ee W. 75,8559 nos,i-oO G. FARLOW HARVARD BOTANI ———— At a meeting of the Botanical Department held Oct. 20, 1903, the following vote was passed: *“ Under the head of Harvard Botanical Memoirs it is proposed to include all quarto publications issuing from the Gray Herbarium, the Cryptogamic Herbarium, and the Botanical Laboratories of Harvard University, including theses presented for the degrees of Ph.D. and S.D. in Botany. Inasmuch as some of the future publications are likely to be continuations of subjects treated in quarto papers already published, it seemed desirable to begin the numbering of the Memoirs with the year 1880, the date of the first quarto publication of any member of the botanical staff at present connected with Harvard University.” At a meeting on Nov. 25, 1916, it was voted to dis- continue the series of Botanical Memoirs. In all, nine numbers have been issued, the titles of which are given below. . I. The Gymnosporangia or Cedar-Apples of the United States. By W. G. Farlow. Anniversary Memoirs, Boston Soc. Nat. Hist. 1880. Pp. 38. Pls. 1 and 2. II. The Entomopthoreae of the United States. By Roland Thaxter. Mem. Boston Soc. Nat. Hist., 1V, No. 6. Pp. 133-201. Pls. 14-21. April, 1888. TII. The Flora of the Kurile Islands. By K. Miyabe. Mem. Boston Soc. Nat. Hist., 1V, No. 7. Pp. 203-275. Pl. 22. Feb. 1890. IV. A NorthAmerican Anthurus: its Structure and Development. By Edward A. Burt. Mem. Boston Soc. Nat. Hist., II, No. 14. Pp. 487-505. Pls. 49 and 50. Oct. 1894. V. Contribution towards a Monograph of the Laboulbeniaceae. By Roland Thaxter. Mem. American Acad. of Arts and Sct. Boston. XII, No. 3. Pp. 189-429. Pls. 1-26. Presented May 8, 1895. Issued Oct. 14, 1896. VI. The Development, Structure, and Affinities of the Genus Equisetum. By Edward C. Jeffrey. Mem. Boston Soc. Nat. Hist., V, No. 5. Pp. 155-190. Pls. 26-30. April, 1899. VII. : Comparative Anatomy and Phyllogeny of the Coni- ferales, Part I. The Genus Sequoia. By Edward C. Jeffrey. Mem. Boston Soc. Nat. Hist., V, No. 10. Pp. 441-459. Pls. 68-71. Nov. 1903. ; VIII. The Comparative Anatomy and Phyllogeny of the Coni- ferales, Part II. The Abietineae. By Edward C. Jeffrey. Mem. Boston Soc. Nat. Hist., VI, No. 1. Pp. 1-37. Pls. 1-7. Jan. 1905. IX. Contributions towards a Monograph of the Laboulbeniaceae, Part II. By Roland Thaxter. Mem. American Acad. of Arts and Sci., XIII, No. 6. Pp. 219-469. Pls. 28-71. June, 1908. CosicAe of thas Prank a hy In he it ae a Pa wn HARVARD BOTANICAL MEMOIRS Under the head of Harvard Botanical Memoirs it is proposed to include all quarto publications issuing from the Gray Herbarium, the Cryptogamic Herbarium, and the Botanical Laboratories of Harvard University, in- cluding theses presented for the degrees of Ph.D. and S.D. in Botany. Inasmuch as some of the future publications are likely to be continuations of subjects treated in quarto papers already published, it seemed desirable to begin the numbering of the Memovzrs with the year 1880, the date of the first quarto publication of any member of the botanical staff at present con- nected with Harvard University. [. The Gymnosporangia or Cedar-Apples of the United States. By W. G. Farlow. Anniversary Memoirs, Boston Soc. Nat. Hist. 1880. Pp. 38. Pls. 1 and 2. The Entomopthoraceae of the United States. By Roland Thaxter. Mem. Boston Soc. Nat. Hist., 1V, No.6. Pp. 133-201. Pls. 14-21. April, 1888. The Flora of the Kurile Islands. By K. Miyabe. Mem. Boston Soc. Nat. Hist., 1V, No.7. Pp. 203-275. Pl. 22. Feb. 1890. A North American Anthurus: its Structure and Development. By Edward A. Burt. Mem. Boston Soc. Nat. Hist., U1, No. 14. Pp. 487-505. Pls. 49 and 50. Oct. 1894. Contributions towards a Monograph of the Laboulbeniaceae. 3y Role Thaxter. Mem. American Acad. of Arts and Sct. Boston. XII, No. 3. Pp. 189-420. Pls. 1-26. Presented May 8, 1895. Issued Oct. 14, 1896. The Development, Structure, and Affinities of the Genus Equisetum. By Edward C. Jeffrey. Mem. Boston Soc. Vat. Hist., V, No. 5. Pp. 155-190. Pls. 26-30. April, MEMOIRS OF . . THE AMERICAN ACADEMY # OF , ARTS AND SCIENCES. me Vou. XII. —No. IIL. | - CAMBRIDGE: JOHN WILSON AND SON. Gnibersity Press. DECEMBER, 1896. as Fai te A Sa HE ONTRIBUTION TOWARDS A MONOGRAPH LABOULBENIACE BY ROLAND THAXTER. - Lge feet, Re an ae ra ty _ PRESENTED May 8, 1895. 4 72 < ; “wet ’ ‘ % "1S > é os Mabel “he 7 a a . 4 ey ‘ ws —— : a: .. oo tiie a yi - Sod ips x af 4, Ye ee a ¢ ‘ “ 2 Care. An , ¥ if Le ; a: - oe: ri : a “e n » 2 . & - _ ~ ve. ra —— gang * bs . “at of: . “ + + a ee TABLE OF CONTENTS. eee rOMae WVCRUM nen yiesteres oS) lk ek the wt tt PAE I, GeNERAL CHARACTERS OF THE LABOULBENIACE® HistoricaL SUMMARY GreNERAL MorpenoLoGy AND DEVELOPMENT Pend hes : a: Met arr mee Ns Tue Spores. Their formation, mode of transference, and germination. Development of the foot; its modifications and function aghe> Aor ORs hs Tue Recerracte. How it is characterized in Pe Its simplest type and some of its ier ey Sea ner ww ee Sew et kt ot ls Tur AppenpAGEs. The general structure, origin, and significance of the sterile and fertile forms . . : , : a rhe F . THe Mate Sexvuan OrGans. Their general character, relation to the appendages and systematic significance. Antheridia producing antherozoids (a) exogenously, and (b) endogenously. The two types of antheridia occurring in the latter category . The simple antheridium. Its development, structure, and the production from it of antherozoids # its variations in form; its disposition on the appendages either definite or indefinite : 6 RE MEE. 5s ae a The compound antheridium. Its structure compared with that of the simple anthe- ridium ; the relative frequency of its occurrence ; its detailed structure as illustrated by the genus Dimeromyces; variations of this type of structure in other genera; its more complicated form in Haplomyces and Cantharomyces Relation in position of the antheridia to the female organ and the probable existence - of cross-fertilization: duration of the period of functional activity in either type . The antherozoids. Their exogenous or endogenous origin and mode of production in either case; their cell characters and rate of discharge aes Ps oe Tue FemaLe Sexuvat OrGaAns. Their origin and detailed development as iiceeatad by Stigmatomyces: they consist primarily of two superposed cells; the primordial cells of the perithecium and of the procarpe, respectively. Development of the procarpe and young perithecium from these primordial cells; their component parts, form, and rela- tions. Fertilization of the trichogyne and the accompanying changes in the young peri- PAGE 193 197 ie 202 202 206 207 209 211 214 190 CONTENTS. THe FEMALE SEXUAL ORGANS continued. thecium and the carpogenic cell; final development of the latter, and production of asci. Structure of the mature perithecium and function of its two cell-layers. Absorption of the ascus and discharge of spores é : Variations in the relation of the perithecium to the receptacle ; general correspondence in the history of its development. Variations in the form of the trichogyne; its fer- tilization, how effected; results of fertilization in different cases; variation in number of ascogenic cells produced; the form and position of the latter. Relation of the asci to the ascogenic cell; their form. Variations in the number of ascospores; their discharge, how effected. Destruction of surrounding cells and of canal cells by spore- and ascus-mass. Variations in the development of the perithecium; the latter further illustrated by the genus Enarthromyces. Variations in the number of wall- cells and their relation to the ascogenic cells; perithecial appendages and out- growths; modifications of the lip-cells. Variations in the development and relations of the parietal and canal-cells; their general function ; assumption of a similar fune- tion by the basal cells of the perithecium; obliteration of the cavity of the wall-cells. Relations of the perithecial cells illustrated by their protoplasmic connections . SUMMARY OF THE DEVELOPMENT OF THE PERITHECIUM AND FEMALE SexuAL ORGANS GENERAL RELATIONS AND CHARACTER OF THE CELLS; the universal envelope surrounding them; character of the cell wall and presence of fibrille in certain cases ; absence of any test for cellulose. Character of the cell contents; protoplasmic continuity; nuclei . . ABNORMAL MORPHOLOGY AND D Evan ORNUHNEU MECN MeN) i). js 6 of» oe er Abnormal septation in the receptacle and branching of the appendages. Accessory peri- thecia, of two kinds. Abnormal increase in number of antheridia; abnormal production of the latter from the basal cells of the perithecium; abnormal filaments similarly pro- duced; substitution of an antheridial appendage for the perithecium. Atrophy of one individual of a spore pair... 2 ©. ies Uc.0 eae ae aan tes tte ey ho) oc an Norman Variations. Such are influenced largely by size and character of host and position of growth 2. en de we en ge = eae Rate oF GRowTH AND DuRATION oF Lire PERIOD GzocraraicaL Distripution +.) 2 teeny ae ae General distribution; distribution by continents. Range of species; local occurrence. . . Hosts OF LABOULBENIACER . oe Gs 0. mu) ee eee teen ag Hosts in general. Table of host-insects. Comparison of representation of the genera among different insect groups. Host habitats. Relation of parasite to host in regard to nutri- tion and to position of growth, the latter often invariably fixed; relation of hosts to type- forms. 6. sw not bet et eR gee PaRASsItEs OF LABOULBENIACE A 20 2. Wie) ya, 06) unde) a eee ne CoLLEcTION OF LABOULBENIACE®. Their cultivation and preparation for examination, or for the herbarium . 206,20 2 a) ik ee ye Sa an PAGE 217 224 233 234 237 237 239 240 241 243 244 244 247 248 CONTENTS. PART I. Inrropuctory Nore concernina tae Systematic Position or THE LABOULBENTACE®, Historical Résumé. Summary of theories regarding the origin of Ascomycetes and the mor- phology of the ascus. Comparison of the Laboulbeniacee with the Floridew ; their genetic Demenmein SUpACRTOO .)e) sucahimens «6 6 he 8 8 wh ew ee ww ww iste (Gy) MERANGEMENT ADOPETEDS (~) =) 5 <6 5 6 6 st ee wt le tl ll GENERAL Synopsis of THE GENERA AND SPECIES. . . - © + + © © © © we we ww eceMOnal UNGSterMIned JOLMIGs a. ‘“‘ Doubtful Ascomycetes ” of his Morphology and Biology of the Fungi, ete. In 1886 occurs the note of Gerke, to which my attention was drawn through the kindness of Professor Giard, in which he gives a figure that, without doubt, is intended to rep- 5 resent the “ Appendicularia entomophila” of Peck, published two years later. In the same year (1886), Karsten published in Hedwigia, under the title “ Doubtful Ascomy- cetes,’ a reassertion of his former observations on the sexuality of Stigmatomyces, in reply to the publications of Peyritsch and De Bary already mentioned, denying their ascomycetous nature and giving what he supposed to be the method by which the spores were formed. In 1889, Berlese again summarized the group, adding to the thirteen species then known a new form (Laboulbena armillaris) found on an acarid from South America. With the exception of the writer’s own notes on the family, but two papers published since the one just mentioned remain to be noticed: that of Giard (1892), in which he describes, as a new genus Thaxteria, a remarkable species of Laboulbenia from the Javan Mormolyce ; and that of Istvanffi (1895), in which he redescribes as Laboul- bema gigantea a large form of L. elongata, giving his impressions of its morphology and development, stating his disbelief in the sexuality of the group, and expressing the erroneous opinion that the individuals are derived from a vegetative portion growing within the body cavity of the insect. GENERAL MorpHoLoGy AND DEVELOPMENT. Spores. The spores of the Laboulbeniacesw present a uniformity of form and structure quite remarkable for so varied a group, being in all cases, without excep- tion, hyaline and fusiform or acicular in shape; and although in the single genus tested MONOGRAPIL OF THE LABOULBENIACEZ. 203 Amorphomyces they are continuous (Plate V, fig. 29), in all others they are divided into two cells by a septum or pseudoseptum. In the great majority of cases the two spore segments are of unequal size; that which is terminal in relation to the axis of growth being, as a rule, much the longest: although in a few cases, as in Zodiomyces and Ceratomyces (Plates XXIV to XXVI), the reverse is true; while in still other instances the septum may be more nearly median, as in Compsomyces (Plate XI, fig. 15). The spore contents usually consists of more or less homogeneous granular pro- toplasm, except in the genus Amorphomyces, the spores of which when living contain numerous, often large, oil globules. In all cases a large spherical nucleus may be, as a rule, readily demonstrated in either segment (Plate I, fig. 13). A gelatinous enve- lope, more or less well developed and characteristically thickened about its base, always surrounds the spore; serving as a protective covering for the latter as well as facilitating its adherence to the host insect when it comes in contact with its surface. In the majority of cases this envelope, though often adherent about the tip of the spore, is continuous around it; but in Ceralomyces furcatus and C. contortus (Plate XXV, figs. 4 and 10) the separation between the spore segments involves the envelope also, which is marked by a corresponding constriction. The spores are produced in the asci in fours or eights (Plate XXIV, fig. 9, Plate XI, fig. 17), in the first instance usually disposed more or less definitely in pairs, one of which is slightly higher than the other; and the members of a given spore pair are discharged together through the pore of the perithecium, the ascus wall having been previously absorbed. The juxtaposition of two individuals at the point of contact with the host, a condition essential for the perpetuation of the dicecious species, is thus insured in a majority of cases (Plate III, fig. 5). The spores are formed in the ascus, and therefore lie in the cavity of the perithecium after the wall of the latter has been absorbed, with the basal half uppermost; and the base is therefore directed towards the substratum on which it is discharged. Having become attached to a proper host, the swollen portion about the base, by its peculiar form, assists the spore in assuming the position necessary for germination; the upper extremity, if at first adherent, soon freeing itself and projecting from the substratum at an acute angle. A conspicuous exception to the conditions just described is found in the genus Mos- ° chomyces, the minute spores of which are irregularly distichous in the eight-spored ascus, and become free in a mass within the distal portion of the perithecium, whence they are discharged, not in pairs but in small masses, each mass probably giving rise to the so-called individual, which may thus in reality be compound (Plate XI, figs. 16 and 18). 