for a considerable distance, otherwise as in Pucciniastrum. No uredo. . lv-idi.-i cylindrical, with a thin peridium; aecidiospores verrucose, without germ-pores, and with no smooth spot. Calyptospora Goeppertiana Kiihn. jEcidium columnare A. et S. Consp. p. 121, pi. 5, f. 1. Peridemiium columnareS. et K.; Cooke, Handb. p. 535 ; Micr. Fung. p. 194, pi. 2, f 27, 28. ? JEcidiurn pseudo-columnare Plowr. Ured. p. 271 (not Kiilm ?). Calyptospora Goeppertiana Kiihn in Hedwigia, 1869, viii. 81. Saec. Syll. vii. 766. Kew Bulletin, 1907, p. 1, with plate. C. columnaris Kiihn (1886); Arthur, X. Amer. Flor. vii. 114. Pticcintastrum Go&ppertianvm Kleb. Wirtswechs. Rostp. j>. 391. Fischer, Ured. Schweiz, p. 466. f. 304. JEcidiospores. iEcidia hypophyl- lous, arranged in two long rows parallel to the mid-rib, cylindrical, white, with torn or slit margin ; spores ellipsoid, uniformly verrucose, orange-red, 21 — 30 x 14 — 18 /n. Teleutospores. Caulicolous (the fusiform attacked part of the stem being swollen and at first clear- pink, passing into brown), developed in the epidermal cells, densely crowded, prismatic, mostly divided by crossed longitudinal walls into four cells, up to 42 /u. high ; epispore yellowish-brown, smooth, thickened (up to 3 /a) at the summit, with a germ-pore at the upper and inner corner of each cell. Fig. 278. C. Goeppertiana. Affected branch of T". Vitis- idaea, Scotland (slightly re- duced); a, leaf of A. pecti- nata, b, leaf of A. Nordman- niana, with secidia, Torquay (slightly enlarged). CALYPTOSPORA 373 ^Ecidia on leaves of Abies pectinata, A. Nonlnxmniana, June — September ; teleutospores on stems of Vaccinium Vitis- idaea, July — September. England, Wales, Scotland. Very rare. (Fig. 278.) The life-history has been experimentally demonstrated by Hartig, Kulm, and Bubak. It has been shown that the iccidia can be developed in artificial cultures on other species of A hies (but not on Tsuga canadensis or Pseudotsuga l><>uglasii), though it is not recorded on any of these in natural conditions. Saccardo's citation of "Abies canadensis" in the Sylloge is probably an error. The teleutospores are recorded also on several other species of Vaccinium, including V. Myrtillus A. Gray, in the United States, but the secidia have not been observed there. The infested branches of the Cowberry stand perfectly erect ; the plant becomes taller and the leaves stunted. No uredo-stage occurs. For the life-history see p. o9. The mycelium is perennial in the Cow- berry, producing fresh crops of teleutospores year after year ; this may be the origin of the (presumably) erroneous statement in the Kew Bulletin (I.e.) that the basidiospores are able to infect the Vaccinium again, as well as the Silver Fir. This statement was originally made by Hartig, but is unsupported by any experimental evidence. When planting any of the species of Abies liable to attack, it would be w«U to look for and burn all infested bushes of Cowberry in the neighbour- hood ; they are easily recognisable by their peculiar habit. The fungus cannot attack the Firs unless the infested Cowberry is near enough to convey the infection. Distribution : Europe, North America. HYALOPSORA Mac mis. Teleutospores in one or two layers, produced in the epidermal cells which are united into crusts; spores with colourless membrane, each divided by vertical septa into 2 — 4 (or more) cells. Uredo-sori subepidermal, without a peridium or with a very rudimentary one, but surrounded by paraphyses; uredospores of two kinds, yellow, sessile, furnished with evident germ-pores. On Ferns. It has been suggested that the species of Hyalopsora are hetercecious, Abies and Pinus being named by Bubak (1906) as 374 IIYAI.ol'SoRA possible alternate hosts. This is in itself improbable and is inconsistent with Arthur's suggestion that of the two kinds of nrcdospoivs the first kind represents aecidiospores. 1. Hyalopsora Aspidiotus Magn. Uredo Polypodii Schrot. Krypt. Flor. Schles. iii. 374. Plowr. Ured. p. 256 p.p. Sacc. Syll. vii. 857 p.p. U. Asj,iili',/n.< Peck in Ann. Rep. X. Y. State Mus. xxiv. 88. Melampsorella Aspidiotus Magn. Ber. Deutsch. Bot. Gesell. xiii. 288. Pucciniastrum Aspidiotus Karsten, Bidr. Finl. Nat. Folk, 1879. xxxi. 1 13. Hyalopsora Aspidiotus Magn. Ber. Deutsch. Bot. Gesell. 1901, xix. 582. Arthur, N. Amer. Flor. vii. 112. H. Polypodii-Dryopteridis Magn. in Hedwig. 1902, p. 224. Fischer, Ured. Schweiz, p. 472, f. 308. Uredospores. Sori amphigenous, scattered, small, round. up to | mm., golden-yellow, without a peridium, dehiscing irregularly, often seated on yellowish spots; spores oblong or ellipsoid, of two kinds, (1) thick-walled (2| — 3^/x), with very Fig. 279. II. Aspidiotus. Part of frond of P. Dryopteris, showing uredo-sori, nat. size ; three spores, all from the same sorus (Scotland). faint, hardly perceptible warts, 36 — 72x30 — 40 p. with 6 — 8 scattered germ-pores, (2) thin-walled (about li f.L), covered uniformly with very faint scattered warts, 28 — 40 x 16 — 26 p, with four indistinct equatorial germ-pores; paraphyses few. Teleutospores. In the epidermal cells, often filling them completely, roundish or irregular, flattened where they are in contact, sometimes arranged in two layers, about 25 p. high, 21 — 35 p or more wide, divided by vertical septa into 3 — 5 EYALOPSOEA 375 (mostly four) cells; epispore thin, smooth, colourless; germ- pore not prnvptihlo. On Polypodium Dryopteris. Uredospores, June — August; teleutospores, May and June, on young leaves (Magnus). Rare. (Fig. 27!).) Fischer, who records it also on P. Robertianum, states that the teleuto- spores germinate in June. Arthur prefers to call the first kind of uredospore the secidiospore ; this is possibly correct but, until something is known about their development, it is premature to decide. According to him, the secidia have no peridium, and the uredo-sori a very rudimentary one, but I have repeatedly found both kinds of spores in the same sorus. They appear perfectly smooth, when seen in water. Distribution : Europe, North America. 2. Hyalopsora Polypodii Magn. Uredo linearis var. Polypodii Pers. Syn. p. 217. U. Filicum Desm. ; Cooke, Handb. p. 526; Micr. Fung. p. 215 p.p. White, Scot. Nat. 1877, iv. 27, pi. 2, f. 7. U. Poly} "" Hi DC. Flor. fr. vi. 81. Plovvr. Ured. p. 256 p.p. Fungus Flor. Yorkshire, p. 204. Sacc. Syll. vii. 857 p.p. Pucdniastrum Polypodii Dietel, in Hedwig. 1899, xxxviii. 260. Hyalopsora Polypodii Magn. Ber. Deutsch. Bot. Gesell. 1901, xix. 582. Fischer, Ured. Schweiz, p. 474, f. 309. Arthur, N. Amer. Flor. vii. 112. Dietel, Annal. Mycol. 1911, ix. 530. Uredinopsis Polypodii Liro, Uredin. Fennic. 1908, p. 496. Uredospores. Sori hypophyllous, minute, scattered, bullate, golden-yellow, without a peridium, rup- turing irregularly; spores more or less globose or ellipsoid, of two kinds, (1) thick-walled (2 — 3 //,), with very faint warts, 26—38x18—29//, with 6—8 scattered germ-pores, (2) thin-walled (1-- l\p), covered uniformly with faint 'u'reS'porS", on T^i!,- distant warts, 22 — 35x13 — 20 a, with pteris fragilis (Shrews- „ • i bury). lour equatorial germ-pores. Teleutospores. In the epidermal cells, often filling them completely, showing as yellowish-brown spots on the under side of the leaf, densely crowded, divided into 2 — 4 cells, each about 14 — 18 /z in diam., single cells subglobose ; epispore thin, colourless; germ-pore perceptible at the upper end. ::ti; MILKS |\ A ( )n Cystopteri* ffits]>ores = ?R>cidiospores ; b, uredospores (no. 2) and parajmysis ; c, teleutospores. From a German specimen, issued by Sydow. £ mm. diam., fiattish, semi-transparent, always situated beneath a stoma, surrounded by a rather tough peridium, rupturing above ; spores more or less obovate-polygonal, on slender 380 CJREDINOPSIS pedicels as long as <.r longer than the spore, colourless, densely verruculose, averaging 20 — 2(j x 12— 14/z: epispore about 1 fi thick: (2) larger, aboul j mm. diam., more elevated, pustular, withoul (?)peridium, but with what looked like hyaline clavate i hiii-wallcd paraphyses; spores ovate-fusiform, colourless, smooth, thickened at the apex with a conical, acute, usually oblique process, 35 — 45 x 10 — 12 /u, : pedicels short; epispore thin. Teleutospores. Scattered singly throughout the mesophyll. Le. >i«>t in definite clusters, roundish-oval or oblong, 1 — 3 (mostly two) celled, 18 — 24x15 — 16//; epispore very thin, almost perfectly hyaline and smooth. On Polypodia in Phegopteris. Europe. Not uncommon; appearing about the time when the fern-sori are being formed. The uredo-sori are of about the same size as a fern-sporangium, and are scattered among the fern-sori. (Fig. 284. ) The description and figures are taken from a German specimen. The fungus will no doubt be found in this country, if looked for ; it is rather conspicuous on account of the uredo-sori. The warts on the uredospores (no. 1) are clearly perceptible even when wet : the peridium is pseudoparen- chymatous, with cells isodiametric near the apex, becoming longer and prosenchymatous towards the periphery; texture rather tough and con- sistent, not friable. Fischer gives a beautiful and accurate drawing of the nature of the peridium and its relation to the epidermis (I.e. fig. 310). The uredospores (no. 2) are described by Arthur in his generic character (N. Amer. Flora, vii. 115) as smooth, except for two longitudinal, thickened, more or less warted ridges, which I could not discern. As in Hyalopsora, Arthur prefers to call the two kinds of uredospores respectively secidio- and uredo-spores : the same remark may be made as under that genus, for the supposed secidiospores occur in company with the teleutospores on the dying parts of the frond, and are said to germinate only after having passed the winter. Uredospores (no. 2) can germinate at once. The teleutospores are to be found in large numbers in the tissues immediately round both kinds of sori, and simultaneously with them. They are borne on short lateral branches of the mycelium. Dietel, who o-ave a long account of this fungus (Ber. Deutsch. Bot. Gesell. 1895, xiii. 326 . showed that in the allied European species, U. Struthiopteridis, the teleutospores germinated easily with a typical basidium (occasionally branched) which bore round basidiospores. Distribution: Germany, Switzerland, Italy. APPENDIX HEMILEIA Berk, et Broome. vEcidia, if any, unknown. Uredospores borne singly on hyphae which protrude in fascicles through a stoma. Teleuto- spores formed later on pedicels in the centre of the same fascicles, one-celled, with apical germ-pore, germinating as in Uromyces. The character given for Hemileia, when only H. vastatrix was known, of having one side of the uredospore smooth, is now known to be not of universal application. Hemileia americana Mass. Hemileia amerieuna Mass. in Gard. Chron. 1905, xxxviii. 153, f. 53; Kew Bulletin, 1906, p. 40, with plate. Uredospores. Sori hypophyllous, forming broadly effused pulverulent deep-orange patches, often several centimetres in ex- tent ; spores perfectly spherical, shortly stipitate, 24 — 32/x diam.; epispore bearing small, rather sparsely scattered, round warts, and with two germ-pores. Teleutospores. Occupying the centres of the heads of uredo- spores, shortly stipitate, colour- less, broadly obovate or turbinate, often with a small obtuse api- Fig 285- /; americana. Tip of cuius, densely verruculose, 30 x CaMZeyaleaf, showing sori (slightly J reduced); a, uredospore, and b, 22 — 25 \x. teleutospore, x (iOO. :*N2 1 1 KM 1 1. HI A ( >n leaves of Cattlei/a Duiciunn I Ijit < -in., imported from Costa Rica, L899. (Fig. 285.) Only a small patch of IJust was present mi the leaf when the plant was received from Costa Rica, but this continued to increase in size and the tailing spores infected other leaves. The uredospores germinated readily, and young Cattleya leaves, inoculated on the under .surface, produced mature uredospores in thirteen days. No success attended the efforts to infect oilier orchids, not belonging to the genus Cattleya. This description is founded on that given in the Kew Bulletin; on referring to the Gardener's ( Ihronicle (I.e.) it will be seen that the particulars there given differ in several respects. The specimen is in the Kew Herbarium. This fungus and the others of the same genus might easily become dangerous parasites in orchid houses, if allowed to spread; though it seems probable, on the slender evidence at present known, that each is confined, like most other Rusts, to a single genus, if not species. Hemileia Phaji Sydow. Credo I'lit'ji Haciborski, Parasit. Alg. und Pilz. 1900, ii. 32. U. Lynchii Adams, Irish Naturalist, 1911, xx. 68. Hemileia Phaji Sydow, Monogr. iii. (ined.). Grove, Journ. Bot. 1913. p. 44. Y,j. 286. H. Phaji. a, epidermis, showing fascicle of uredospores emerging from a stoma, x 180 ; b, the same in section ; c, uredospores, x 600 ; d, a portion of a leaf, with uredo-sori, in Herb. Kew, x ^. Uredospores. Sori hypophyllous, densely scattered, round, distinct, formed of little dense fascicles of hyphae (20—25) which issue through a stoma; pedicels divergent, clavate up- wards, each surmounted by a spore ; spores subglobose or rarely II EMI LEI A . 383 obovate, strongly but sparsely echinulate-verrucose all over, 13— 15 /a diam. or 18—25x12—16/*. On Phajus Walliclrii. Botanic Gardens, Glasnevin, Dublin (Sir Frederick Moore). April. (Fig. 286.) The description and figures a, b, and c are derived from some slides mounted by Sir Frederick; 1 have not seen the leaves, which were not preserved, but there is an exactly similar form on Phajus sp. in Herb. Kew, tinlocalised but apparently sent up by some gardener for identifica- tion. This also was named Uredo Li/iidiii at first. Hemileia Oncidii Griff, et Maubl. Hemileia Oncidii Griff, et Maubl., Bull. Soc. Myc. Fr. 1909, p. 138, pi. vi. Uredospores. Sori hypophyllous, numerous, pulverulent, orange, minute, 50 — 100 \x diam.; fertile hyphse issuing from stomata, fasciculate, branched, 25 — 30 /u. long, clavate above; Fig. 287. H. Oncidii. Uredospores, a, wet, b, dry : c, part of leaf, showing uredo-sori, x i. From the specimen in Herb. Kew. spores globose or rarely obovate, echinulate-verrucose, filled with orange drops, 16 — 18 /x diam., occasionally 20 /x long; epispore hyaline. [Teleutospores. Sori growing in the centre of the spots, pallid-brownish ; spores subglobose or pyriform, at first hyaline, aculeate, then pallid-brown and somewhat smooth, 20 — 23 x 15—20/*.] On Oncidium varicosum, imported by Messrs Stuart Low & Co. from San Paulo, Brazil, August, 1909. Specimen in Herb. Kew. (Fig. 287.) In the Kew specimen, the spots occupied by the crowded groups of uredo-sori are more or less oval, ^ — 1 cm. across, and covered with orange-yellow dust, but as the sori are limited, each by the stoma, through which it issues, they never become confluent. The description is founded upon that of Griffon and Maublanc; only uredospores were seen. What is doubtless the same species has been found at the Botanic 384 (HUVSOMYXA Gardens, Glasuevin, Dublin, on Onddium varicomm and 0. Forbesii. Others of the same character were also seen there by Sir Frederick Moore ..11 Epidendrum viteUinum and Lycaste Skinneri. These tropical parasites are imported with the plants, and occasionally spread to a small extent, under glass, in this country. Persistent examination of such imports would no doubt discover still others of the same kind. Chrysomyxa Rhododendri De Ban. JEcidium abietinum A. et S. Consp. p. L20 p.p. Uredo Rhododendri DC. Flor. fr. vi. 86. Chrysomyxa Rhododendri De Bary, in Bot. Zeit. 1879, p. 809, pi. 10, f. i_6. Sacc. Syll. vii. TOO. Fischer, Fred. Schweiz, p. 126. [.Er'nIinxjHii-cs. .Ecidia irregular, membranaceous, com- pressed and elongated parallel to the mid-rib of the leaf (up to 3 mm. long), erumpent on transverse yellowed zones, at length irregularly torn, whitish; spores more or less ellipsoid, yellowish, 17 — 45x12 — 22 ^ : epispore thin, thickly verruculose except for a smooth longitudinal stripe.] Uredospores. Sori almost always hypophyllous, on yellowish or brownish spots, minute, roundish or oblong, scattered or in small groups, at length pulverulent, orange ; spores in chains with intercalary cells, irregular or oval, verruculose, orange- yellow, 17 — 28 x 15 — 22 ijl, without perceptible germ-pores. [Teleutospores. Sori similar, in densely clustered groups, brownish-red . spores single or in the centre of the sorus several (4 — 6) superposed in a row, prismatic, 20 — 30x10 — 14 fj.: epispore colourless, thin, but with an annular thickening at the summit of the uppermost cell.] [iEcidia on leaves of Picea excelsa, August — October ;] uredospores on Rhododendron hirsutum, Douglas Castle, Lanark- shire, June, 1913 (D. A. Boyd). While this book was passing through the press, Mr D. A. Boyd kindly communicated specimens of this most interesting find. It occurred in small quantity, and only uredospores were observed. Description of the other forms after De Bary. This parasite is very common in Switzerland, in all parts where the Fir and the Alpine Boses occur together. The teleutospores germinate in June or July and the spermogones and secidia appear on the Fir leaves from that time onwards. The secidiospores can at once infect the Bhododendron leaves, where the mycelium winters, UREDO 385 producing its spores in the following spring. The fungus is recorded also on 11. ferrugineum and R. dahuricum, but does not attack the American or Himalayan species. DISTRIBUTION : Central Europe, Siberia. Uredo Lynchii Plowr. Trichobasis Lynchii Berk, in Gard. Chron. 