204 MONOGRAPH OF THE LABOULBENIACE. oe The transfer of the spores from one host to another is probably accomplished, as a rule, by the direct contact of two insects; as, for example, during coitus, perhaps never otherwise in the aquatic species, as might be inferred from the remarkable con- stancy with which some of these forms occur in definite positions on the elytra or else- where ; but may doubtless be otherwise effected, at least in cases where more or less gregarious hosts inhabit or hide during the day in moderately moist situations. Under these conditions it is not improbable that spores discharged upon materials with which such insects have come in contact may subsequently adhere to other individuals on which they may develop. Although a gelatinous envelope is always a protection of extraordinary efficacy, it does not seem probable that the spores can retain their power of germination for any considerable time, at least in a dry condition. Having reached a proper host, and having adhered to it by virtue of its generally viscous character, the sporé begins to germinate at once. Germination. The first indication of germination in the spore usually consists in the modification of its lower extremity into a blackened organ of attachment, the foot ; the blackening resulting from a change which takes place in the gelatinous envelope in this region by which it becomes converted into a black, opaque, hardened, more or less elastic medium of attachment to the host. This conversion of the lower portion of the basal spore segment into an indurated organ by which the growing plant ad- heres firmly to the substratum on which it grows, is apparently unconnected with any effect resulting from contact with the chitin of the insect; since, in exceptional in- stances, where the usual discharge of spores has been prevented from any cause, the latter, while still within the perithecium, may begin to germinate and even attain an advanced development (Plate V, figs. 1 and 19). In such cases the first step in the process consists, as in normal germination, in the formation of a blackened foot of the usual type. A foot of this nature is not, however, invariably present. If the figures of Peyritsch are to be relied upon, there is no such blackening in the case of Helmin- thophana (Plate VIII, fig. 10), which is represented as penetrating the integument of the insect on which it grows by the intrusion of a papillate haustorium, there being no blackening whatever of the basal cell. The typical foot is also conspicu- ously absent in certain other genera. In Moschomyces also, to which reference has been made above, this organ is not differentiated ; and the plant penetrates the soft integument by means of a cellular haustorium which, expanding within the body cav- ity of the host, holds the parasite firmly attached. The most striking exception, however, is presented by the genus Rhizomyces, in which the penetrating haustorium reaches a development quite beyond that of any other form. In this case (Plate MONOGRAPH OF THE LABOULBENIACE. 205 IV, figs. ] and 3) the basal cell sends into the body cavity of the host a copiously branched and well-developed rhizoid-like organ, the interior of which is apparently continuous with that of the basal cell. In cases where a typical foot is formed, it may constitute a cell distinct from the basal cell of the individual, as is at least often the case in Stigmatomyces Beri (Plate I, fig. 14), a condition which may very likely exist in many other instances, although obscured by the blackening already mentioned. In some instances this blackening involves more than the basal cell of the mature plant, as in some species of Cerato- myces (Plate XXV, figs. 15 to 17). In forms like those just mentioned in which a definite haustorium is present, it is quite evident that this organ constitutes the means by which the parasite absorbs from its host the nourishment necessary for its growth, and at the same time serves to fasten it securely. In the great majority of instances, however, it is probable that there is no intrusion of any kind from the ordinary form of blackened foot into the substance of the host. That this is the case may be clearly seen in preparations in which the parasite is shown attached to some transparent portion of the host’s integu- ment, as in fig. 4, Plate III. In the specimen here represented several individuals were fixed to the surface of the thin integument of one of the abdominal segments of the host, the substance of which was perfectly transparent, so that by inverting the piece to which they were attached, so as to view its lower surface, one could observe with accuracy the relation of the adherent portion of the foot to the chitinous surface to which it is applied. In such a specimen it is evident that this sucker-like adherent portion consists of a flat area in the closest contact with the chitin of the integument, and consisting of a thin membrane through which the absorption of nutriment takes place, bordered, as in the figure cited above, by the thick base of the indurated exter- nal wall of the foot proper. It should be mentioned, however, that Peyritsch figures sections of a fly’s integument which indicate a slight penetration through the pore canals in the case of Stigmatomyces. In addition to its function of attachment and absorption the foot may, in some of the aquatic genera more particularly, perform the office of a fulerum on which the plant is supported and by means of which it is allowed a certain freedom of motion which it could not otherwise attain. This is effected through the more or less sudden and distinct enlargement of the foot above its point of attachment. The rounded ful- crum thus produced rests on the surface of the host, allowing the body of the plant, although appressed, to lie free from the insect and to roll upon it from one side to the other, as far as the elasticity of the attached portion will permit. ‘This is most clearly 206 MONOGRAPH OF THE LABOULBENIACE. seen in Chitonomyces (Plate XXVI) or Hydroeomyces, both of which genera inhabit hosts that live, for the most part, submerged, and which are rapid swimmers. Under these conditions of life the advantages of such a contrivance, to allow a certain free- dom of motion, are sufficiently apparent. The same office of a fulerum for the sup- port of the perithecium is effected in some species of Ceratomyces by the conversion of a considerable portion of the receptacle into a foot-like organ (Plate XXIV, fig. 1). After having become attached to the insect, and during, or sometimes before, the formation of a definite foot, the spore elongates more or less distinctly and becomes further divided by the formation of tranverse septa into a series of superposed cells, varying in number in the different species and genera, from the further development -of which result the three fundamental parts of which these plants are usually com- posed: namely, a main body, the receptacle ; one or more spore-producing portions, the perithecia ; and lastly, one or more appendages which, in the majority of cases, are asso- ciated with the formation of the male sexual organs. The Receptacle. The tern “receptacle” has been used to designate that portion of the fungus on which the appendages, together with the perithecia or their stalk-cells, are inserted ; but it is necessarily used with some looseness, and is sometimes unavoid- ably applied to series of cells which are neither homologous in origin nor similarly related to the other essential organs of the plant. In the genus Laboulbenia, for example, the whole body of the individual, exclusive of the appendages and perithe- cium, is spoken of as the receptacle ; although, in this instance, it consists fundamen- tally of the usual two superposed basal cells, while distally it is formed from a consolidation of the stalk-cell of the perithecium (cell VI), which has become laterally united with what is in reality the base of an appendage (cells III-V). In other cases it is often difficult to determine exactly how the receptacle should be limited, as, for example, in the genus Chetomyces (Plate XI, fig. 20), in which it consists of a single series of superposed cells which give rise directly to perithecia or to appendages. The simplest type of receptacle, which is found in more than half of the genera, consists of only two superposed cells, the upper of which bears the appendage, at first terminally; while the perithecium, or perithecia, if there are several, are lateral pro- ductions from the same cell. This type is well illustrated by such genera as Haplo- myces and its allies (Plate VIL) Compsomyces (Plate XI, fig. 7) and similar instances ; while, as has been above indicated, even genera like Laboulbenia are fundamentally similar. In other genera various degrees of complication are found in the develop- ment of the receptacle which passes gradually from the simple two-celled type to MONOGRAPH OF THE LABOULBENIACEZ. 207 such highly developed forms as are present in Zodiomyces (Plate XXIII) in which it attains its maximum development as compared with other known genera of the group. Other multicellular forms may be illustrated by such genera as Cheetomyces or Enar- thromyces (Plate XI, fig. 20 and IV, fig. 8), in which it consists of a single row of super- posed cells, and Rhachomyces (Plate XII), in which, from a similar simple axis a series of appendiculate cells is cut off on one side, the predominance of transverse divisions resulting in an elongate form. As an illustration of the reverse condition where a predominance of longitudinal divisions is present, genera like Dichomyces or Diplomyces may be mentioned, in which the habit is stout and compact, this type reaching its greatest complication in the genus Zodiomyces just referred to. Although the number and arrangement of the cells which form the receptacle in a given genus is often very constant, this is by no means invariably the case ; and, even in the same species, fixity in this respect does not always exist. In Peyritschiella, for example (Plate VI), although the individuals of a given species do not vary greatly, in so far as the number and arrangement of the cells is concerned, no two species are alike in this respect. The same is true to a more marked degree in Rhachomyces (Plate XII), a genus which is further remarkable from the fact that the main body of the receptacle is in the nature of a lateral proliferation from the sub-basal cell of what may be termed the primary receptacle, as well as from the fact that this pro- liferation may be once or several times repeated, as will be presently noted. Again, in Ceratomyces there may be wide variations in this respect, not only between different species, but in individuals of the same species ; while in other instances in this genus the number and arrangement of the cells of a given species may be invariable. In the simple as well as in the more complicated forms, the receptacle is more or less flattened, usually in an antero-posterior plane, the side bearing the appendage in the mature individual being considered for convenience “ posterior,’ where this distine- tion is possible. In some instances, however, the flattening is in a plane at right angles to that just mentioned, as in Dichomyces and Diplomyces. In a majority of ‘instances this flattening is well marked; but it is most pronounced in forms which have a distinctly appressed habit of growth in relation to their substratum. Appendages. With but a single exception in the whole group of Laboulbeniacex, the receptacle gives rise to one or more appendages which, though not invariably, are, as a rule, clearly distinguished from it, as well as from the perithecium. These appen- dages, though extremely variable, and affording, in many cases, excellent specific dis- tinctions, are chiefly important from the fact that they are, with few exceptions, asso- ciated with the production of the antheridia or male sexual organs of the plant. In 208 MONOGRAPH OF THE LABOULBENIACE. ul cases the primary appendage is originally a terminal structure, developed at least in part, often entirely, from the terminal spore segment, which is as a rule distinctly smaller than the basal, and never under any circumstances concerned in producing any portion of the true receptacle. While this terminal character of the primary appen- dage is often obscured, as in Zodiomyces, Rhachomyces and other genera, it is in many cases sufficiently evident; for example in such instances as Stigmatomyces (Plate I, figs. 2-13), Cantharomyces (Plate VII), and the like. In the genus Rhachomyces the primary appendage is represented by the single small bristle at the very base of the whole series of appendages, which are thus nearly all secondary and quite different in origin from the originally terminal primary one. Again, in Zodiomyces the origi- nally terminal primary appendage is soon sloughed off, its place being taken by a multitude of secondary ones, also quite different in origin. Although they are very important from a systematic point of view, it would be quite superfluous in the present connection to enter into any detailed description of the many variations of form and structure which the appendages, both primary and secondary, exhibit ; and further details should be sought in the special descriptions of the genera. It is sufficient for the present purpose to say that the appendages, using the term to include both primary and secondary, may be solitary or very numerous: greatly elongated or consisting merely of single short cells; simple or branched in a great variety of ways, either sympodially or monopodially, or more or less irregularly ; hyaline or deeply colored ; stiff and bristle-like or flexuous and slender; in short, as may be seen by a glance at the accompanying plates, showing wide variations as to minor details even in nearly related forms. The sterile appendages, or the sterile portions if such are present of fertile appen- dages, which are always more or less filamentous and composed of successive cells placed end to end, may become very highly developed, forming a dense tuft which is