1877, viii. 242. Berk, et Br. Ann. Nat. Hist. 1878, ser. 5, i. 26. W. G. Smith in Gard. Chron. 1885, xxiii. 603, f. 154. Uredo Lynchii Plowr. Ured. p. 259. Sacc. Syll. vii. 852. Uredospores. Sori subepidermal, erurapent, on small pallid spots, scattered, rarely confluent ; spores yellow, obovate, beau- tifully echinulate, with short pedicels, 28 — 35 x 20 — 30 /x. Fig. 288. U. Lynchii. Sori on leaf of Spiranthes, surrounded by the ruptured epidermis, x 40 ; two uredospores. From the original specimens. On a Spiranthes from Trinidad, Kew Gardens, August, September; present on the plant when imported (R. Irwin Lynch). On living leaves of Spiranthes in hothouse, Kelvinside, Glasgow, September, 1890 (D. A. Boyd). Specimens in Herb. Kew. (Fig. 288.) Uredo Plantaginis B. et Br. Uredo Plantaginis B. et Br. Ann. Nat. Hist. ser. 5, vii. 130 (1881). Plowr. Trod. p. 259. Sacc. Syll. vii. 857. Uredospores. Sori epiphyllous, on round yellowish spots 3 — 6 mm. wide, erumpent, surrounded by the epidermis; spores roundish to obovate, faintly echinulate with sharp but distant opines, yellowish, varying from 19 x 20 yu. up to 27 x 16 /m, with (apparently) two or three equatorial germ-pores. On Plantago lanceolata(i), P. major. Very rare. (Fig. 289.) g. u. 25 386 UREDO Plowrighi mentions this with doubt ; Berkeley and Broome (I.e.) record it as found on Plantago al Wood Newton, and mi /'. lanceolata at Dolgelly Ralfs). In the Trans. Worcester. Nat. Club, L910, p. 291, it is recorded at Pirton Pool. These I have no1 seen. Plowrighl suggests that the Dolgelly specimen in.i\ be a Synchytriu'm, and some of those I have Fig. 289. U. Plantaginis. seen in herbaria under the name U. Plantaginis Uredosporea, on P. are n,,t (Jredines. But in the British Museum major, from the Isle of ,, „ . ,, , . ,. . Wight specimi n there is one on /'. major, collected in the Isle of Wighi by J. F. Rayner, October, 1907, which appears to belong to this class; the foregoing description is taken from this specimen, the son lie in the centre of thickened yellow spots which are sometimes confluent. Uredo Tropaeoli Desm. Uredo Tropaeoli Desm. Ann. Sci. Nat. l^."ii;. vi. 243. Cooke, Journ. Bot. iv. 97; Handb. p. 528; Micr. Fung. p. 216. Plowr. Ured. p. 258. Sacc. Syll. vii. 862. Uredospores. Sori hypophyllous, on pale-yellow spots. minute, roundish, scattered or confluent : spores subglobose to ovoid, orange-yellow, 20 fx diam. On leaves of Tropaeolum aduncum (= T. canariense). Very rare; Shere, near Guildford, October, 1865 (Dr Capron). This species has been found only once in England. Desruazieres who records it on Tropaeolum minus, in August, says that it causes, on the upper face of the leaf, numerous irregular spots, barely 1 mm. across. It has been recorded from France and Belgium. o iEcidium Hellebori Fischer. jE. Hellebori Fischer, Ured. Schweiz, p. .">26. Spermogones. Epiphyllous, or hypophyllous amongst the secidia, immersed, with projecting paraphyses, about 135 — 150/* diam. jEcidiospores. zEcidia hypophyllous, crowded, in roundish groups, cup-shaped, with a torn spreading margin ; spores densely verruculose, 18 — 24 x 18 — 21 /j,. On Helleborus vi?-idis. May. Included with all reserve. Winter mentions an secidium on Helleborus foetidus (Pilze, p. 269), which he places under JEc. Ranunculaeearum. .ecidium 387 Rev. C. Wolley-Dod records (Journ. Roy. Hort. Sue. xii. p. liii.) an aecidiurn allied to JEc. Ficariae on his Hellebores at Malpas. I have not met with any other reference to an secidium on that genus. The whole of the description is taken from Fischer, who adds that it belongs presumably to a heteroecious species. JEcidium Ranunculacearum I >C. var. Linguae. JE. Ranunculacearum DC. Blor. fr. vi. 97 p.p. Grev. Flor. Edin. p. 446. Plowr. Tied. p. 266. JEcidiospores. ^Ecidia hypophyllous, in roundish or elongated clusters of various sizes, cup-shaped, whitish, with a brittle margin; spores polygonal, orange-yellow, 17—28x14 — 20 fi. On Ranunculus Lingua. Very rare: Duddingston Loch (Dr Greville). It was at one time suggested by the brothers Sydow that this secidium belonged to their P("»v////" Galamagrostidis on Calamagrostis neglecta, hut they have since seen reason to doubt the truth of this idea. iEcidium Poterii Cooke. JEcidiwm Poterii Cooke, Journ. Bot. ii. 39, pi. 14, f. 3 ; Handb. p. 540; Micr. Fung. p. 198. Plowr. Ured. p. 268. See also Berkeley, Eng. Fl. v. 373. JEcidiosjJores. zEcidia hypophyllous, in subrotund or elon- gated clusters on the nerves of the leaves, also upon the petioles, scattered or circinate, immersed, edges torn into minute fugacious teeth ; spores oval, yellowish. On Poterium Sanguisorba. May and June. A species of which nothing else seems to be known. & iEcidium Dracontii Schweinitz. JEcidium Dracontii Schwein. Trans. Amer. Phil. Soc. 1834. Cooke, Bandb. p. 538; Micr. Fung. p. 200. Plowr. Ured. p. 266. Sacc. Syll. vii. 831. jEcidiospores. /Ecidia on extensive pallid spots on the leaves, sometimes almost covering them, arranged without order, elongate ; spores orange. 25—2 MSN c.KliMA On Arum triphyllv/m. In gardens, Melbury, 1863 (Rev. M. .). Herkeley). An intrndueed species, from North America. Caeoma Ari-italici Rud. Uredo Ari-italici Requien in Duby, Bot. Gall. ii. 899. Caeoma Art Winter Pilze, p. 256. C. Ari-italici Rud. in Linnsea, iv. 512. Sacc. Syll. Fung. vii. 868. Trans. Brit. Myc. Soc. i. 60. J-'.riiliospores. Caeomata hypophyllous, irregular, flattened, scattered or concentric, often confluent, without a peridium, orange-yellow ; spores round or elliptic, often somewhat angular, verrucose, orange, 15 — 30 x 15 — 20 /x. On Arum maculatum. Very rare; near Salisbury, April, 1897 (Mr E. J. Tatum). Found also in France and Germany. Requien's description is as follows : Uredo hypophylla, maculis Litis lutescenti-albidis sparsis, acervulis rufis, annulatiui digestis, orbicularibus ovatisve, compactis, planis, epidermide rupta cinctis, sporidiis dilute rufis, grossis, pellucidis, subglobosis. Phragmidium tuberculatum J. Miiller. This species is recorded for Britain by Sydow (Monographia, iii. 114), on the ground that it was distributed by Baxter, ■' Stirp. Crypt. Oxon." no. 37. There are specimens of this ixsiccatum both at Kew and in the British Museum, but in both these the fungus on Rose-leaves is typical Ph. disciflorum, almost all the teleutospores having six septa, not 3 — 5 as in Ph. tuberculatum. The latter species, being widely distributed in Europe, is likely to be found here, but the evidence of its occurrence is at present insufficient. There is in Herb. Kew an secidium on "Atriplex littoralis, Maldon, Essex, M. A. Irvine," June 1st, 1864. iEcidia hypo- phyllous, covering the whole leaf, densely crowded, shortly cylindrical, with a slightly torn margin. Of this nothing else seems to be known. EXCLUDED SPECIES 389 EXCLUDED SPECIES. I'miMViKs Pakxassi.e Cooke, Grevillea, vii. L34. p. 128. Plowr. Ured. This species does not exist. The secidiutn on Parnassia, which was at first assigned to it, is now known to belong to Puccinia uliginosa (q.v.), and the other stages are merely Uromyces Valerianae on Valeriana dioica, as has been pointed out by Sydow. The leaves issued by Cooke under this name (Exsicc. i. 74) are obviously the radical leaves of the Valerian (Fig. 290) which grows in similar places to those suitable for the Parnassia. In Sutton Park (Warwicks.) the two plants grow side by side, and when I first found the Uromyces there I put it down as U. Parnassiae. Rusty marks do frequently appear on the leaves of the Parnassia which are placed in herbaria as U. Parnassiae, but on examination no spores will be found in them. Fig. 290. Radical leaves of Va- leriana dioica (left) and Parnassia Jin I ust r is (right), nat. size. Uromyces Urtic^e Cooke, Grevillea, vii. 137. Plowr. Ured. p. 142. This also is a non-existent species. The origin of the mistake can now never be unravelled. Uromyces scutellatus Cooke, Grevillea, vii. 137. Plowr. Ured. p. 134. This is probably nothing but an error in identification. Puccinia asarixa Cooke, Haiidl). p. 504. Plowr. Ured. p. 202. This is merely Puccinia Fergussoni; the leaves on which the specimens are in Cooke's Exsicc. i. 110 are obviously those of Viola jxdustris. It is rather strange that the spores of this species are exceedingly like the spores of the continental P. asarina, and the sori are almost identical in appearance. 390 EXCLUDED SPECIES Puc< i\i \ Cl \Ni'i>ii\ \ < 'inn. in Berk. Engl. El. v. ::*■."■. -on Scabiosa succisa"; Plowr. Ured. p. ls">. Perhaps the host-plant was mistaken; no Puccinia on Scabiosa is known. X kxodochcs curtus Cooke, Micr. Fung. p. 201. Plowr. Ured. p. 228. Not a Oredine, but may be a species of Hyphomycetes. .lVinir.M P.AKi'.AUK.i: Cooke, Grevillea, x. 115. Plowr. Ured. p. 265. This record depends, so far as this country is concerned, od a single leaf which was sent to M. C. Cooke by a correspondent . who informed him (on the authority of another person) that the leaf belonged to Burbarea jiraecox. But on examination the leaf is seen to be one of the radical leaves of Lapsana communis, of which it has the peculiar hairs, and the fungus is obviously sEcidium Lapsanae. The true jE. Barbareae is quite different. The leaf is preserved in the Kew Herbarium. Mi ii'iUM ixcarceratcm B. et Br. Plowr. Ured. p. 207. Plowright's suspicion that this is merely Doassansia Sagit- tariae (on Sagittaria sagittifolia) is perfectly correct. jEcidium pseudo-coloixare J. Kiihn. Recorded in Plowright, Ured. p. 271, on "Abies pectinata, nordmanniana, amabilis,cephalonica. Lyme Regis, Mr Munro." Nothing else seems to be known of this as British ; most likely the specimens belonged to Calyptospora Goeppertiana, or were introduced on imported plants. jE. pseudo-columnare differs from the secidium of C. Goeppertiana in having oblong white spores which are larger, more irregular in form, unevenly warted or even smooth at the end (see Hedwigia, 1885. xxiv. 108). Uredo Oxalidis. There is a specimen bearing this name in the Herbarium of the British Museum, on leaves of Oxalis Acetosella (W. Phillips, Juh- 31, 1880) from " Orton " ; and a similar one, part of the same gathering, in the Kew Herbarium, from Orton Longueville, Hunts. (Rev. M. J. Berkeley, August, 1880). Both these are not Uredines, but show marks probably of insect-bites. GLOSSARY aculeate. Covered with needle-like projections. alveolate. Having depressions all over the surface, like a honeycomb. amphigenous. Growing upon both sides of a leaf. autcecious. Having all spore-fortns upon the same species of host. basipetal. Having each new part (e.g. spore) formed nearer to the base than the similar preceding one. brachymeiosis. A modification of meiosis in which the separation of the chromosomes is not preceded by a contraction of the nuclear material, capitate. Surmounted by a nearly globular head. chlorenchyma. Parenchymatous tissue containing chloroplasts. circinate. Arranged in a circle cuneate. Tapering downwards, with straight sides. dendritic. Having a branched form, like a tree. denticulate. Provided with small teeth-like projections. digitaliform. Having the shape of a finger of a glove. echinulate. Covered with spiny projections, ellipsoid. Having an oval outline, rounded equally at both ends. endokaryogamy. The intracellular fusion of nuclei after a series of conjugate divisions. epiphyllous. Growing on the upper side of a leaf. erumpent. Bursting through the tissues of the host, and becoming superficial, fusoid. Having the shape of a shuttle, tapering at each end. gamete. A cell specialised for reproducing the species, by fusing with another gamete. haustoria. Short mycelial branches which penetrate from the inter- cellular -paces into the cells of the host, and absorb their nutritive contents, hetercecious. Having some spore-forms upon one species of host and the others upon another species of a different genus. hypophyllous. < Growing on the underside of a leaf. infection. The successful attack of the mycelium upon the cells of the host. inoculation. The entry of a germ-tube into a host-plant. 392 GLOSSARY intercalary. Occupying a position between other bodies in a row. karyogamy. See endokaryogamy. laciniate. Torn into a ragged form at the edge. rneiosis. A special type of nuclear division by which the number of chromosomes in each daughter nucleus is reduced to half the number present in the nucleus before meiosis. metoecious. The same as hetercecious. monophyletic. Descended from a single species or closely allied group df species. pallid. Of the colour of fresh chamois leather. paraphysis. A more or less thread-like organ which grows by the side of the s j lores. plectenchyma. A kind of pseudo-parenchyma, formed by a mass of intcirw ined hyphse. polyphyletic. I a-M-ended, in distinct lines, from widely different ancestors. pulverulent. Having a powdery appearance from the loo>e >pores. pulvinate. Having the shape of a cushion. punctate. Marked with little dots, like pin-pricks. punctiform. Having the form of a small pin's head. reticulate. Covered with a network of lines. scrobiculate. Having the surface hollowed out into little shallow pits. semi-apogamy. A fusion of cells for reproduction, where one at least of the fusing cells is still more or less sexually specialised, but the cells are not of opposite sexes. If both cells represent female gametes, it may be called parthenogamy. trichogyne. A long hair-like projection from a female cell, suitable for arresting a passing male cell, truncate. As if cut off' at the top, with rather square corners, verrucose. Warted, i.e. covered with rounded elevation-. verruculose. Covered with minute rounded elevations. viable. Able to put forth a germ-tube. BIBLIOGRAPHY of Literature referred to, chiefly of recent works and additional to those quoted under the several species ALLEN, C. E. Die Keimung der Zygote Lei Coleochrete. Ber. Deutsch. Bot. Gesell. xxiii. 285 (1905). Arthur, J. C. North American Flora, vol. vii. parts 2, 3, Uredinales (1907—12). Bachmann, F. M. A New Type of Spermogonium and Fertilization in Collema. Annals Bot. xxvi. 747 (1912). Balls, W. L. Infection of Plants by Rust-Fungi. New Phytologist, iv. 18 (1905). DeBary, A. Recherches sur le developpement de quelques Champignons parasites. Ann. Sci. Nat. Bot. 4, xx. 83 (1863). Baur, E. Zur Frage nach der Sexualitat der Collemaceen. Ber. Deutsch. Bot. Gesell. xvi. 363 (1898). Biffex, R. H. Mendel's Laws of Inheritance and Wheat Breeding. Journal Agricult. Sci. i. 4 (1905). - Studies in the Inheritance of Disease Resistance, I. Journ. Agricult. Sci. ii. 109 (1907); II, ibid. iv. 421 (1912). Blackmax, V. H. On the Conditions of Teleutospore germination and of Sporidia formation in the Uredinea?. New Phytologist, ii. 10 (1903). On the Fertilisation, Alternation of Generations, and General Cytology of the Uredinese (preliminary notice). New Phytologist, iii. 23 (1904). On the Fertilisation etc. (as above). Annals Bot. xviii. 323 (1904). On the Relation of Fertilisation, "Apogamy" and "Parthenogenesis." New Phytologist, iii. 149 (1904). Blackmax, Y. H., and Fraser, Miss H. C. I. On the Sexuality and Development of the Ascocarp in Humaria granulata. Proc. Roy. Soc. B. lxxvii. 354 (1906). Further Studies on the Sexuality of the Uredinea). Annals Bot. xx. 3.") (1906). Blackmax, V. H., and Welsford, E. J. The Development of the Peri- thecium of Polystigma rubrum DC. Ann. Bot. xxvi. 761 (1912). Brooks, F. T. The Cytology of the Uredinese (review). New Phytologist, viii. 74 (1909). o!>4 liir.LIoGRAl'IlY r.RooKs, r. 'I'. The Life History of tlie Plum Rust in England. New Phytologist, \. 207 1911). Bubak, Fr. Lnfektionsversuche mit einigen (Jredineen. Centralbl. f. Bakter. 2, ix. 913 (1902 ; ibid. 2. xii. ill (1904); ibid. 2, xvi. L50 (1906). Buller, A. H. R. Researches on Fungi (1909). Butler, E. J. The Bearing "f Mendelism on the Susceptibility of Wheat to Rust. Journal Agricult. Sci. i. 361 (1905). Carleton, M. A. Cereal Rusts of the United States. U. S. Dept. Agric. Bulletin 16(1899). Investigations of Rusts. Ibid. Bulletin 63 (1904). Carruthers, D. Contributions to the Cytology of Helvella crispa. Ann. Hot. xxv. 243 (1911 . Christman, A. H. Sexual Reproduction of the Rusts. Botan. Gaz. xxxix. 267 (1905). The Nature and Development of the primary Uredospores. Trans. "Wisconsin Acad. Sci. xv. 517 1907). Alternation of Generations and the Morphology of the Spore-forms in Rusts. Bot. Gaz. xliv. 87 (1907). Clatjssen, P. Zur Entwicklungsgeschichte der Ascomyceten. Pyronema confluens. Zeitschr. f. Botanik, iv. 1, pi. 1—6 (1912). ( Viokk, M. C. Handbook of British Fungi (187 1 . Microscopic Fungi (fourth edition) (1878). Coons, G. H. Some Investigations of the Cedar Rust Fungus. Gymno- sporangium Juniperi-virginianse. 