often much more conspicuous than the main body of the plant itself. The function of such highly developed sterile branches is doubtless primarily that of protection for the delicate trichogyne, which is subsequently developed, and they may, perhaps, have a further office in facilitating the fertilization of this organ by holding around it a drop of water, which is usually found at times condensed on the surface of the hosts, the majority of which are apt to hide by day, at least, in cool moist situations. It might be supposed that since these sterile portions of the appendages extend, in many cases, beyond or around the tip of the perithecium, that they were of some assistance in the dissemination of the spores; but it is a curious fact that the latter, despite their generally viscous nature, are only in exceptional cases found adherent to any portion ~ o MONOGRAPH OF THE LABOULBENIACEZ, 209 of them; and it seems certain that they do not thus act as brushes for the more con- venient transfer of these bodies. In general the ultimate filamentous sterile branches, or branchlets, consist of a single series of superposed cells which appears to elongate, at least in so far as I have been able to determine, through the successive division of the terminal cell. One curious structure, the significance of which is as yet undetermined, should be mentioned before leaving the sterile portions of the appendages. ‘This structure con- sists of a spine-like process laterally developed from the primary appendage, and has been observed in only a few instances. It is most pronounced in an apparently undescribed genus parasitic on Zachinus pallipes, which has not yet been found in a mature condition. It also oceurs in Sphaleromyces Lathrobii (Plate XI, fig. 19) and in Phadinomyces cristatus (Plate 1X, fig. 22), in both of which it seems to disappear at an early period. The process which is persistent at the summit of the antheridium in Haplomyces may also be of a similar nature. Male Sexual Organs. As has been previously mentioned, the essential function of the appendage, apart from the secondary function of protection which it undoubtedly subserves in many cases, is as a rule connected with the production of the male sex- ual organs; although in the following genera, Amorphomyces, Dimorphomyces, Dimeromyces, Enarthromyces, Peyritschiella, and Dichomyces (probably also in Chitonomyces and Hydrzomyces), the latter are wholly independent structures un- connected with the sterile appendages which usually accompany them. A comparison of the character of these male organs in the different genera makes it apparent that they afford the best, indeed the only basis for the natural separation and grouping of the members of the family as a whole, which are thus dis- tinguished into two main categories: one including those forms having male organs from which the male elements are produced exogenously; the other including those forms in which the male element arises endogenously. Of these groups the first may be further subdivided, on a similar basis, into forms in which the male organs are borne on specialized male individuals, and those in which the sexes occur together on the same individual. Again, as will be seen presently, these moncecious and dice- cious groups are further distinguished according as the male organs are simple or compound; while their relative position, distribution, ete., afford opportunities for minor subdivisions. Since they are undoubtedly homologous with similar structures in the Floridex, the male organs and male elements in the Laboulbeniacez may be properly designated as antheridia and antherozoids, respectively ; the former, as will be seen, consisting of a 14 210 MONOGRAPH OF TIE LABOULBENIACE. single “ antheridial cell,” or a group of such cells, the latter of a single naked or thin- walled cell. If we separate the genera of Laboulbeniacex, on the basis above outlined, into forms with endogenous and those with exogenous antherozoids, it will be found that it is only in two aquatic genera that the forms included in the first-mentioned cate- gory occur: Zodiomyces and Ceratomyces being the only instances in which this type has been definitely observed. In general the antheridial branches are not highly differentiated even in the more typical instances; while were the demonstration of the existence of male elements of this type dependent on the data afforded by the less well-marked examples, one might be inclined to doubt the presence of any sexuality in such cases, despite the presence of a well-developed trichogyne. In Zodiomyces, however, the antherozoids are produced in the form of buds which arise from the tips of short special branches (Plate XXIII, figs. 21-23) which assume a rod-like form and eventually fall from their attachment. These bodies have a definite wall and seem to be sought by the tip of the trichogyne in a fashion to which reference will be made below. A second well-marked instance is found in Ceraloniyces rostratus, from the an- theridial branches of which are developed rods of definite form and size, which become separated and adhere to the trichogyne. In this instance the rods are formed successively from a definite point at the distal end of the fertile cells of the anthe- ridial branch (Plate XXIV, figs. 23-24), each rod usually becoming detached from its point of origin before its successor has begun to form. In other species of the genus, however, these bodies are not so clearly differentiated, and seem to be replaced by slender, often long, filaments which eventually break up into rods that are presum- ably functional as antherozoids. The adherence of the antherozoids to the mother- cell, or to one another, as in the last two instances mentioned, may be assumed to render fertilization more certain ; since it is evident that were such bodies separated, as soon as they were mature, from a plant growing, as in the present instance, on an isolated and rapidly swimming host, the chances of their ever coming in contact with and adhering to the trichogyne would be reduced to a minimum. In view of the fact that this separation does not seem to take place at once, and that the long slender trichogyne must, from its position, inevitably be continually brought into contact with them while still i sw, through the motion of the plant which would necessarily result from the activities of its host, it seems quite probable that such antherozoids, in order to be functional, must become detached at the moment when they come in con- tact with and adhere to the trichogyne. In all other genera of the family the antheridia are more complicated in MONOGRAPH OF THE LABOULBENIACEZ. 211 structure, consisting of highly specialized cells, or groups of cells, within which the antherozoids are formed endogenously and from which they are discharged through a special orifice in the form of free, naked, or nearly naked protoplasmic masses, Among such antheridia two distinct classes may be clearly distinguished, which I have termed simple and compound, respectively : although instances occur in which antheridia ” of the “simple” type are so closely associated that they may be regarded as tran- sitional forms between the two types. In the first class, the simple form is character- ized by the fact that the antheridial cell is quite independent of any similar cells, however closely they may be united to it, and discharges its antherozoids into the surrounding medium through its proper mouth. In the second instance, on the other hand, several such cells, closely associated to form a specialized organ, discharge their antherozoids into a common cavity from which the latter make their escape into the surrounding medium through a single aperture. The simple antheridium (Plate I, fig. 15) is usually a more or less flask-shaped cell, which may be solitary or associated with similar cells grouped together with or with- out regularity.. Although in Amorphomyces, a genus wholly destitute of appendages, it results from the direct modification of the terminal portion of the germinating spore (Plate V, figs. 20 and 23), it originates in other genera as a terminal or lateral outgrowth from the appendage or its branches. In a majority of genera, if it does not terminate the appendage or one of its branches or branchlets, the antheridium is itself a branchlet, as, for example, when it is sessile ; but in a few instances this termi- nal character is lost and the organ is formed as a definitely intercalary cell, as in Rhadinomyces. In the compound type, these cells seem always to be intercalary in origin, although material has not been available for the study of their development. The form of the simple antheridium is remarkably constant, the single cell of which it is composed being distinguished, more or less abruptly, into a basal, somewhat in- flated portion or venter ; and a terminal, more slender, usually sub-cylindrical portion, the neck, originally developed as a terminal outgrowth, which, at maturity, becomes perforate at its apex for the discharge of the male elements. The cavity of the ven- ter is separated from that of the neck by a kind of diaphragm, abruptly distinguished on the venter side, and much less so on the neck side. This diaphragm, which seems to be formed by the deposition of a ring of cellulose in: the position indicated, is per- forate in the middle so as to allow the contents of the venter to pass out into the neck through an opening which is much smaller than the diameter of the cavity of the neck itself (Plate I, fig. 26). The cavity of the neck therefore, which may be conveniently called the canal, though of about the same diameter throughout the 7 Ws MONOGRAPH OF THE LABOULBENIACE. greater portion of its length, is more or less abruptly narrowed just before passing into the venter. The contents of the venter during its active period, as it increases in volume, pushes through the narrow opening in the diaphragm into the cavity of the neck, and the portions thus extruded, when they have reached a certain definite size, become separated from the mass whence they were derived ; and, assuming the form of short cylindrical rods, the antherozoids pass into the general cavity of the neck, the diameter of which is but slightly larger than their own, and thence make their final exit through the terminal pore. This process of abjunction, by which small but uniform pieces become separated from the contents of the venter as it is pushed into the neck, continues for a period which varies somewhat in different cases, but may begin some time before the female organs are mature, and continue long after they have been fertilized. Although one sees many cases in which the neck contains a continuous series of antherozoids which are evidently pushing one another out through the terminal opening, these bodies seem to be able to make their exit quite inde- pendently of one another, although at the same time they also appear to lack any indication of a power of independent motion, amoeboid or other. As has been already mentioned, the form of the simple antheridium is subject to inconsiderable variations which depend in general on the relative development of the neck or of the venter, the latter being sometimes short and stout and abruptly dis- tinguished, while in other cases it may run gradually into the neck without any such clearly marked differentiation. The neck, too, may be short and rather stout, or long and slender, straight or curved, the extremes in these respects being illustrated by such instances as the following: Laboulbenia decipiens (Plate XX, fig. 20), LZ. Llon- gata (Plate II, fig. 15), Teratomyces (Plate X, figs. 6 and 11), Compsomyces (Plate XI, fig. 14), or Stigmatomyces Bueri (Plate I, fig. 26). The disposition of the antheridia, and their relation to one another on the same appendage or branch, is a matter of much importance in affording generic distinctions, and even, in a few cases, is of service in defining species. On this basis all the gen- era having simple antheridia, with the exception of Amorphomyces, which has in- variably a single antheridium, might be separated into two categories: those in which the antheridial cells are disposed in definite series on the appendage, and those in which they are more or less indefinitely placed. In the first category are found forms in which the series is a single one, as in Stigmatomyces (Plates I, figs. 8-12 ; VIII, fig. 3), the antheridial cells succeeding one another in a single vertical row, while a somewhat more complicated condition exists in Idiomyces, where three vertical rows are present (Plate IX, figs. 18-19). Again, MONOGRAPH OF THE LABOULBENIACE. 