25th Ann. Report, Agric. Exper. Station, Univ. of Nebraska (1912 . Cruchet, P. Contribution a l'etude biologique de quelques Puccinies sur Labiees. Centralbl. f. Bakter. 2, xvii. 212 (1907 . Dietel, P. Ueber Rostpilze mit wiederholter Aecidienbildung. Flora, lxxxi. 395 (1895). Ueber die Arten der Gattung Phragmidium. Hedwigia, xliv. 112, 330. with one plate and two text-figures (1905). Dittschlag, E. Zur Kenutnis der Kernverhaltnisse von Puccinia Falcarise. Centralbl. f. Bakter. 2, xxviii. 473, f. 1—6, plates i— iii (1910). Duggar, B. M. Fungous Diseases of Plants (1909). Eriksson, J. Die Hauptergebnisse einer neuen Untersuchung fiber den Malvenrost, Puccinia Malvacearum. Centralbl. f. Bakter. 2, xxxi. 93 (1911). Fungoid Diseases of Agricultural Plant- 1912). Eriksson, J., und Hexxixg, E. Die Getreideroste, ihre Geschichte unci Natur, sowie Massregeln gegen dieselben (1896). Evans, I. B. P. Infection Phenomena in various Uredineae. Rep. Brit. A>soc. p. 595 10').-) . The Cereal Rusts. I. The Development of their Uredo mycelia. Ann. Bot. xxi. 44.3, pi. 40—3 (1907). BIBLIOGRAPHY 395 Evans, I. B. P. South African Cereal Busts, with Observations on the Problem of Breeding Rust-resistant Wheats. Journal Agricult. Sci. iv. 95 (1911). Faull, .1. II. The Cytology of Laboulbenia cbaetophora and L. Gyrini- darum. Ann. Bot. xxvi. 325, pi. 37—40 (1912). Fiscii ER, Ed. Ueber Beziehungen zwischen Uredineen welche alle Sporen- formeD besitzen und solchen von reduziertem Entwicklungsgang. Entwk-klungs-cscliichtliclie Untersuch. ii. Rostpilze, p. 109 (1898). Die Uredineen der Sehweiz (1904). Beitrage zur Entwickelunesgeschichte der Uredineen. Centralbl. f. Bakter. 2, xv. 227 (1906); ibid. xvii. 203 (1907). Eraser, H. C. I., and Brooks, W. E. St-J. Further studies on the Cytology of the Ascus. Ann. Bot. xxiii. 537 (1909). Frek.max, E. M. Experiments on the Brown Rust of Bromes (Puccinia dispersa). Ann. Bot. xvi. 487 (1902). Freemax, E. M., and Johxsox, E. C. The Rusts of Grain in the United States. U. S. Dept. Agric. Bulletin 216 (1911). Fromme, F. D. Sexual Fusions and Spore Development of the Flax Rust. Bull. Torr. Bot, Club, xxxix. 113, pi. 8, 9 (1912). Gibson, C. M. Xotes on Infection Experiments with various Uredinese. New Phytologist, iii. 184, pi. v, vi (1904). Grove, W. B. The Evolution of the Higher Uredine*. New Phytologist, March (1913). < Juillermond, A. Les Progres de la Cytologie des Champignons. Progressus Rei Botanicse, p. 389 (1913). Harper, R. A. Sexual Reproduction in Pyronenia confluens and the Morphology of the Ascocarp. Ann. Bot. xiv. 321, pi. 19—21 (1900). Harper, R. A., and Holdex, R. S. Nuclear Division and Nuclear Fusion in Coleosporium Sonchi-arvensis Lev. Trans. Wisconsin Acad. Sci. xiv. 63 (1903). Hasler. A. Kulturversuche mit Crepis- und Centaurea-Puccinien. Centralbl. f. Bakter. 2, xv. 257 (1906). Hoi-tmaxx, A. W. H. Zur Entwicklungsgeschichte von Endophyllum Sempervivi. Centralbl. f. Bakter. 2, xxxii. 137 (1911). 11 ime, H. H. Some Peculiarities in Puccinia Teleutospores. Bot. Gazette, p. 418 (1899). Eumphries, A. E., and Biffex, R. H. The Improvement of English Wheat. Journal Agricult. Sci. ii. 1 (1907). JACKT, E. Beitrage zur Kenntniss der Rostpilze. Centralbl. f. Bakter. 2, ix. 796 (1902). Der Chrysanthemum-Rost. II. /A/-/. 2, x. 369 (1903). Jaczewski, A. von. Studien iiber das Verhalten des Schwarzrostes des Getreides in Russland. Zeitschr. f. Pflanzenkr. xx. 321 (1910). Kakstex, ( 1. Die Entwicklung der Zygoteu von Spirogyra jugalis. Flora, xcix. 1, pi. 1 (1909). Kleiiahx, H. Die wirtswechselnden liostpilze (1904). -'i-'i; BIBLIOGRAPHY Kiuki;, W. Versuche mit lianunculaceen bewohnenden Aocidien. Cen- tralbl. t'. Bakter. 2, xv. 258 (1906 ; ibid. xvii. 208 (1907 . - Neue tnfektionsversuche mit I'romyrcs I >act ylidis. Centralbl. f. Bakter. 2, xxv. 430 (1909). Lewis, I. F. The Life-History of Grimthsia lionictiana. Ann. Hot. xxiii. 639, with pi. 49—53 (1909). Li t.\i an, B. V. Some < 'niitriliutions to the Life-History and Cytology of the Smuts. Trans. Wisconsin Acad. Sci. xvi. [it. 2, p. 1 l!)l 1910). McALPINE, D. The Rusts of Australia (1906). Magnus, 1'. Wcitere Mittheilung iibcr die auf Farnkrautern auftretenden Uredineen. Ber. Deutsch. Bot. Gesell. xix. 578 (1901). • Bemerkungen iiber einige Gattungen der Melampsoreen. Ber. Deutsch. Bot. Gesell. xxvii. 320 (1909). Maire, R. La Biologie des Uredinales. Progressus Rei Botanicae, iv. 109 (1911). Massee, G. On the Presence of Sexual Organs in iEcidium. Ann. Bot. ii. 47 (1888). - A Text-book of Plant Diseases caused by Cryptogamic Parasites (1 903). Diseases of Cultivated Plants and Trees (1910). Moreau, Mme Fernand. Sur l'existence d'une forme ecidienne uni- nucleee. Bull. Soc. Myc. France, xxvii. 489 (1911). Muhi.ethaler, F. Infcktionsversuche mit Rhamnus befallenden Kronen- rosten. Centralbl. f. Bakter. 2, xxx. 386 (1911). Muller, "Wilhelm. Versuche mit Uredineen auf Euphorbien und Hypericum. Centralbl. f. Bakter. 2, xvii. 210 (1907). Neumann, R. Ueber die Entwickelungsgeschichte der Aecidien u. Spermo- gonien der Uredineen. Hedwigia, xxxiii. 346 (1894). Olive, E. W. Sexual Cell Fusions and Vegetative Nuclear Divisions in the Rusts. Ann. Bot. xxii. 331 (1908). Origin of Hetercecism in the Rusts. Phytopathology, i. 139 (1911). Plowright, C. B. A Monograph of the British Uredineoe and Ustilaginese (1889). Pole-Evans. See Evans, I. B. P. Pritchard, F. J. A Preliminary Report on the yearly Origin and Dissemi- nation of Puccinia Graminis, with plate. Bot. Gazette, lii. 169(1911). The Wintering of Puccinia Graminis Tritici E. et H., and the Infection of Wheat through the Seed. Phytopathology, i. 150 (1911). Probst, R. Die Spezialisation der Puccinia Hieracii. Centralbl. f. Bakter. 2, xxii. 676 (1909). Rawitscher, T. Beitrage zur Kenntnis der Ustilagineen. Zeitschr. f. Botanik (1912). Richards, H. M. On the Structure and Development of Choreocolax Polysiphonise. Proc. Amer. Acad. Arts, new ser. xviii. 46 (1891). On some Points in the development of the iEcidia. Proc. Amer. Acad. xxxi. 2.").") (1896). BIBLIOGRAPHY 397 Schneider, 0. Weitere Versuche mit schweizerischen Weidenmelam- psoren. Centralbl. f. Bakter. 2, xv. 232 (1906). - Experimentelle Untersuchungen iiber sehweizerische Weidenrost- pilze. Centralbl, f. Bakter. 2, xvi. 71, 159, 192 (1906). Semadeni, 0. Beitrage zur Kenntniss der Umbelliferen bewohnenden Puccinien. Centralbl. f. Bakter. 2, xiii. 73—81 etc. (1904). Sharp, Lester W. Nuclear Phenomena in Puecinia Podophylli. Bot. Gazette, li. 463 I L911). Spauxding, P. The Blister Rust of White Pine. U. S Dept. Agric. Bulletin 206 (1911). Stahl, E. Beitrage zur Entwickelungsgeschichte der Flechten (1877). Sturch, H. H. Harveyella mirabilis. Ann. Bot. xiii. 83, pi. 3, 4 (1899). Sydow, P. et H. Monographic Uredinearum. Vols. I, II, Vol. Ill, pt. 1 (1904—12). T wrexhaus, J. J. A Contribution to our Knowledge of the Morphology and Life- History of Puecinia Malvacearum Mont. Phytopathology, i. 55(1911). Troxdle, A. Ueber die Kopulation und Keimung von Spirogyra. Bot. Zeit. lxv. 187 (1907). Ward, H. M. On the Relations between Host and Parasite in the Broines and their Brown Rust, Puecinia dispersa (Erikss.). Ann. Bot. xvi. 233 fl 902). On the Histology of Uredo dispersa Erikss. and the "Mycoplasm" Hypothesis. Phil. Trans. Roy. Soc. B. exevi. 29 (1903). Further Observations on the Brown Rust of the Broines, Puecinia dispersa, and its adaptive parasitism. Annal. Mycol. i. 132 (1903). Recent Researches on the Parasitism of Fungi. Ann. Bot. xix. 1 (1905), with a very full bibliography. Werth, E. Zur Kenntnis des Sempervivum-Rostes. Centralbl. f. Bakter. 2, xxxvi. 395 (1913). Wkrth, E., und Ludwigs, K. Zur Sporenbildung bei Rost- u. Brandpilze (Ustilago antherarum und Puecinia Malvacearum). Ber. Deutsch. Bot. Gesell. xxx. 522 "(1912). Weir, J. R. Review of the Characteristics of the Uredinea?, with notes of a variation in the promycelium of Coleosporium Pulsatilla? (Str.). New Pbytologist, xi. 12!) (1912). Wurth, Th. Rubiaceen-bewohnende Puccinien vom Typus der P. Galii. Centralbl. f. Bakter. 2, xiv. 209, 309 (1909). Zach, Franz. Cytologische ' Untersuchungen an den Rostflecken des Getreides und die Mycoplasmatheorie J. Eriksson's. Sitzungsber. Kais. Akad. Wissensch. Wien, cxix. 307 (1910). Zalewski, A. Ueber Sporenabschniirung und Sporenabfallen bei den Pilzen. Flora, lwi. 228, 2 1'.) (1883). [NDEX OF EOST-PLAJSTTS MENTIONED Abies alba = A. pectii aniabilis ri'.H) cephalonica 390 Nordmanniana '■'^'■'<. 