213 in Helminthophana there are four rows symmetrically arranged (Plate VIII, fig. 10) on the single appendage. In all these cases the antheridial cells are themselves branch- lets from the axis of the appendage. In Rhadinomyces and Corethromyces, how- ever, the entire axis of the antheridial branch is formed from the superposed venters of the antheridial cells, the necks alone being free and projecting in a vertical row (Plate IX, figs. 5, 9, and 14). In Teratomyces a somewhat similar condition is brought about through the re- peated sympodial branching of the appendage, the false branchlets (Plate X, figs. 6 and 11) being either antheridia or short sterile outgrowths, terminated by a charac- teristic beak-like cell. Although, in this last instance, a regular series of antheridial cells, either alone or mingled with sterile branchlets, is produced, the type is not strictly comparable with that of the previously mentioned instances; in which the an- theridial cells or branchlets originate as a result of the septation of an axis already formed and the production of monopodial branchlets from the resultant cells. In Teratomyces it is evident that the serial arrangement is necessitated by the relative position of the crowded appendages. In the second category may be included all the remaining genera, comprising forms in all of which the simple antheridia are not thus disposed in regular series, but are more or less irregularly placed on the appendages. Although never serially arranged through monopodial branching, individual examples occur, however, in which the grouping of the antheridia is very characteristic and specifically distinctive. Among the instances included in this general category, Laboulbenia is by far the best illustration, since it comprises a greater variety of forms than any other genus. Among those in which there is no definite relation in position between the antheridia, Laboulbenia elongata or L. Plerostichi may be mentioned, while among forms in which there is a more definite grouping, Laboulbenia variabilis (Plate XXI, fig. 3) or L. pro- liferans (Plate XVI, fig. 23) offer the best examples of more or less regular and dis- tinctly characteristic clusters. Species in which the antheridia are borne in pairs, or are irregularly grouped, are common; while in one instance, the curious L. zanzibarina (Plate XVII, fig. 5), a single short series may result, as in Teratomyces, from continued sympodial branching. In regard to the number of simple antheridia that may occur in individual cases, it may be mentioned that there are often wide differences, not only between different species, but between individuals of the same species. Laboulbenia texana, for instance (Plate XX, figs. 16-18), has one, rarely two, antheridia, while Z. elongata may have from five to fifty or more. Wherever fertilization has failed, as not infrequently hap- 214 MONOGRAPH OF THE LABOULBENIACE®. pens, the production of antheridia is greatly increased (Plate II, fig. 8). Further ref- erence to this circumstance will be made in connection with the “abnormal forms” described below. Turning to the compound type of antheridium we find much less uniformity than in the simple type just described, although the antheridial cells which make up the essen- tial portion of this organ are practically identical with the simple antheridium. As has been already mentioned, these cells seem to be intercalary in origin as far as can be determined in the absence of a knowledge of their early development, and instead of being wholly or partly free, both the necks and venters are closely united below, beside or around a common cavity into which they discharge. Of the twenty-seven genera thus far described, about one-third are characterized by possessing this compound type of antheridium, and since that occurring in Dimeromyces is one of the largest and most easily studied, it may be taken as an illustration. In this genus the species are dioecious, and the male individual reaches a maximum differen- tiation, being as well developed as the female, and bearing more than one anthe- ridium (Plate IV, fiz. 16), which possesses a stalk-cell and four basal cells lying below the antheridial celis. The antheridial cells in this case are six in number, symmetri- cally arranged in two rows, their venters large and emptying through long narrow canals into the somewhat inflated base of the long and slender “ secondary” neck which serves as acommon medium for the final discharge of the products of all the antheridial cells. The latter here correspond closely to those previously described in connection with the simple antheridia, and the formation from them of antherozoids is also similar in all respects. The canal, however, does not enlarge, as in the simple form, immediately after leaving the venter; but continues about the same diameter till it has nearly reached the general cavity at the base of the secondary neck, when it expands slightly. As a result, the antherozoids remain in connection with the proto- plasm of the venter till they project some Tittle distance into the cavity of the sec- ondary neck, eventually separating from it and falling free into this general receptacle, whence, as represented in the figure, they may be seen at various points making their way out. The antheridium of Dimorphomyces is essentially identical with that just described, the male individual in this genus, however, producing but a single antheridium (Plate V, figs. 8-9 and 14-15). In Peyritschiella (Plate II, fig. 12), Dichomyces ‘and Enarthromyces (Plate III, fig. 19), although the antheridium is somewhat different in form, the secondary neck being less prominent and less abruptly distinguished, its general structure is also essentially the same; the four to six antheridial cells in the last mentioned genus lying somewhat obliquely side by MONOGRAPH OF THE LABOULBENIACE. 215 side in two rows, below a common cavity into which they empty. In these instances the antheridium has no stalk-cell, and is closely united to the receptacle. In Campto- myces and Eucantharomyces it terminates the appendage and is somewhat different in character. In Camptomyces the antheridial cells are placed around and beside a general cavity, and are arranged in several nearly vertical rows, emptying upwards through short necks into this cavity, whence they escape through the terminal pore of a short secondary neck (Plate VI, figs. 5 and 6). In Eucantharomyces the antheridial cells are also arranged in nearly vertical rows, and are more numerous than in the last mentioned genus. They empty into a general cavity, which is cen- tral and terminal, and are discharged through a well-developed, though somewhat irregular, secondary neck (Plate VII, fig. 27). In both these genera the material has been so limited in amount, and the antheridia are so difficult to observe by focussing through the asymmetrical arrangement of their numerous antheridial cells, that I have been unable to obtain a figure that would show satisfactorily the exact details of structure and arrangement presented by the latter. The gross structure is, however, made out without difficulty, and the general cavity is usually filled with very numerous antherozoids, which here and there may be seen in process of formation from the antheridial cells in a fashion exactly resembling that which occurs in the instances previously described. A single type of compound antheridium remains to be mentioned, which occurs in Haplomyces and in Cantharomyces, and in this instance also, owing to the lack of sufficient material and to the complicated structure of the organ, much remains to be learned concerning its exact structure. The type is distinguished from those already mentioned from the fact that the secondary neck opens, as far as can be ascer- tained from the material available, through a lateral pore, and consists of a central cavity almost completely surrounded by the very numerous antheridial cells which open into it. This general structure may be made out by focussing through the organ; but no further details have been visible in the specimens examined. The an- theridium (Plate VII, figs. 8, 9, and 22) is identical in the two genera mentioned, except that in Haplomyces it is terminated by a thorn-like cell, while in Cantharo- myces it is placed below a well-developed sterile branching portion. With the exception of Cantharomyces pusillus, which may possibly prove to represent a new generic type, none of the species of these genera have been seen in a fresh condition, and they are the only ones possessing compound antheridia in which the discharge of the antherozoids has not been observed. The antheridium in these forms is the most highly developed that has thus far been noticed, and further observations upon it are greatly to be desired. 216 MONOGRAPH OF THE LABOULBENIACE. The antheridia are usually so placed that the antherozoids are discharged very near to, or even directly upon, the female organ when the latter is mature (Plate I, fiz. 15; Plate Il, fig. 2; Plate III, fig. 18). When they are associated with long and well-developed appendages, they are usually borne near the base of the latter (Plate Il, fig. 5, anth.), and where also, as in the case of Laboulbenia, there are inner and outer appendages or branches, the male organs usually occur on the inner ones, that is, on those nearest to the female. There are, however, some exceptions to this rule in which the antheridia and trichogynes are not thus closely associated ; as, for example, in Teratomyces, in which the former are borne some distance below the latter, and are curved away from them. In many cases also, though the two sexes may be closely associated, there is often a marked tendency in the male to turn away from rather than towards the female, as in Stigmatomyces and Eucantharomyces. In the dioecious genera, the male and female individuals are always in close proximity, their invariable association resulting from the fact that the spores always become attached to the host in pairs, corresponding to those which are formed in the ascus, and that, of any given spore pair, one member produces a male while the other produces a female (Plate V, figs. 2, 5, 17, 20, and 23). Notwithstanding the fact that the male and female organs are in general so closely associated, it 1s more than probable that cross-fertilization occurs quite as frequently, if not more frequently, than close fertili- zation; since not only are the species as a rule more or less gregarious in habit, but the maturition of the antherozoids invariably precedes that of the trichogyne, and the former continue to be produced long after the latter has been fertilized, in many cases after the perithecium has matured and begun to discharge its spores. This is true of forms having but a single perithecium ; but more strikingly so in those which produce several successive perithecia. In Dimorphomyces, for example, the antheri- dium of the male individual continues to produce antherozoids indefinitely, while the female may produce two or even four sets of perithecia. The same extension of the functional period of the male is also seen in all the genera having compound anthe- ridia. In forms having simple antheridia the same extension of functional activity in the male is often effected by the production of new antheridial cells or new fertile branches after the fertilization of the trichogyne. As has been previously mentioned, the antherozoids are formed, in those genera which produce them exogenously, as lateral branchlets, the whole or portions of which become separated in the form of long slender rods (Plate XXIV, figs. 21, 24; Plate XXIII, figs. 21-25), having a definite cell wall, while in the genera producing them en- dogenously the contents of the venter of the antheridial cell are protruded through MONOGRAPH OF THE LABOULBENIACEZ. 217 the opening in the diaphragm into the cavity of the neck, and this protruded portion, having reached a definite size, becomes separated as a free mass of protoplasm, which, having been, as a rule, moulded to a cylindrical form in the cavity of the neck, makes its exit through the terminal pore. When free, the endogenous antherozoids exactly re- semble bacilli or cocci in appearance, becoming rounded at the ends so that the short forms are almost spherical (Plate I, fig. 26; Plate VI, fig. 5; Plate II, fig. 2; Plate III, fig. 19). Their substance is refractive and homogeneous, and I have been unable, by staining, to differentiate a definite nucleus. Although, when first separated from the contents of the venter, they are undoubtedly naked protoplasmic masses, in some cases at least a thin wall seems to be secreted around them after their exit from the antheridial cell. In Enarthromyces, for example, which possesses antherozoids larger than those of any other known form, such a wall appears to be present, and the same may very probably be the case in other forms in which the antherozoids are so minute that exact observations in this respect are made with difficulty, even after they have become attached to the trichogyne. When freshly discharged they are usually suffi- ciently characteristic in appearance to be readily distinguished from the bacteria and yeasts which are often so numerous on certain individuals as to completely corticate the appendages. The discharge of the antherozoids from the antheridial cell is a slow process, and probably does not occur more frequently than once every two or three hours; at least this is about the rate observed in individuals which have been separated from the host and placed in water. It may be, however, that the rate is more rapid under natural conditions; since the parasite does not survive very long after separation from the host, which doubtless disturbs not only its nutrition, but also its general turges- cence. From the secondary neck of compound antheridia the discharge is, of course, often much more rapid at times, and a dozen or more antherozoids may be seen to make their exit within a few minutes after an individual of Camptomyces, for ex- ample, has been mounted in water. Cultures of the antherozoids in water, continued for many days, have never shown any indication of an attempt at development. Reference has already been made to the duration of the active period of the anthe- ridia, and it remains to note the fact that the numbers of antherozoids formed during this period from a single antheridium must be counted by hundreds, or even thou- sands, in the case of the more highly-developed compound forms. Female Sexual Organs. — It has been previously mentioned that in a majority of forms the antheridial appendage is developed from the terminal cell of the germina- ting spore. The female organs, however, are always formed from the products of 218 MONOGRAPH OF THE LABOULBENIACEZ. division of the basal cell, never in any case from the terminal cell, where this is pres- ent. Although the products of the division of the terminal cell are invariably sterile or male, it is not true, as might be supposed, that the basal cell or its derivatives have any inherent female character; since, in many cases, both normal and abnormal an- theridia and antheridial branches may arise below the point of imsertion of the female organ, or even, in exceptional cases, replace it entirely (Plate II, figs. 7-8). While, then, the primary appendage is, as a rule, terminal, the trichogyne, as well as the perithecium which follows it, are always lateral, with the single exception of Amor- phomyces, where both are developed terminally from the unsegmented spore. This lateral origin is, however, very often obscured in the fully developed plant, from the fact that the perithecium, as it develops, is apt to push the appendage more or less to one side, and assume an apparently terminal position; as, for example, in the genus Stigmatomyces or in Laboulbenia. We have already seen that the basal cell of the germinating spore begins its development by the formation of a cross partition which divides it into an upper and a lower cell, and that the base of the latter becomes modified to form the foot. The development of the upper, although varying considerably in the different genera, may, perhaps, be best illustrated by reference to the series of figures (Plate I, figs. 1-24) of Stigmatomyces, which is typical of the more simple forms. In this series, fig. 2 represents the spore after it has become attached to the host, the foot is beginning to form and the upper half is considerably enlarged. Figs. 3—5 illustrate the further development of the upper cell, the lower still remaining as at first. In fig. 6, the lower cell has become divided into two superposed cells, the upper of which (4) forms the basal cell of the appendage. The lower cell (v) then divides in two by a trans- verse partition, forming the basal cell and the sub-basal cell (a). This sequence of divisions is not apparently constant, and in other forms, at least, it more frequently happens that the partition which separates (a) from the basal cell is formed before that which separates (a) from (2). In fig. 7, the nucleus of cell (@) has already divided, although no wall has formed between the daughter nuclei. In fig. 8, this wall has been formed and the cell (a) has become divided into an upper and a lower cell (a and a”), cell (6) remaining permanently without further division as the base of the appendage. Of these two cells (a and a"), the lower remains without further change as the terminal cell of the receptacle; while the upper (a) alone continues to develop into the female sexual organ and the perithecium. It will be noticed that even at as early a stage as is represented in fig. 8, the antheridial cells have begun to mature and to discharge their antherozoids. Cell (a’) next begins to grow upward and MONOGRAPH OF THE LABOULBENIACE2. 219 outward (fig. 9), and by the formation of a septum is soon divided (fig. 10) into an upper and a lower cell (d and c). From the upper of these cells, which may be called the primordial cell of the procarpe, is formed the whole of the female sexual organ, while from the lower (c), which may be similarly called the primordial cell of the perithecium, are developed the cells which form the perithecium proper. This lower cell (c) is the first to show further signs of development, and becomes divided by a more or less obliquely longitudinal septum (fig. 11) into two cells (¢’ and ¢). The cell (c’) then begins to grow upward, and becomes divided into an upper and a lower portion, fig. 12 (z) and (p). The lower cell (fig. 12, p) constitutes the “stalk- cell,’ while the upper continues to divide, as will be presently described. The cell (c’) of fig. 11, on the other hand, becomes separated into two upper cells (fig. 12, 7,7), lying on opposite sides so that only one is shown in the figure, and a lower cell (4), which may be called the secondary stalk-cell. The cell (z) and the cells (2, 2) then continue to grow up around the base of the cell (d), the primordial cell of the procarpe. One of the cells (7, 7) then becomes separated into a single lower and two upper cells, while the other becomes separated into a single lower (figs. 13, 0) and a single upper cell (n), as also does the cell (z); the cells (o’) and (x) of fig. 13, resulting from its divisions. There are thus formed three proper basal cells of the perithecium, two an- terior (0, 0) and one posterior (0'), but two of which are shown in the optical section (fig. 15), and from them, four cells are separated above which continue to grow up- ward and surround the cell (d), the base of which is now (fig. 15) completely enclosed and has become separated by a cross partition as the cell (f/) from the part (e) which still remains free above it. We then have this basal part (fig. 15) constituting a central cell (f), distinguished from the free part above (e) by a cross partition, and completely surrounded by the seven cells (0, o’ and n, 2), but four of which are, of course, shown in the optical sec- tion. The central cell (f) then remains without further development until fertiliza- tion has taken place; the terminal portion of cell (e) in the mean time becomes separated from the part below (fig. 14, e”) as a usually very small cell (e’), which im- mediately begins to produce a terminal outgrowth, the young trichogyne. This small cell is more marked in the genus illustrated than in most others in which it is not, as a rule, so clearly distinguished; the filamentous portion usually growing directly from the apex of cell (e), from which it is separated by a septum (Plate II, fig. 2). Cell (¢) and the projection (é) from it must therefore be considered as con- stituting together the trichogyne proper. As the latter develops, the cell (¢’) becomes less well marked, while the filamentous portion reaches a development that varies in 220 MONOGRAPH OF THE LABOULBENIACE. different specimens, but is usually not greater than is represented in fig. 15 (ér) and in fig. 16. Before fertilization it can be clearly seen, as in the last-mentioned figure, that the trichogyne and the basal cell (c’) constitute a single cell, and the nucleus is usually readily made out in the position indicated at the base of the trichogyne. Meantime the cells (x 7) and (0 0’) of fig. 13 have increased in size, and the former have begun to grow up still further around the base of cell (e), fig. 13, (e”), fig. 14. With the maturity of the trichogyne the female organ has completed its develop- ment, and we have the condition represented in fig. 15, from which it will be seen that the latter, which may be conveniently termed the procarpe, consists of three distinct parts, which, so far as is known, are present in all genera of Laboulbeniacese. Of these the uppermost may be properly called the érichogyne, and in the present illustra- tion is unicellular, though often, as will be seen presently, far more complicated in structure; the middle portion in this, as in all other instances, also unicellular, which we may call the ¢richophorie cell (e”), and the lowest portion (f/), which may be termed the carpogemc cell, being that portion of the procarpe which is fertilized, and the only part which persists and undergoes further development. As is shown in the figure, the carpogenic cell is completely surrounded by eight cells, four of which (0¢) lie be- low and around it, and, remaining undivided, form the basal cells of the mature perithecium ; while the four others (n 7) completely surrounded it, and, as will be seen later, form by further division the wall- and lip-cells of the perithecium. These eight cells are arranged in four longitudinal rows, and, in the stage represented, the upper four have already begun to grow up around the trichophoric cell, the lower half of which is now enclosed by their advancing tips. In the stage represented in fig. 15, the antherozoids, which have been continuously escaping from the antheridia since the stage represented in fig. 7 was reached, begin to adhere to the trichogyne, often in larger numbers than are represented in the figure, and fertilization is accom- plished, probably with considerable rapidity if one may judge from the rarity of conditions which are intermediate between that represented in fig. 15 and that shown in fig. 19, where the trichogyne has entirely disappeared. In the comparatively in- frequent instances in which one finds an adherent trichogyne belonging to a procarpe in which fertilization has evidently been accomplished, slight elevations may be seen which coincide with the position of usually more than one of the antherozoids (figs. 17 and 18). In such cases, although it is very difficult to determine the actual presence of conjugation in the case used for illustration, it is evident that a wall has been formed around the antherozoid, which often seems inflated and nearly empty of contents. MONOGRAPH OF THE LABOULBENIACL&. 231 In brief, then, we find the female organ developed at first as a lateral outgrowth from one of the cells of the receptacle. From this outgrowth a terminal and a sub- terminal cell are cut off. From the former of these by further division is produced the procarpe, consisting of a terminal receptive portion, the trichogyne, a middle con- necting portion, the trichophoric cell, and a lower essential portion, the carpogenic cell, which alone develops further; while from the latter arises by further division the whole of the perithecium proper. The subterminal cell thus forms the basal and wall-cells of the perithecium, while the terminal one, although at first quite free, forms its contents. The further development of the young perithecium after the fertilization of the trichogyne may be also best illustrated by reference to the same series of figures of Stigmatomyces (Plate I, figs. 17-24). In fig. 17, which represents a condition in which fertilization has been completed, the procarpe remains unchanged, except that the carpogenic cell (f/) has become somewhat enlarged and elongated. In fig. 18, the first indications of development are seen in the carpogenic cell, which, through the formation of two transverse partitions, has become divided into three superposed cells, while the trichogyne has begun to wither. As a rule, however, it has entirely disappeared when the first divisions of the carpogenic cell are visible. In fig. 19, this division of the carpogenic cell has become still more pronounced, and nothing remains of the trichogyne but its insertion. Disregarding the accompanying development of the wall-cells of the perithecium, and following only the divisions of the carpogenic cell, we may distinguish the three cells into which it first divides as follows: the lower of the three (7s) may be termed the inferior supporting cell, while the upper con- stitutes the superior supporting cell (ss). The remaining cell, which lies between the two, may be conveniently termed the ascogonium (am), and is the only one of the three which undergoes any further development; the two supporting cells eventually dis- appearing entirely. Up to this-point the development of the procarpe is similar in all the genera, so far as they are known; but the further divisions of the ascogonium show certain variations in different genera and even, apparently, in different speci- mens of the same species, although it is improbable that individual variations of this nature are at all common. In the present instance the ascogonium divides into a lower and an upper portion, the latter at the same time dividing, by somewhat irregu- larly longitudinal septa, into four cells. The lower portion (figs. 20-25, is ¢) remains unchanged, being eventually destroyed, like the two supporting cells (ss and 7s), and may be called the secondary supporting cell. We have then the ascogonium dividing simultaneously into five cells, one of them, the secondary supporting cell, remaining 222 MONOGRAPH OF THE LABOULBENIACE. sterile, while the others constitute the ascogenic cells, and at once begin to bud up- ward; the buds (fig. 21, as) developing into asci (fig. 22, as), and in this, as in a majority of cases, arising in a more or less distinctly double row. (See Plate LI, fig. 88; Plate III, fig. 1; Plate V, fig. 18.) In Stigmatomyces the four ascogenie cells, but two of which appear in the figures, are at first symmetrical neither in form, size, nor arrangement; but, as the asci begin to develop, become so placed that one is anterior, one posterior, and one lateral on either side. In fig. 23, which represents an antero-posterior view, the two lateral ascogenic cells are shown, placed more or less symmetrically with reference to one another, the anterior and posterior ascogenic cells (not shown in the optical section) occupying a similar relative position in front of and behind them, Returning now to the perithecium proper, which we left in the condition repre- sented in fig. 15, it will be remembered that it originated as a single cell (fig. 10, c), which has divided several times, and that the upper products of these divisions have grown up around the base of cell (d), from which, as we have seen, the female organ is developed. In this stage (fig. 15), it will be seen to consist of the stalk-cell (p), the secondary stalk-cell (2), and three basal cells (0), but two of which are visible in the figure, and four primary wall-cells (x, 2), which surround the carpogonium (/) and the base of the trichophoric cell (e”). At a stage slightly earlier than that represented in fig. 15 a further development from the three basal cells takes place, which is not indi- cated in the optical section. This development consists in the upgrowth from the basal cells (0) of four cells corresponding to the wall-cells, but alternating with them and lying partly between them and the carpogonium. As they continue to grow upward and to increase in size, they separate the wall-cells completely from the structures de- veloped from the carpogenic cell, growing up around the latter in a fashion exactly resembling that of the wall-cells. There are thus developed from the three basal cells, eight cells arranged in two series; an outer, the primary wall-cells, four in num- ber; and an inner, also consisting of four cells. The further growth and the succes- sive divisions of the cells of these two series, although its course is identical, is, nevertheless, quite independent in either case; the divisions of the cells of the inner series occurring in general after those of the outer series have taken place. The further development consists simply in a continued upward growth around the products of the division .of the female organ, accompanied by the separation of a terminal portion. The latter is then again separated into two portions, the upper of which divides again, and so on, until the number of cells characteristic of the genus or species has been formed. This process may be made somewhat clearer by reference to figs. MONOGRAPH OF THE LABOULBENIACEZ. 