390 pectinata 361, 363, 367, 373, 390 Aehilh a Millefolium 131 Ptarmica 131 Adianturn Capillus- Veneris 378 Adoxa Moschatellina 161, 162, 205 JEgopodium Podagraria 185 Jjimsi Cynapium 190 Agrimonia Eupatoria 365 Agropyron 282 caninum 251, 259, 263 repens 251, 254. 259, 263, 283 Agrostis alba 251, 254, 276 caniua 2-31 stolonifera 2-51 vulgaris 251, 254, 276 Aira casspitosa 251, 265 flexuosa 251, 265 Ajuga reptans 172 Alchemilla vulgaris 106 Allium Cepa 236, 345, 348 Schcenoprasum 122, 236 Scorodoprasurn 122, 236 sphaH'ocephaluni 122 ursinum 268, 345, 346, 348 Alopecurus pratensis 251, 256, 271 Alsine rubra 111 Althaea rosea 207 Amygdalus communis 209 Anchusa arvensis 201 Anemone coronaria 209 decapetala 210 nemorosa 209, 210, 330 ranunculoides 209 Angelica silvestris 193, 225 Anthoxanthum odoratum 270 Anthriscus silvestris 195 Anthyllis Vulneraria 95 Apium graveolena L85 Aquilegia vulgaris 276 Arctium Lappa 138 Arenaria trimrvis 219 Armeria maritima 89 Arrhenatherum elatius 251, 250. 285 Artemisia Absinthium 134 campestris 134 maritima 131 vulgaris 134 Arum maculatuin 209, 388 triphyllum 388 Asparagus officinalis 234 Asperula odorata 164 Asi:>idium spinulosum 378 Aster Tripolium 130. 248 Atriplex littoralis 388 Avena elatior (see Arrhenatherum) fatua 251, 250 pratensis 256 sativa 251, 256 Barbarea prascox 390 Bartsia Odontites 326 Bellis perennis 237 Berberis vulgaris 251, 285 Beta maritima 113 vulgaris 113 Betonica officinalis 174 Betula alba 359 Blechnuni Spicant 377 Brachypodium pinnatum 251, 281 silvaticum 255, 259, 281 Bromus 202 arvensis 251 asper 251 erectus 251 giganteus 251 mollis 251. 259 1XDEX OF HOST-PLANTS MENTIONED 399 Bromus {rout.) secalinus 2">1 sterilis 262 Bu nium Bulbocastanum 186 flexuosum 187 Bupleurum Ealcatum 190 tenuissimum 189 Buxus sempervirens 206 < acalia hastata 325 Buaveolens 325 Calamagrostis lanceolata 25 4 Calamintha Clinopodium 171 officinalis 171 Caltba palustris 217, 218 Campanula jdomerata 329 latifolia 329 Kapunculus 160 rotundifolia 160, 329 Trachelium 329 turbinata 329 Carduus arvensis 145 crispus 141, 144 nutans 141 pycnoeephalus 142 Carex acuta 243 acutifonuis 242, 243 arenaria 246, 247, 248 binervis 2 12 caespitosa 243 caryophyllea 246 Davalliana 244 dioica 241 extensa 248 fulva 249 Goodenovii 243, 249 hirta 242 montana 133 muricata 152 paludosa 242, 243 panicea 249 pendula 242 preecox 246 Pseudocyperus 212 remota 2 15 riparia 242, 243 stricta 243, 249 vulgaris 249, 250 Carlina vulgaris 138 Carpiuus Betulus 360 Carina Bulbocastanum 186 Carui 225 Carum {cont. I majus 187 Cattlcya Dowiana 381 ( lentanrea Cyan us 141 montana 141 nigra 140, 248 Scabiosa 140 Cerastium arvense 361 triviale 361 viscosum 361 Chserophyllum temulum 195 Cbelidonium majus 353 Chrysanthemum indieum 132 Leucanthemum 133 sinense 132 Chrysosplenium aiternifolium 214 oppositifoliurn 214 Cicborium Intybus 148 Cicuta virosa 184 Cineraria 321 Circsea alpina 198 intermedia 366 lutetiana 198, 3613 Cirsium arvense 14"> heterophyllum 146 lanceolatum 113, 144 palustre 143, 244 pratense 143, 244 Clematis Vitalba 282 Cnicus (see Cirsium) Colchicum speciosum 122 Conium maculatum 196 Conopodium denudatum 187, 225 Convallaria majalis 267 Convolvulus sepium 176 Corydalis 353 Cotoneaster integenimus 307 Cotyledon Umbilicus 211 Crataegus monogyna 306, 307 Oxyacantha 306, 307 Crepis paludosa 158 virens 157 Cydonia vulgaris 306, 307 Cystopteris fragilis 376 Dactylis glomerata 126, 251, 254 Deschampsia (see Aira) Dianthus barbatus 109, 219 Caryophyllus 109 chinensis 109 Digraphis arundinacea 266 too INDEX OF HOST-PLANTS MKXTloXED Echium vulgare 265 Elyinu- an aaj ius 259 Empetrum Digram 311 Endymion non-scriptam 121 Epidendrum vitellinam 383 Epilobium angastifolium 367 hirsutum 199 montanum L99 palustre 200, 367 tetragonum 199 Eranthis biemalis 209 Ervum hirsntum 96 Euonymus europsens 340 Euphorbia amygdaloides 334 Cyparissiiis 93, 95, 100 exigua 103, 35 1 Helioscopia 354 Peplus 354 silvatica 334 Euphrasia officinalis 326 Faba vulgaris 97 F( -tuca duriuscula "257 elatior 256, 284 My urns 251 ovina 251, 257 rubra 251 silvatica 254 Fragaria vesca 290 Fritillaria Meleagris 119 Gagea lutea 120 Galanthus nivalis 345 Galium Aparine 168 Cruciata 165, 167 Mollugo 165 palustre 165 saxatile 165, 167 uliginosum 165 verum 165, 370 Gentiana acaulis 179 Geranium dissectum l11! molle 104, 105, 228, 229 phffium 228 pra tense 104, 228 pusillum 104, 105, 229 pyrenaicum 104, 105 rotundifolium 104, 229 silvaticum 104, 228 Glaux maritima 125 Glechoma hederacea 174 Glyceria aquatica 256 fluitans 279 Goodyera repens 344 Gymnadenia conopsea 3 1 1 Gypsophila elegans 219 ll'llcborus foetidne 386 viridis 386 Eepatica acntiloba 209 Heracleum Sphondylium 194 Hieraciam boreale 159 murorum 159 Pilosella 159 nmbellatum 159 Holcus lanatus 251. 254, 256, 263 mollis 251, 254. 256, 263 Hordeum murinum 251, 259 pratense 251 vulgare 251, 259, 265 Hydrocotyle vulgaris 182 Hypericum Androsaemum 355 humifusum 355 montanum 355 perforatum 355 pulchrum 355 Hypochoeris radicata 149 Impatiens fulva 205 Xoli-tangere 205 Iris fcetidissima 231 Pseudacorus 231 160 123 Jasione montana Jonquil 232 Juucus obtusinorus Juniperus communis 306, 308 Sabina 307, 309 Koeleria cristata 286 Lactuca muralis 152 Lapsana communis 147 Larix europsa 339, 341, 349, 350, 359 Lathyrus macrorrhizus 99 montanus 99 pratensis 90, 97, 100 vein us 98 Leontodon autumnalis 150 birtus (see Additions, p. xii) hispidus 150 Lilium candidum 119 Limnanthemum nymphaeoides 240 Limonium (see Statice) INDEX OF HOST-PLANTS MENTIONED 401 Linum catbarticum 356 usitatiBsimum 356 Listera cordata 344 ovata 344 Loliuni perenne 251, 256 Lunicera Periclymenum '257 Lotus angustissimus 95 corniculatus 95 Luzula campestris 237, 239 maxima 23'J pilosa 239 silvatica 237 and p. xii Lycaste Skinneri 383 Lychnis diurna (dioica) 219, 222 vespertina 219 Lycopsis arvensis 2(51 Mahonia Aquii'olium 251 Malva silvestris 207 Malveas 207 Mangels 113 Medicago 93 Melampyrum 277 arvense 327 pratense 327 Mentha aquatica 171 arvensis 171 eitrata 171 rotundifolia 171 silvestris 171 viridis 171 Mercurialis perennis 352 Mespilus germanica 306, 307 Milium effusum 251 Molinia ccerulea 251, 277 Myrrhis odorata 195 Narcissus poeticus 232 Pseudonarcissus 232 Nepeta Glechoina 174 Nvinphoides peltatum 240 Oncidium Forbesii 383 varicosum 383 Orchis latifolia 268, 344 maculata 344 Origanum vulgare 171 Ornithogalum umbellatum 235 Orobus tuberosum 99 Oxalis Acetosella 390 Oxyria digyna (reniformis) 224 G. U. Paonia officinalis 314 Paris quadrit'olia 267 Parnassia palustris 250, 389 Pedicularis palustris 249 Petasites officinalis 324 Petroselinum sativum 190 Peucedanum palustre 193 Phajus Wallichii 382 Phalaris arundinacea 254, 266 canadensis 251 Phaseolus vulgaris 102 Phillyrea latifolia 332 media 333 Phleum nodosum 284 pratense 284 Phragmites communis 272, 274, 275 Picea excelsa 313, 369 Pimpinella magna 188 Saxifraga 188 Pin us austriaca 321 Cembra 317 montana 325 monticola 317 silvestris 314, 321, 323, 324, 325, 326, 327, 329, 351 Strobus 317 Pisum sativum 97, 100 Plautago lanceolata 385 major 385 Poa annua 127, 279 aquatica 256 compressa 251 fluitans 279 nemoralis 127, 251, 279 pratensis 127, 251, 279 trivialis 127, 251, 279 Polygonum amphibium 228 aviculare 117 J5istorta 225 Convolvulus 229 lapathifolium 228 viviparum 226 Polypodium Dryopteris 375 Phegopteris 380 Kobertianum 375 vulgare 377 I'uplllllS alba 350, 351, 352 balsamifera var. 348, 349, 350 canadensis 349 canescens 351 26 402 INDEX OF IIOST-l'LANTS MENTIONED Populus (cont.) nigra (scrntina) :ils, lil'.l (iiiiariensis (candicans) 3 19 pyramidalia (italica) 349 tremula 350, 351, 352 Potentilla aigentea 292 • recta 302 Fragariastnim 290 sterilis 290 Tormentilla 302 verna 292 Poterium Sanguisorba 293, 387 Primula acaulis (vulgaris) 180 elatior 180 veris 180 Prunella vulgaris 174, 278 Prunus domestica 209 insititia 209 Padus 369 spinosa 209 Pulicaria dysenterica 123 Puhnouaria montana 262 Pyrola minor 312, 368 rotundifolia 312, 368 Pyrus Aucuparia 308, 330 communis 306, 307, 308 torminalis, etc. 331 Quercus pedunculata 316 Kanunculus acris 126, 271 auricomus 127 bulbosus 126, 127, 272 Ficaria 107, 115, 127 Lingua 387 repens 126, 127, 272 Pih a mn us catharticus 209, 256 Frangula 2-54 Eheum officinale 274 Rhinanthus Crista-galli 326 Ehododendron dahuricum 385 ferrugineum 385 hirsutum 384 Eibes alpinum 343 • Orossularia 243, 342, 343 nigrum 243, 317, 342 rubrum 212, 317, 342 Eosa alpina 295 canina 294, 388 spinosissima 