233 17-20. In fig. 17, the primary wall-cells (fig. 15, ,) have undergone their first division, having each separated into an upper and a lower cell (wandz’). The four cells of the inner series (7) have also been developed from the three basal cells (9, 0) ; but, although they extend upward above the septum which has divided the wall-cells, they are not themselves as yet septate. In fig. 18, the wall-cells remain as before, but each of the cells of the inner series has divided into two, (pc) the parietal cells and (ne) the primary canal-cells. In fig. 20 the primary canal-cells have again divided into two (nc! and ne”), and this division has been preceded by a correspond- ing separation of the upper wall-cells (w,w) into the two cells (wa and wa’). This condition continues until the asci have reached a considerable development (fig. 25) ; but before any of them are mature a last division takes place, by which the cell (7c) of the same figure is separated into the cells (cc) and (¢e) of fig. 24, and the cells (w x’) are separated into (wy) and (wz). In Stigmatomyces, then, there are present in the mature perithecium, before the spore discharge has commenced, an outer series of wall-cells disposed in four longitudinal rows of four cells each, the terminal cells of which may be conveniently called the lp-cel/s, and also an inner series of cells alternating with the outer, and also arranged in four longitudinal rows, each made up of four cells, the lower of which may be conveniently termed the parietal cells, the three others being distinguished as the canal-cells. There are thus four parietal cells, twelve canal-cells and sixteen wall-cells, making a total of thirty-two cells in the perithecium proper, exclusive of the three basal cells, the stalk-cell and the secondary stalk-cell previously alluded to (0,2, and p of fig. 15). The asci,in the mean time, have continued to bud from the ascogenic cells, so that their total bulk has greatly increased, and as a result the ascus mass begins to exert a considerable pressure im all directions on the surrounding cells. In this way the superior supporting cell, the secondary inferior supporting cell, and the parietal cells are gradually destroyed, and in most instances, though not usually in that which has been used as an illustration, the inferior supporting cell is eventually obliterated by pressure. In Stigmatomyces the lower series of neck-cells (fig. 24, nc’) become gradually inflated towards their distal ends and their walls are somewhat thickened, so that they act as guard-cells which control, to some extent, the passage of the spores from the general cavity of the perithecium to that of the neck. The asci as they mature are sloughed off from the ascogenic cells, and rapidly disappear, their walls being absorbed as soon as the spores are completely formed, so that the latter lie free in the cavity of the perithecium. The spore mass thus formed, being constantly augmented, pushes between the guard-cells (z c’) just mentioned, and being forced still 224 MONOGRAPH OF THE LABOULBENIACE. further upward, destroys the two remaining series of canal-cells (ec and ¢¢), and finally forcing their way between the lip-cells (wz) the spores make their exit through the permanent pore thus formed. The special instance which has been selected as an illustration of the development of the female organ, although it may be considered typical of the process as it occurs in the family generally, does not, as has been noted, represent the invariable course of development in all cases, when the details of the successive changes are considered ; and it will therefore be necessary to compare the processes described with the corre- sponding conditions presented by certain other genera. The exact point of origin of the bud which is to develop into the perithecium, in so far as concerns its position with reference to the cells of the receptacle, is, as has been previously mentioned, subject to many variations in the different genera, although that which has just been described is the most common. The genus Amor- phomyces, to which reference has several times been made, presents the most essential difference in this respect; since the terminal and subterminal cells of the germinating spore constitute the primordial cells of the procarpe and of the perithecium proper, respectively ; the latter dividing and growing up around the former as in the case of Stigmatomyces just described (Plate V, fig. 23,d,c). In several other cases, as in Rhadinomyces and Enarthromyces (Plate III, figs. 13-18), the female organ first appears as a free bud, developed from a cell, not necessarily the sub-basal cell, of the receptacle ; and this bud having become divided by a cross partition into two super- posed cells, the same changes which have already been described in detail, take place in essentially the same way, as will be presently noted. A very remarkable variation from the method above described by which the primordia of the perithecia and sexual organs arise from ‘the ‘receptacle, occurs in Zodiomyces ; a genus in which these organs, instead of originating as superficial out- growths, are formed as buds from a layer of cells which line the bottom of the cup-like extremity of the receptacle. This cup-shaped portion, though open at maturity, Plate XXIII, fig. 8, originates as a closed cavity below the base of the primary appendage, fig. 5, , which becomes open as a result of the destruction of the super- ficial cells above it, which is effected by numerous sterile appendages that make their way out, fig. 6, 7. The cells which give rise to the perithecia are thus primarily derived from the central parenchyma of the body of the receptacle. The course of development of the perithecia, in this instance, does not appear, however, to differ very materially from that already described. Apart from these differences in origin, the development of the female organ corre- _ MONOGRAPH OF THE LABOULBENIACE. 225 sponds very closely in all the genera, up to the formation of the procarpe, with certain differences of detail; and in all cases the mature procarpe consists, as in Stigma- tomyces, of the three essential parts above mentioned. The carpogenic and tri- chophoric cells are practically identical in all cases ; but the trichogyne is subject to very considerable variations, even in the same genus. In its most simple form it is unicellular, as in the case of Stigmatomyces, without branches, and of no great length. Other unicellular trichogynes may be more or less branched, consisting of an abruptly enlarged portion from which radiate more or less irregular short lobes or branches, which are the receptive portions, sometimes quite numerous, as in the genera Amor- phomyces and Dimorphomyces (Plate V, figs. 4, 5, 20, and 24), Camptomyces (Plate VI, fig. 4); and, to a less extent, in Peyritschiella and Dichomyces, both of which have trichogynes which are nearly simple or but slightly lobed (Plate VI, figs. 16 and 82). A similar trichogyne appears to be characteristic of Dimeromyces (Plate IV, fiz. 17); but sufficient material is needed to determine this point. The multicellular trichogynes, which are the more numerous, may be branched or simple, even in the same species, and sometimes reach a very remarkable degree of development; becoming many times septate and copiously branched, the free extrem- ities being either straight or more or less definitely spirally twisted (Plate I, figs. 1-5, ¢r,and Plate XXI, fig. 15), The terminal portion of the trichogyne alone is receptive, and it is this part which is subject to the spiral twisting, the most striking instance of which occurs in the genus Compsomyces (Plate XI, figs. 9 and 10), well developed specimens of which, like that represented in fig. 9, producing more highly developed trichogynes than are found elsewhere in the family. In all cases the receptive tips have the same refractive appearance when mature that is noticeable in the correspond- ing organs of the Floridez, and the continuity of the protoplasm of successive cells is readily demonstrated, as in other parts of the plant. However highly the trichogyne may be developed, it disappears with great rapidity as soon as fertilization has been accomplished, collapsing and breaking off, its point of insertion sometimes remaining as a scar-like prominence (Plate II, fig. 14, ¢7); while less frequently its base becomes somewhat indurated and persists even in the mature individual (Plate XXI, fig. 12 and Plate XVII, fig. 18). In a majority of cases, however, all signs of it have disap- peared at an early stage in the development of perithecium after fertilization. The process of fertilization, in so far as concerns the adherence of the antherozoids ° to the trichogyne, has been already referred to; the former in almost all cases being carried to the latter, whether they are discharged directly upon it or make their way to it by floating passively through the water, which, as we have seen, is apt in a 15 226 MONOGRAPH OF THE LABOULBENIACEZ. -~ majority of instances to surround the individuals while their hosts are hiding in moist situations. The genera in which the antherozoids are exogenous in origin probably form an exception to this rule, and the transfer may be effected as above described (p. 210). At all events, the trichogyne in the genus Zodiomyces always grows down- wards (Plate XXIII, fig. 16) as it develops and seems to seek the antherozoid which is almost invariably found attached to its tip (figs. 17, 18), and it is only after contact with the antherozoid that it turns upward as is shown in the two last-mentioned ficures. In this instance several specimens have been examined in which there seemed to have been a definite conjugation between the two organs, as in fig. 17; but asa rule this union cannot be satisfactorily demonstrated owing to the very small size of the male element. The fertilized trichogyne usually disappears before any definite change takes place in the other cells of the procarpe ; but soon after this disappearance the divisions of the carpogenic cell already described succeed one another with considerable rapidity, and correspond at first in all the genera so far as they are known. The carpogenic cell divides by two transverse septa into three superposed cells, the superior and infe- rior supporting cells and the ascogonium; and the latter, at least in many of the instances observed, becomes separated into a lower sterile part which has been above described as the secondary inferior supporting cell; while the upper part either remains without further division, constituting the single ascogenic cell, as in Amorphomyces, Sphalero- myces, Peyritschiella (Plate I, figs. 28, 29) and a few other genera; or becomes divided into two such cells, as in Laboulbenia (Plate I, fig. 35, ac) and in many other genera in which this is the usual number. Less frequently the number of ascogenic cells may be four, as in the case of Stigmatomyces above described ; while in only one genus, Haplomyces, have eght such cells been definitely observed. Although the number of ascogenic cells is moderately constant in a given genus, it does not appear to be inva- riable, and I have seen rare instances in which two were present in forms having typi- cally only one ; while in a single instance a specimen of Rhadinomyces was observed in which three were distinct within the perithecium in place of the usual four; a con- dition doubtless due to the abortion of one of the original products of the division of the ascogonium, The form of the ascogenic cell, though generally similar in most of the genera, is subject to certain variations, and the position which it occupies in the cavity of the perithecium is sometimes characteristic in given instances. In Rhizomyces, for exam- ple, it lies somewhat obliquely in relation to the axis of the perithecium, so that the ascus mass has the appearance represented in Plate III, fig. 3; while in Chitonomyces MONOGRAPH OF THE LABOULBENIACEZ. ‘ 22:7 paradowus it is nearly horizontal, so that the asci growing inward, downward, and up- ward are much bent and distorted when mature. When they occur in pairs the asco- genic cells are more or less symmetrically arranged, and when there are but two, lie facing one another in a plane at right angles to that in which the perithecium is flat- tened. Where there are two or four pairs the perithecium is but slightly if at all flattened, and the ascogenic cells lie facing one another: one anterior, one posterior, and two lateral, or approximately so, The form of the ascogenic cell is subject to little variation, being, as a rule, oval in outline in face view; but in a few genera it is much elongated, as in Amorphomyces, (Plate IV, figs. 25-28), or Sphaleromyces, and in the former case may possibly become divided by a transverse septum in certain instances. In Stigmatomyces, as has been mentioned, the asci bud from the ascogenie cell, alternating first from one side, then from the other; so that, as a result, two definite rows are formed, such as are illustrated in Plate IV, fig. 18, in Plate I, fig. 38 (at the left), and in Plate III, fig. 1, all of which represent dorsal views of the ascus mass, the ascogenic cell lying away from the observer and being consequently invisible. In the last mentioned figure, a slight irregularity is noticeable from the fact that an extra ascus has been abnormally produced at one point lying in the median line between the two rows normally formed. The most conspicuous exception to this biseriate arrange- ment of the asci is found in the genus Moschomyces, the very large ascogenic cell of which gives rise to an enormous number of asci arranged in many vertical rows; and, although such variations are exceptional, there seems in a few other cases to be some slight variation from the biseriate type. The form of the ascus at maturity is subject to unimportant variations, being practically identical in nearly all the genera. It varies from a somewhat stout and short type (Plate VIII, fig. 28 ; Plate II, fig. 11; Plate I, fig. 25) to more slender forms (Plate XI, fig. 17 and Plate XXIV, fig. 9). At the time when the spores are fully formed, the ascus is commonly short-stalked; but as the latter begins to be pushed upward by the ever-increasing mass of asci below it, the basal part often becomes drawn out into a long slender pedicel, which finally sloughs off as the ascus wall itself begins to dissolve. In almost all cases the asci are four-spored at maturity, as in the majority of the figures cited, and are distinctly flattened, the only known exceptions to this rule being presented by the two genera, Moschomyces (Plate XI, fig. 17) and Compsomyces, in both of which they are definitely eight-spored and more nearly eylindrical. The asci are never naturally discharged from the perithecium, having wholly dissolved some time before they reach the terminal pore of the latter; but by 228 MONOGRAPH OF THE LABOULBENIACE. . crushing perithecia in some aqueous stain like eosin, perfect asci in various stages of maturity are very easily obtained, either free or still attached to the ascogenic cell. Before they separate from their attachment, the asci are generally much distorted by mutual pressure, but assume a more or less regular form after they have become free. As the ascus mass or masses increase in size, their upward pressure soon destroys the superior supporting cell, as we have already seen; while their downward pressure in most instances destroys the primary and secondary inferior supporting cells, at the same