294 Eubus cresius 297 and p. xii fruticosus 296, 297, 301 and p. xii Idaeus 299 saxatilis 297 Eumex Acetosa 116, 223, 275 Acetosella 116, 118 acutus 274 conglomerates 115, 274 crispus 115, 271 Hydrolapathum 115, 274 nemorosus 115 obtusifolius 115, 274 Sagina nodosa 219 procumbens 221 Sagittaria sagittifolia 390 Salicornia europiea 114 herbacea 114 Salix alba 346 aurita 339, 340, 341, 344 Caprea 339, 340, 341 cinerea 340, 341 fragilis 341, 345 herbacea 346 nigricans 341 pentandra 345 purpurea 341, 343 repens 344 reticulata 341 Smithiana 341 viminalis 341, 342 Sanguisorba officinalis 303 Sanicula europaja 183 Saponaria ocymoides 109 Saxifraga Aizoon 213 Cotyledon 213 elatior 214 granulata 213, 357 longifolia 213 stellaris 213 umbrosa 213 Scabiosa 390 Scilla bifolia 121 campanulata 121 nutans 121 Scirpus lacustris 240 maritimus 125 Scolopendrium vulgare 378 Scrophularia aquatica 87 nodosa 87 Secale cereale 252, 259, 261 INDEX OF HOST- PLANTS .MENTIONED 403 Sedum acre 287 reflexum 287 Bhodiola 211 roseum 211 Sempervivum araohnoideum 335 caleareum 335 globiferum 335 montanum 335 tectoium 335 Senecio aquaticus 136 Jacobasa 136, 137, 247, 321 palustris 321 pulcher 321 sarracenicus 321 silvaticus 321 viscosus 321 vulgaris 321 Serratula tinctoi'ia 139 Sherardia 370 Silaus pratensis 191 Silene Cuoabalus 110, 222 inflata 110, 222 latifolia 110, 222 maritima 110 Sniyrniuin Olusatrum 197 Soldanella alpina 181 Solidago Virgaurea 130 Sonchus arvensis 156, 325 asper 156, 325 oleraceus 156, 325 Spergula arvensis 219, 220 Spergularia rubra 111 Spiraea Filipendula 288 Ulmaria 288 Spiranthes 385 Stachys Betonica 174 Statice Limonium 88 Stellaria graminea 361 Holostea 219 media 219, 361 uliginosa 219 Suaeda fruticosa 112 maritima 112 Symphytum officinale 262, 363 Tanacetum vulgare 135 Taraxacum officinale 153, 154, 245 Teucrium Chamaidrys 175 Scorodonia 175 Thalictrum 209 flavum 215, 283 minus 215, 283 Thesium humifusum 230 Thrincia hirta (see Additions, p. xii) Thymus Serpyllum 173 Tragopogon pratensis 151 Trifolium agrarium 93 hybridum 90 incarnatum 90 medium 90 minus 93 pratense 90 procumbens 93 repens 91, 92 Trisetum flavescens 252, 264 Triticum (see also Agropyron) glaucum 282 junceum 282 vulgare 252, 259, 263 Tropseolum aduncum 386 canariense 386 minus 386 Tussilago Farfara 279, 323 Urtica dioica 242, 389 Vaccinium Myrtillus 371, 373 uliginosum 371 Vitis-idasa 371, 373 Valeriana dioica 86, 389 officinalis 86 Veronica alpina 170 montana 169 officinalis 170 Vicia Cracca 97 birsuta 96 sativa 97 sepiirm 97 Vinca major 177 minor 177 Viola canina 201 cornuta 203 hirta 201 lutea 203 odorata 201 palustris 204, 389 Riviniana 201 silvestris 201 tricolor 203 CKXKKAL INDEX Synonyms and specie* incidentally mentioned are in italics. }llien more than one page is given, the one on which the description of the species occurs is printed in heavy type. aecidiospores 4 • iium 3, 30 / CIDIUM 386 nlii, tinum A. et S. 384 Adoxae Opiz 1G0 albescens Grev. 162 Allii Grev. 268 Anchusae E. et H. 261 Aquilegiae Pers. 275 argentatum Schultz 204 Ari Desm. 269 Asperifolii Pers. 261 aviculariae Kze. 117 Barbareae Cooke 390 Behenis DC. 109, 222 Berberidis Gmel. 250 Bulbocastani Cum. 186 Bunii DC. 186, 188 Buniiv&T.Smyrnii-OlusatriDG, 197 Calthae Grev. 216 Gathartici Sebum. 255 Chenopodii Duby 111 Cirsii DC. 244 Clematidis DC. 281 columnare A. et S. 372 Compositarum var. Bellidis DC. 236 Gompositarwm var. Jacobaeae Cooke 246 Compositarum var. Prenanthis Cooke 157 Compositarum var. Tussilaginis Cooke 278 conorum-Picese J?eess 313 Convallariae Sebum. 267 crassum Pers. 253, 255 crassum var. Ficariae Pers. 127 trassum var. Periclymeni Cooke 257 crassum var. Phillyreae Cooke 332 ( yparissiae DC. 99 depauperans Vize 202 Dracontii Scftic. 387 elatinum A. et S. 360 Epilobii DC. 198 AYt/ Wallr. 96 Euphorbiae Gmel. 102 Euphorbiae Pers. 333 MCIDIUM (cont.) Euphorbiae-silvaticae DC. 333 Falcariae var. Bupleuri-falcati DC. IS'.I Frangulae Schum. 253 fuscum Pers. 215 Gahi A. et S. 164 Gentianae Jacz. 178 Gmom DC. 103, 227 GZaucis D. et M. 124 graveolens Shut. 252, 284 Grevillei Grove 152 Grossulariae DC. 242 Hellebori Fisch. 386 incarceration B. et Br. 390 Jacobaeae Grev. 246 Lapsanae Schultz 147 leucospermum DC. 329 Melampyri K. et S. 276 Menthae DC. 170 Mespili DC. 30(5 Nymphoidis DC. 239 Orchidearum Desm. 268 ornithogaleum Bub. 235 Oro&i Pers. 99 Parnassiae Grav. 249 Pedicularis Libosch 248 Periclymeni Schum. 257 Phaseolorum Wallr. 101 Phillyreae DC. 332 Pimpinellae Kirch. 188 Poterii Cooke 387 Prenanthis Pers. 151 Primulae DC. 179 pseudo-columnare Plowr. 372, 390 Pulmonariae Thiim. 262 punctatum Pers. 207, 331 I',irolae DC. 312 Pyrolae Gmel. 367 quadrifidum DC. 207 Ranunculacearum'DG. 125, 127, 271 Ranunculacearum var. Aquilegiae DC. 275 Ranunculacearum var. Clematidis Cooke 281 Banunculacearum rar. Linguae 387 GENERA], INDEX 405 JECIDIVM (mnt.) Ranunculacearum var. Thalictri Cooke 282 Ranunculi-acris Pers. 125 rubellum Gmel. 273, 271 Salicorniae DC. 11 1 sitiiiiitinolciititiii Lindr. 104, '227 Saniculae ('arm. 182 Scrophitluriuc DC. 87 Senecionis Fisch. 246 Soldanellae Horn. 180 Sonchi Johnst. 155 Statices Desm. 88 strobilinum Wint. 368 Suaedae Thiim. Ill Symphyti Thiim. 262 Taraxaci K. et S. 245 77„'.v/7 Desv. 229 Tragopogi Pers. 150 Tragopogonis Cooke 150 Tussilaginis Gmel. 278 Urticac Solium. 241 Valerianearum Duby 86 Violae Schum. 200 zonule Duby 123 alternation of generations 27 amphispores 11, 34 ascogonium 22 Ascomycetes 77, 80 Ascophora disciflora Tode 293 Ascospora Scolopendrii Fckl. 378 azygospores 23, 31 Barclayella 14, 80 basal cells 18 Basidiomycetes 29 basidiospores 14, 37 basidiospores, germination of 15, 38 basidium 14, 37 basidium, internal 28, 39, 77, 318, 329, 332 Blackman 17 Blackman and Fraser 18 Blackman and Welsford 25 brachymeiosis 82 bridging-species 65, 66, 71 CMOMA 30, 388 Alliorum Link 269, 344, 347 Ari-italici Had. 388 Armeriae Schl. 89 Euonymi Plowr. 339 Galii Link 370 Ilydrocolnle* Link 181 Laricis Plowr. 338, 348, 349 l.ilii Link 118 Mercurialis Link 351 oblongatum Link 238 obtegens Link 145 Orchitis Wint. 268, 343 CMOMA {rout.) pinitorquum A. Br. 350 Ribesii Dink 342 Saxifragae Wint. 213, 357 CALYPTOSPORA 74, 372 columnaris Kiibn 372 ( iocppcrl iana Kiilin 59, 363, 36 I, 367, 372 Christman 18 CHRYSOMYXA 75, 310 albida Kiibn 300 Empetri Pers. 311 PyrolaB Rost. 312 Rhododendri De By. 311, 384 Woronini Tranz. 311 classification 73 Claussen 81 Coleosporiaceae 73, 75, 318 COLEOSPORIUM 75, 319 CacaliaE! Fckl. 325 Campanula? Lev. 328 Euphrasia) Wint 326 Melampyri Karst. 327 miniatum Pers. 293 ochraceum Fckl. 364 Petasitis Lev. 323 pingue Lev. 293 Rbinanthacearum Lev. 326 Senecionis Fr. 320 Sonchi Lev. 156, 322, 323, 324, 325 Symphyti 363 Tussilaginis Tul. 322 Collemaceas 25, 77 conjugate division 5 Cronartiacea? 73, 75, 310 CRONARTIUM 75, 313 asclepiadeum Fr. 313 asclepiadeum var. Quercnum Berk. 315 fiaccidum Wint. 314 Paeoniae Cast. 313 Quercus Arth. 315 Quercuum Miy. 315 ribicola F. de W. 55, 316 Darluca Filum Cast. 33, 178 dikaryon 5 Dittschlag 19 Doassansia Sagittariae Fisch. 390 Endophyllacea) 73, 75, 333 ENDOPHYLLUM 75, 333 Euphorbiae Plowr. 333 Euphorbke-silvaticse Wint. 333 leucospermum Sopp. 331 Sedi Lev. 287 Sompervivi De By. 53, 335 Epitea Baryi B. et Br. 280 Fairy Ring of Carnations 109 26—3 400 GENERAL IN'DliX Fromme 20 fusion of urn l< i 12. 28 gametophyte 27 germ-pores 7, 76" germ-tube 8 GYMNOSl'OBAXliHM 7.".. :;0J bermudianum Earle 304 clavariaeforme DC. 51, 304 confusum Plowr. 306 coniutinii Artll. 307 Jmiiperi LfraA; 307 juniperinum Fr. 307 Sabina: M7h/. 308 haustoria 8 HEMILEIA 381 amrricana lints. 381 Oncidii G. e« If. 383 Phaji Syd. 382 hetercecism 16 Heterosporium 109 heterotype mitosis 26, 28 Hoffmann 20 HYALOPSORA 74, 373 Adianti-Capilli-Veneris Syd. 378 Aspidiotus Magn. 374 Polypodii Magn. 375 Polypodii-Dryopteridis Magn. 374 immunity 71 inoculation 63 intercalary cell 6, 10 intercalation 5 KUEHNEOLA 75, 299 albida Magn. 300 obtusa Arth. 301 Tormentilke Arth. 301 Uredinis Arth. 300 Kurssanow 18 Laboulbenia 77, 81 lactic acid, action of 84, 289, 291, 297, 298 LECYTHEA Baryi Berk. 280 epitea Lev. 340 Euphorbiae Lev. 353 gyrosa Berk. 298 Lini Berk. 356 Poterii Lev. 292 Rosae L£v. 293 Ruborum Lev. 295 Licea strobilina A. et S. 368 Lutman 82 Lycoperdon caryophyllinum Schr. 108 manipulation 84 Melampsoraceae 73, 74, 336 MELAMPSORA 74, 336 aecidioides Plowr. :'>">1 Allii-fragilis Kleb. 344. 346 Allii-populina Kleb. 346, 347 Allii-Salicis-albae KLh. 345 arctica Rost. '■'< 16 betulina Desm. 358 ( 'erastii Wint. 361 Circaeae Schrot. 365 elatina Arth. 361 epitea Thiim. 