time freeing the ascogenic cells from one another, if there are more than one; so that the latter eventually lie in the cavity of the perithecium, free and unconnected with any other cells. In some cases the inferior supporting cell persists after the ascogenic cells have freed themselves from their attachments, as is the case to a certain extent in Stigmatomyces, the supporting cell in this instance being so placed as to be protected by the basal cells of the perithecium which surround it. In a similar manner the inferior supporting cell in the species of Laboulbenia allied to LZ. palmelia persists permanently, for the reason that it is surrounded by the lower series of wall-cells of the perithe- cium, which are modified to form a perithecial stalk, and corticate it completely. The further destructive action of the ascus masses on the parietal and canal cells of the perithecium has already been described in connection with Stigmatomyces. With the formation of the spores and the disappearance of the ascus-wall, the his- tory of the female organ and its products is completed; but, as we have seen, the changes which it has undergone are accompanied by changes in the cells of the peri- thecium proper which are also subject to certain variations from the course of development described as characteristic of Stigmatomyces. In all cases the perithecium proper, by which is meant all portions of it exclusive of the female organ and its products, originates as a single cell (Plate I, fig. 10, c; Plate III, fig. 14, ¢) that lies wholly below the terminal cell which gives origin to the female organ in the manner above described. This cell, which has already been alluded to as the primordial cell of the perithecium, divides, in cases which have been followed out and probably in most if not in all of the other genera, into two cells more or less obliquely superposed (Plate I, fig. 11, ce”, and Plate III, fig. 15, c’,c”); the divisions of which follow in general the same course which has been described in Stigmatomyces, and may be briefly recapitulated with reference to the genus Enar- thromyces. Comparing figs. 15-17 of Plate III, which represent three successive stages of development, we have in fig. 15 the condition just referred to, in which the primordial cell of the perithecium (fig. 14, c) has divided into two obliquely superposed cells (¢’) and (¢”). In fig. 16 cach of these has divided, (c”) into the stalk- MONOGRAPH OF THE LABOULBENIACE. 229 cell (p) and an upper cell (z); while the cell (¢’) has separated into a lower cell (A), the secondary stalk-cell, and two upper cells (7,2), only one of which is seen in the ficure, the second lying opposite it on the reverse side. The two cells (7,7) and the cell (z) then continue to grow upward around the base of the primordial cell of the procarpe (d), and the cell (2) becomes separated into a lower cell (fig. 17, 0’), the pos- terior basal cell, and an upper cell (x); while of the cells (7,7), one becomes separated into a lower (0) and one upper cell (7), the other into a lower (0) and two upper cells (n,n). The two cells (0,0) constitute the anterior basal cells, and thus, together with the posterior basal cell (0’), give rise to four cells (n,n), two only of which are seen in the optical section (fig. 17). The cells (7,7) in the figure cited have already grown up around the base of the primordial cell (d) of the procarpe, the base of which is completely enclosed, and is separated from the free portion above as a distinct cell (/), the carpogonium. In fig. 18, the cells (0, 0’) have also given rise to the inner series of perithecial cells which develop as in Stigmatomyces, and are eventually all destroyed. The free part above has in the mean time developed a terminal trichogyne (tr), which is separated from the carpogonium by the trichophoric cell (e"). The four cells (n,n) then continue to grow upward, and their upper portion becomes separated by a septum. This upper portion then in turn divides into an upper and a lower part, and the process is repeated several times, until the number of wall-cells characteristic of the genus has been produced. During the formation of these wall- cells a corresponding development of the inner series of perithecial cells, fig. 18 (not shown in fig. 17), has taken place, resulting in the formation of the parietal and canal- cells previously described, which correspond in number to that of the wall-cells, and, like them, are derived as upgrowths from the three basal cells (0, 0, and 0’). The number of wall-cells which occur in a single row is usually four, as in Laboulbenia and many other genera, while in Moschomyces, Compsomyces and a few others, the number is five. As far as I am aware, the genus Ceratomyces (Plates XXIV and XXV) is the only one in which the number may not only vary in differ- ent species, but also in individuals of the same species. In this genus: the perithecium attains in some instances a most extraordinary development, and in C. rostratus the number of wall-cells in a single row may reach seventy. These rows of wall-cells may be either straight, or spirally twisted, as in Shgmatomyces Bacri, or in some of the species of Laboulbenia (Plate XIV, fig. 27; Plate XX, fig. 19), and vary in specific eases both in form, relative size, and position. For example, in the genus Laboulbenia, the lower members of the series of wall-cells, as a rule, lie opposite the ascogenic 230 MONOGRAPH OF TIIE LABOULBENIACE. cells; but in some species they may lie wholly below them, being prolonged into a stalk, as in L. longicollis, L. Kunkel, and their allies (Plate XIX, fig. 6, ete.; Plate XVIII, fig. 9). In such instances the ascus and spore masses, as they lie in the perithecium, are almost completely surrounded by the sub-basal cells of the series. In Stigmatomyces Baer, on the other hand, the basal cells of the series occupy a corresponding position (Plate I, fig. 23), while in S. eiescens (Plate VIII, fig. 2) the basal cells of the perithe- cium (not the wall-cells) extend up above the ascogenic cell and the base of the ascus mass. In a few cases the wall-cells may give rise to projections or definite appen- dages, which originate as lateral branches. Such a projection from one of the basal wall-cells is found in Chitonomyces spinigerus (Plate VIII, fig. 16). C. appendiculatus offers an example of a similar projection from one of the sub-basal cells; while in- stances of outgrowths from the terminal cells of the series are sometimes found, as in Stigmatomyces virescens (Plate VIII, fig. 1), Laboulbenia Gyrinidarum (Plate XXII, fig. 37), and a few other cases. In the genus Ceratomyces, multicellular appendages may be thus developed which may be even copiously branched (Plates XXIV and XXV); but in all the instances cited these structures are only of specific importance. The terminal cells of the series of wall-cells have been already referred to as the “lip-cells”’; since it is between them that the spores eventually force their way out from the perithecium. These lip-cells are often somewhat modified in shape, and are not, as a rule, similar and symmetrical, though sometimes so. They are usually modified to form a more or less elastic margin to the pore; as, for example, in some species of Laboulbenia (Plate II, fig. 14), the lip-cells of which are in general very irregular in form, and have their walls so modified as to give them great elasticity and thus regulate the discharge of spores. In this genus, and perhaps in some others, there is a peculiar structure at the apex of at least one of the lip-cells, represented in the figure just cited at (v), which may perhaps act as a valve, allowing the lip-cell, the cavity of which it terminates, to be more readily compressed, and at the same time to recover its form as soon as the pressure is removed. That the spore discharge may be regulated by modifications of cells other than the lip-cells has already been seen in the case of Stigmatomyces (Plate I, fig. 24); and in Sphaleromyces (Plate III, fig. 1), several sets of cells appear to assume this office. Rhadinomyces (Plate IX, fig. 13) affords an instance in which the lip-cells perform this function without any consider- able modification. Turning now for a moment to the inner series of cells which have been described as arising from the basal cells of the perithecium within, and alternating with, the wall-cells, we have seen that in Stigmatomyces their development follows that of MONOGRAPH OF THE LABOULBENIACE. 231 the wall-cells, although quite independent of it, and the same is in general true of all the genera, In forms in which, like Ceratomyces, the wall-cells become very numer- ous, a corresponding increase in the number of the canal-cells takes place ; although the number formed may be fewer, as is indicated in Plate XXIV, fig. 8, which repre- sents the growing apex of a perithecium in section, the divisions of the wall-cells (w ¢) being evidently more numerous than those of the canal-cells (¢ ¢). There seem, also, to be certain variations in the relative position of the canal-cells. In Sphaleromyces, for exampte (Plate III, fig. 2), the cells (#) are probably the lower canal-cells which have, by pushing between the wall-cells, assumed an apparently external position; and some similar modification of the more normal course of development may account for the peculiar arrangement of the cells at the tip of the perithecium in this genus, when seen antero-posteriorly as in fig. 1. It is not improbable that there are, in certain cases, variations of the course of development above described : yet in all the instances in which a careful examination has been made, it seems to correspond in all essentials. While the wall-cells of the perithecium have a definite protective function, persist- ing and forming an envelope around the ascus and spore masses, and regulating and directing the dispersion of the spores; the inner series, as we have seen, performs an entirely different function which may well be compared, in so far as concerns the cells which I have called the parietal cells, to that of the so-called tapetal cells in the sporangia of the higher cryptogams; while the cells which I have called the canal-cells bear a similar resemblance in function to the neck-cells, for example, of an archegonium. It is needless to say, however, that there is not the slightest homology between the two structures in either case. The function then of the parietal cells is to make room for the developing ascus masses, and allow them to float free within the cavity of the perithecium ; while that of the canal-cells is to afford a channel of exit by means of which the spores may make their way out through the pore of the perithecium. For, as we have seen in the case of Stigmatomyces, which is typical of the family generally, the growing ascus masses press upon and destroy not only their own connections with other cells above and below, but also the thin-walled parietal cells around them; while the mass of mature spores, which is constantly being pushed upward, has a similar effect upon the canal-cells, the latter, as a rule, having disappeared entirely when the spore discharge commences. In exceptional cases, as in that of Stigmatomyces, the walls of certain of the canal-cells (Plate I, fig. 24,c) appear to become somewhat indurated, and to regulate, to some extent, the spore discharge ; but I have observed no other instance than the one just mentioned, unless it be in the case of Sphalero- myces (Plate III, fig. 1) above alluded to. 232 MONOGRAPH OF THE LABOULBENIACEZ. It should be mentioned in this connection that in two genera (Dimorphomyces and Dimeromyces) not only do the parietal cells lose their individuality as cells, but also the basal cells of the perithecium, the stalk-cell, and the secondary stalk-cell ; so that the cavity of the stalk-cell and that of the body of the perithecium are continuous when the latter is fully mature. In these cases, as in many other genera, the cavities of the lower wall-cells of the perithecium are gradually obliterated, as the spore mass increases ; the terminal and subterminal wall-cells often being the only ones, in old individuals, in which the cell cavity can be made out. Before leaving the subject of the derivation of the two series of cells which consti- tute the outer and inner portions of the body of the perithecium proper, and of their relation to the cells below, it must be confessed that my first impression concerning the origin of the inner series was that they arose from the primary wall-cells (Plate I, fig. 15, »), through the formation of longitudinal tangential septa, it being a matter of great difficulty to make out their exact position and relations at the earliest period of their development ; and it was not till the protoplasmic connections of both the outer and inner series was observed with some exactness, in specimens of Laboulbenia the cells of which had been separated by treatment with potash, that the true condition of things suggested itself. These connections, in so far as they have been absolutely seen in given instances, are represented in figs. 16 to 18 of Plate II, which should be compared carefully with the account above given of the successive origin of the stalk-cell, the secondary stalk-cell, the basal cells, and the wall-cells; the lettering corresponding in general to that of the figures above cited on Plates I and III. In figs. 16-17 (Plate II), the parietal cells, as well as the inferior supporting cell below the ascogenic cells, had been destroyed, so that the protoplasmic connection with these cells had also disappeared; but in fig. 18, which represents a young individual, some of them were distinctly visible, and it was apparent that the inferior supporting cell (is), and the two anterior basal cells (gy) and (d), were definitely connected with the secondary stalk-cell (#), and that the two visible parietal cells ( p ¢) were similarly connected with the two anterior basal cells. A second preparation showed with equal distinctness that the posterior basal cell (o’) was similarly connected with one of the parietal cells. It was, however, impossible to determine which of the two anterior basal cells gave rise to two and which to one parietal cell; yet it is safe to assume that the same cell (d) which (fig. 17) gives origin to two wall-cells, is also connected with two parietal cells. In the preparation, as will be seen (fig. 18), three connections were vis- ible from this cell: one of which was evidently with a parietal cell, another with the wall-cell (f ), the third running beneath and connecting on the opposite side either with MONOGRAPH OF THE LABOULBENIACE. 