340 Euonymi-Caprearum Kleb. 339 Euphorbiae ( 'ast. 353 farinosa Schrot. 338 Galii Wint. 370 Helioscopiae Wint. 353 Hypericorum Wint. 354 Klebahni Bub. 350, 353 Larici-Caprearum Kleb. 338 Larici-epitea Fiscli. 340, 343 Larici-pentandrae Kleb. 345 Larici-populina Kleb. 348 Larici-tremulffi Tul. 349 Lini Desm. 355 liniperda Korn. 356 Magnusiana Wagn. 350, 353 mixta Plowr. 343 Orchidi-repentis A'Zei. 343 P«Ji Cooke 369 pinitorqua Rost. 57, 350 populina Lev. 347, 348 pustulata Schrot. 366 Pyrolae Schrot. 367 Quercus Schrot. 315 Ribesii-auritae Kleb. 343 Ribesii-pnrpureas Kleb. 342 Ribesii-viminalis Kleb. 342 Rostrupii Wagn. 351 salicina Lev. 338 Saxifragarum Schrot. 357 TremulsB Tul. 349 ]'i Arth. 124 Trifolii Arth. 91 nuclear division 25 OCHROPSORA Sorbi Diet. 75, 329 Olive 18 Perichaena strobilina Fr. 368 PERIDERMIUM 31 acicolum Link 320 Boudieri Fisch. 323 columnar e, Cooke 372 conorum-Piceae 313 Coram Kleb. 314 deformans Tub. 315 elatinum Link 360 Fischeri Kleb. 324 giganteum Tub. 315 haricis Kleb. 359 Magnusianum Fisch. 325 Magnusii Wagn. 325 oblohgisporium Fckl. 328 Pini Chev. 320 Plowrightii Kleb. 322 Rostrupii Fisch. 328 Soraueri Kleb. 327 StoAiw Kleb. 326 ,S7 /,-/-/ Kleb. 316, 359 peridium 3, 6, 30, 79 Phelonitis strobilina Pers. 368 PHBAGMIDIUM 75, 289 acuminatum Fr. 2(.)2 albidum Lagh. 300 bulbosum Schl. 295 hull, it, nu West. 293 carbonarium Wint. 303 disciflorum James 293 Fragariastri Schr'ot. 290 fusiforme Schr'ot. 294 gracile Cooke 298 iiiiittius Arth. 299 mucronatum Fr. 293 obtusum Link 290 Potentillse A'ars*. 291 Potentillae-canadensis Diet. 301 Rosae-ulpinue Wint. 294 PHRAGMIDIUM (co»«.) Bubi JPmi*. 297 Rubi-IdsBi A"«r.sf. 298 Uubi-saxatilis Lirp 297 Sanguisorbffl Schr'ot. 292 siihi-orticiiun Wint. 293 Tormentillae Fckl. 301 tuberculatum Mull. 388 violaceum Wint. 295 PHYLLACHORA graminis 127 7 7w/ 349 phylogeny 75 plectenchyma 2 PODISOMA Juniperi Fr. 304 Sabinae Fr. 308 promycelium 14 Protomyces filicinus Niessl 379 Pseudopeziza Trifolii Fckl. 92 PUCCINIA 75, 128 Absinthii DC. 134 Acetosaj K'orn. 223 Actaeae-Agropyri Fisch. 283 acuminata Fckl. 167 Adoxas fl^edw;. /. 160, 162 Mcidii-Leucanthemi Fisch. 133 JEcidii-Melampyri Liro 276 iEgopodii Mari. 185 fegra Grove 202 .Fthusffi Mart. 190 Agropyri 23. et 23. 281 agropyrina Eriks. 263 Agrostidis Plowr. 275 albescens PZow. 161, 162, 205 Allii-Phalaridis Kleb. 268 ambigua Lagh. 168 Andersoni B. >\ 146 Anemones Pers. 215 Anemones var. Betonicae A. et S. 174 Angelica? FcW. 192 Angelicae-Bistortae Kleb. 225 annularis ,S'c/f/. 175 Anthoxanthi FcR 269 Apii Desm. 184 Arenariae Wint. 218, 221 arenariicola Plowr. 247 argentata JFini. 161, 163, 204 Ari-Phalaridis Kleb. 269 Arrhenatheri Eriks. 284 arundinucea Hedw. 273 asarina Cooke 204, 389 Asparagi DC. 233 Asperulae Fckl. 163 Asperulae-odoratas Wurth 163 ,lx/,-r/x Duby 129 Asteris var. ( 'hrysanthemi-Leucan- themi Massal. L33 Bardana? Cord. 138 LOS <; i:\eual i\i)i:x IMCCIXIA {rant.) Baryi Wint. 280 Behenia OMft 220, 222 Betonica DC. 171 Bistortae DC. 225, 226 bromina Eriks. 262 bromina, specialisation of 70 Brunellarum-Moliniae Cruch. '277 Bulbocastam Fckl. L86 bullaria Link 193, 196 bullata Wint. 190, 191, 193, 196 Burnt Wint. 187 Bupleuri /.'(/7 faHens Cooke 90, 91 Fergussoni /-'. <■? /;/•. 203 Festaca /'/,,,o-. -.'57 / ragariastri DC. 290 fuBca Wint. 215, :;:;i galatica Syd. 1 12 <;,//// Scliw. 163. 161, IDs Galiorum Link 164 Gentians Linft 178 Glechomatis DC. 173 glomerata Grew. 136. 320 glumarum E. et 11. 258 glumarum, specialisation of 67 graminis Pers. 41, 250 graminis, specialisation of 64 graminis var. Arundinis Cooke 271 Heraclei Gn-r. 194 Hieracii Mart. 137, 138, 141, 148, 149, 158 holcina Eriks. 263 Hordei Fckl. 259 Hydrocotyles Cooke 181 Hypochceridis Oud. 148 intermixta Friend 267 Iridis Wallr. 230 Lapsanae Fckl. 147, 157 Leontodontis Jacky 149 Leucanthemi Pass. 133 Liliacearum D»^// 234 linearis Bob. 266 Lolii Niels. 255 Lolii, specialisation of 68 longissima, Schrot. 286 Luzulae Lib. 238 Lychnidearum Link 218, 221 Magnusiana K'orn. 128, 271 major 7>/Vf. 157 Malvacearum Mont. 48, 206 ■mammillata Schrot. 225 M.nthae Pers. 170 Millefolii V-H7. 131 mixta f. simplicior Korn. 121 Moehringiae Fckl. 219 Molinise TuZ. 276 mucronata var. J2u&£ Pers. 297 nemoralis Juel 276 neurophila De Toni 92 Noli-tangeris Cord. 204 oblongata Jf'/Hf. 238 obscura Schrot. 236 obtegens 2'»/. 145 Opusii Bub. 152 Orchidearum-Phalaridis Z7e&. 268, 344 Oxyriae Fckl. 224 GENERAL INDEX 400 ITCCIXIA (cont.) paliformis Fell. 2^(5 paludosa PI out. 248 I'.iridis Plowr. 207 Pazschkei Diet. 213 perplexans Ploirr. 121), 270 perplexansf. drrJienatheriJZleb. 284 persistens Plotrr. 282 Petro&elini Lindr. L90 Phalaridis 77oin\ 209 PJiaseoli var. Taraxaci Reb. 154 l'hlei-pratensis £. ei //. 283 Phxagmitis K'^rn. 27"> Piloselloidarum Probst 159 Pimpiuelke ilfort. 188, 192, 194, 195 Poarum X7<7«. 45, 278 Polygoni Pers. 227, 228 Polvgoni-ampkibii Pers. 104, 227, 228 Polygoni-Convolvuli DC. 104, 228 Polygoni-vivipari Karst. 226 Polygonorum Link 227 Porri JFiret. 122, 235 Potfiitillae Pers. 291 Prenanthis Plowr. 151 Primulas Duby 179 Pringsheirniaua Kleb. 242 /Vm>h DC. 208 Pruni-spinosas Pers. 207 Prunorum Link 208 pulverulenta Grer. 198 punctata Link 163, 164 Rhodiolas B. et Br. 210 Ribesii- Cartels Kleb. 243 Ribis DC. 212 Rubigo-vera Plowr. 258, 260 Rubigo-vera var. simplex Korn. 264 Rubi-Idaei DC. 298 Saginae 7i. et S. 221 Sanguisorbae DC. 292 Saniculse Grew. 182 Saxifragae Schl. 212 Saxif rag arum Cooke 160, 212 Schmidtiana Diet. 267 Schneideri Schrot. 172 Srhu.lrriima P. et M. 216 Schroeteri Pass. 232 Scirpi DC. 239 rrodoniae Link 175 secalina < s t . 145 suaveolens var. Cyani Wint. 140 Symphyti-Bromorum Mull. 262 Syngenesiarum Cooke 144 Tanaceti 7>C. 134, 135 Taraxaci Ploirr. 154 Thalictri Ghev. 214 Thesii CAaiZi. 229 tinctorias Magn. 139 tinctoriicola Magii. =P. tinctoriae Magn. 139 Tragopogi 0. 211 vaginalium Link 117 Valantias Pers. 167 variabilis Grew. 152, 154 Veronicas Schrot. 169 Veronicarutn DC. 169 Vincsa /.V/7». 176 violacea Schultz 295 Violas DC. 200 Violarum Link 201 Virgaureae Lib. 130 Winteriana Mo^w 268, 348 Zopfii Wint. 217 Pucciniaceas 74, 75, 85 PUCCINIASTRUM 74, 364 Abieti-Chamaenerii Kleb. 366 Agrimonies Tranz. 364 Aspidiotus Karst. 374 Circaeae Speg. 362, 365 Epilobii Otth 366 ';, Toni 370 Padi Grove 368 Pyrolae Karst. 367 Vacciniorum Karst. 371 Trachyspora Alchemillae Fckl. 106 TRANZSCHELIA cohaesa Arth. 210 pinictittii Arth. 208 TREMELLA clavariaeformis Jacq. 304 junipenna Linn. 307 Sabiuttc Dicks. 308 TRICHOBASIS Ana, Ucae Cooke 192 4jm Wallr. 184 Artemisiae Berk. 134 Betae Cooke 113 caricina Berk. 241 Cirsii Lasch 142 Clinopodii DC. 170 Com// Cooke 196 < tynapti DC. 190 Epilobii B^rk. 198 Fabae Cooke 97 fallens Cooke 90, 91 Galii Lev. 164 Geranii Cooke 103 glumarum Lev. 258 Heraclei Berk. 194 Hieracii Cooke 158 Hydrocotyles Cooke 181 Impatientis Bab. 204 irta/x Cooke 230 Lab in tar it m Lev. 170 Lapsanae Cooke 147 linearis Lev. 250 Lynchii Berk. 385 oblongata Berk. 238 77,7' HOBASIS (cont.) Petroselini Berk. L90, 197 Pimpinellae Cooke L88 Polygonorum Berk. 117. 227 Primulae Cooke 17'.» Pyrolae Berk. 312, 367 Iihniiiiii Cooke 209 Eubigo-vera LeV. 260 Rumicum DC. 114 suaveolens Lev. 1 1") Symphyti Lev. 363 Umbellatarum LeV. 196 Vinnif Berk. 176 Violation Lev. 200 fcrichogyne Ls, i;i, 77 TRIPHKAGMIUM 7:.. 287 Filipendulae Pass, l'^7 Ulmariae H'taf. 287 UBEDINOPSIS 71. 379 filicina Magn. 362, 378, 379 Polypodii Liro 375 Struthiopteridis Diet. 380 UREDO 378, 385 Acetosae Schum. 223 JEgopodii Schum. 185 Agrimoniae Plowr. 364 Alchemillae Pers. 106 Alliorum Cooke 121, 235. 341 ambigua DC. 121 AwjrUrae Schum. 192 annula via Str. 175 Anthyllidis Grev. 95 apiculata var. Trifolii Str. 93 appendiculata var. Phaseoli Pers. 101 appendiculata var. i7.7 Pers. 99 Aspidiotus Peck 374 .Betae Pers. 113 Betae var. Convolvuli Pers. 17o &i/r\ et Pr. 385 Polygonoruru DC. 227 Polypodii Schrot. 374, 375 Polypodii var. Pkegopteris Wint. 379 pojiiilimi var. betulina Pers. 358 porphyrogenita Kze. 369 /'"/•W Sow. 235 Potentillarum DC. = chiefly the caeoma-spores of Phragmidium Fragariastri Potentillarum var. Agrimoniae DC. 3(54 Primulae DC. 179 Pteridum White 378 pustulata Pers. 366 Pyrolae Grev. 367 Quercus Brond. 315 Rhododendri DC. 384 Rumicis Schum. 114 Saxifragarum DC. 357 Scillarum Grev. 120 Scirpi Cast. 124 Scolopendrii Sc/nvit. 377, 378 Sempervivi A. et S. 335 Soldanellae DC. 180 Sonchi Pers. 324 Symphyti DC. 363 Tropseoli Desm. 386 Tussilagini8 Schum. 322 Vim, in, i,- Schum. 287 Vaccinia rum, Link 371 vagans var. Epilobii-tetragoni DC. 1 98 Fmc