230 a wall-cell or with a parietal cell. The fourth connection could not be distinctly made out, and may well have been broken by the crushing which was resorted to in order to separate the cells from one another. The connections of the basal cells with the wall- cells, as they are represented in figs. 16 and 17, are very readily demonstrated. It will be noticed that in fig. 17 a portion of the posterior basal cell (0’), as well as its connection with the wall-cell (7), is indicated through the anterior basal cell (d), the connections of which with ¢wo wall-cells (e) and (/) are very distinct. The connection of the basal cell (vy), shown in fig. 16, is not visible in this instance. Having considered these special cases and their modifications as far as they are at present known, it may not be superfluous, even at the risk of tedious repetition, briefly to summarize the general development of the perithecium and of the structures which it contains, since it involves matters of such considerable importance that a clear understanding of it is essential. Summary of the development of the perithecium and of the female sexual organs. The perithecium arises as a lateral, rarely as a terminal organ, and consists at an early stage of two superposed cells: an upper, which is alone concerned in the formation of the female organ, and a lower, from which is developed the perithecium proper. The upper cell elongates, by terminal growth, and is converted into the procarpe through the formation primarily of two cross partitions, by which it is separated into a lower portion, the carpogenic cell, always a single cell; a middle portion, the trichophoric cell, also always a single cell, and a terminal portion, the trichogyne, which may con- sist of a single cell, or, through the formation of cross partitions often accompanied by copious branching, of very numerous cells. The free extremities only, of the tri- chogyne, are receptive, and conjugate with the antherozoids which adhere to them. As a result of this union, the trichogyne soon withers and disappears, while the carpo- genic cell undergoes a series of divisions. First, by the formation of two transverse septa, it is separated into three superposed cells: the upper and lower constituting the superior and inferior supporting cells, respectively, which undergo no further changes; while the middle cell of the three, known as the ascogonium, divides, by more or less oblique partitions, into from two to nine cells, one of which lies at the base of the others and is called the secondary inferior supporting cell; while the one, two, four, or eight remaining cells are known as the ascogenic cells. Each ascogenic cell then begins at once to produce asci, which bud from it downward outward and upward, and soon becomes quite free in the cavity of the perithecium; destroying, as a rule, both supporting cells, and eventually the remains of the trichophoric cell, as well as the cells of the perithecium proper (parietal and canal cells) which lie immediately about and above it. 234 MONOGRAPH OF THE LABOULBENIACE. The cells which make up the body of the perithecium proper, all arise, on the other hand, from the primordial cell of the perithecium, which lies wholly below that of the procarpe. Its first division, with a few possible exceptions, separates it into two more or less obliquely superposed cells. From each of these cells upgrowths arise; one from the lower and two from the upper, which still retains its protoplasmic connection with the primordial cell of the procarpe. The lower becomes separated from its single upgrowth, and constitutes the primary stalk-cell; while the upper is similarly separated from its two upgrowths and constitutes the secondary stalk-cell. There are thus two stalk-cells surmounted by three upgrowths which have become separated from them and form three distinct cells, that begin to grow up around the base of the primor- dial cell of the procarpe. The body of each of these cells constitutes one of the three basal cells of the perithecium. From these three cells then arise eight upgrowths, four of which are external, while the remaining four are included by them and form an inner series. As a result of continued terminal growth, accompanied by the forma- tion of a variable number of septa by which the growing tip is successively separated from the portion below, an envelope is formed which completely encloses the female organ and its products: consisting of an outer series of cells arranged in four rows, each made up of a definite number of superposed cells, which constitute the wall-cells of the perithecium ; and of ansinner series of cells similarly arranged, the lowest mem- bers of which have been called the parietal cells, while those above them are termed the canal-cells. Of the two series, the outer forms the perithecial wall, the inner being destroyed, with unimportant exceptions, to make room for the ascus and spore masses, and to provide a channel through which the spores may pass to their point of exit between the apposed terminal, or lip-cells, of the outer series. General Relations and Character of the Cells. We have already seen that the gelati- nous membrane which surrounds the ripe spore persists, in the growing and in the mature plant, as a general envelope within which the cells undergo their independent divisions and modifications. This envelope, though thin, is tough and very impervious, resisting the action of reagents and staining fluids in a manner which greatly increases the difficulties associated with any attempt to observe the nuclear and other changes that take place in the cells within it. In some cases it may be separated without much difficulty, by treatment with potash, or even, when the specimen is dry, by the application of water; the action in either case resulting in the swelling of the outer layers of the cell walls and the bursting of the envelope, through which the cells themselves may protrude while still connected, as is represented in Plate III, figs. 10 and 11. In other cases the envelope appears to be more firmly and intimately con- MONOGRAPH OF THE LABOULBENIACE%. 235 nected with the outer layers of the cell walls, and, though always readily demonstrated by crushing, is not affected by the application of potash. The cells of the Laboulbeniacex, with the exception of those which lie within the wall-cells of the perithecium, and of the receptive portions of the trichogyne, are char- acterized by possessing thick walls that, when treated with potash, are readily seen to consist of a number of layers, the outer of which are less dense, and soon become swollen and gelatinous. In a majority of instances they do not seem to be connected with the general surrounding envelope by any special organs of attachment; but, in certain cases, the innermost layers of the cell wall give rise to fibrilla which, passing through the outer layers, are attached to the inner surface of the envelope. The latter, in such instances, is usually more or less conspicuously punctate, especially in cases in which it is suffused with some dark color, as, for example, in the sub-basal cell of Laboulbenia Oberthui (Plate XXII, fig. 39). It may be mentioned in passing that the suffusions, usually of black or brown, which characterize many of the species, appear to be for the most part, if not entirely, confined to the envelope ; the cell walls within them being, in general, hyaline. In some cases, however, this blackening involves at least the outer layers of the walls themselves, as, for example, in the ordinary “ foot.” The same is true also in cases where, for instance, the septum separating two cells is thus modified. This occurs very frequently in the appendages of Laboulbenia and other genera, the black- ening extending inward so that it appears to form a disc with a central perforation corresponding to the passage through which the protoplasm of adjacent cells is con- tinuous. These blackened septa were noticed by Berlese, and described in his account of Laboulbena armillaris as “ black dises or rings,” the nature of which was not determined. The fibrillz just mentioned, which, in the normal cell wall when it is viewed in optical section, give rise to the appearance described by Istvanffi as being due to the presence of “ pore-canals,’ are more or less twisted when freed by treatment with potash, and are characterized by the presence here and there of granular thickenings (Plate ILI, figs. 11 and 12), their extremities being attached to the inner surface of the general envelope in a fashion that varies in different instances. In Laboulbenia Kunkeli, for example (Plate XVIII, fig. 9), they are characteristically attached in short rows, running transversely in the sub-basal cell, but less regularly disposed in the cells above it. In other cases they may adhere in more compact groups, as in LZ. Nebrie, or singly without any definite and characteristic aggregation. These points of attachment tend to become dark-colored, especially in areas where the envelope itself 236 MONOGRAPH OF THE LABOULBENIACE2. is suffused, and give to the latter the punctate appearance peculiar to many species. The presence of these fibrille has not been definitely determined in genera other than Laboulbenia, and even in this case, although so conspicuous in many species, they no not seem to be of invariable occurrence. The substance of the cell walls as well as of the general envelope give, when treated with iodine and sulphuric acid, or with Schultze’s chloroiodide of zinc solution, no blue color, The contents of the cells, especially those of the receptacle, are usually more or less characteristic while the plant is alive, and consist of rather dense granular proto- plasm, in which, as a rule, certain highly refractive spherical oily masses are conspicu- ous. In some instances these masses are few in number, one or more of them being often very large (Plate V, figs. 4 and 5), while again they may be more numerous and uniform in size, completely filling the cells, as in the case of Laboulbema Harpali, L. Philonthi, and many others. In general, however, they are more or less variable in size, and present the appearance indicated in fig. 15, Plate II. In glycerine they soon become indistinguishable, and are thus not represented in the accompanying figures, which were drawn, with few exceptions, from glycerine preparations. The protoplasm of adjacent cells, the origin of which is the same,.is connected by a conspicuous strand of the same substance, which passes from one cell to the other through a well marked perforation of the cell wall, the connection being demonstrated with great ease by treatment with potash and subsequent staining (Plate III, figs. 11- 12; Plate II, figs. 16-18). In many instances, also, it may be seen in the living plant without the use of reagents. This protoplasmic connection is found in all the cells, in- cluding those of the trichogyne, when this organ is multicellular. In many cases in which the cells had been separated by potash, and the connecting protoplasmic strand stretched between them, I have seen a slight enlargement like that indicated in fig. 12, Plate III, recalling the similar structure through which the strands of pro- toplasm pass in the Floridex. I have, however, been as yet unable to determine its exact nature. A single, usually large, nucleus is found in the contents of every cell, and is often readily seen without the use of reagents; while in other cases it is by no means easy to demonstrate, owing to the difficulty which is usually experienced in staining the cell contents. The nuclei are spherical or nearly so, and usually contain a large nucleolus (Plate V, figs. 20 and 24; Plate I, figs. 7, 15, 16; Plate I, figs. 3 and 13). The nuclear changes accompanying cell division have not been determined ; but from the occurrence of conditions like that represented in fig. 7, Plate I, it would appear MONOGRAPH OF THE LABOULBENIACE. 237 that the nucleus divides, and that the two resultant nuclei separate before the wall is formed between them. Although a considerable amount of time has been expended in an endeavor to determine the nature of the nuclear changes which take place in the female organ during and after fertilization, I do not as yet feel in a position to make any definite statement concerning them, and have not ventured to give any drawings of the often- conflicting phenomena observed. Any one having an opportunity to study an un- limited series of specimens of Stigmatomyces Buert, for example, in a fresh condition, would probably meet with no great difficulty in determining these matters ; since from the considerable size of this species, its densely gregarious habit, and the nature of its trichogyne, it is peculiarly well adapted for this purpose. Of all the Laboulbeniaceze none, however, is better suited for study in these respects than the form described below as ELnarthromyces indicus, by reason of its large size and great simplicity ; but, as a rule, observations of this nature are made with the greatest difficulty, owing chiefly to the lack of unlimited material in the proper condition of development and the impervious character of the envelope, which so seriously interferes with the action of stains. Abnormal Morphology and Development. In examining a large body of material, cases are often met with in which the course of development, usually characteristic of a given species, is modified in various ways, and some of these modifications are sufficiently curious to call for mention in this connection. Among the simpler in- stances, abnormal septation in individuals of the cells, for example, of the receptacle, frequently occur in genera like Laboulbenia, in which the latter consists, with few exceptions, of an invariable number of cells. Such an instance is represented in fig. 9, Plate XX, the basal and sub-basal cells being thus divided, while in some cases a much more complicated cell division has been observed. An abnormal production of branches in the appendages, due, as a rule, to breakage and subsequent renewal, is very common, especially in species in which the latter are more or less filamentous, and result in an irregularity of form and branching which does not occur in normally developed plants. The multiplication of appendages, nor- mally single or definite in number, sometimes occurs, however, not as a result of in- jury ; as in Stigmatomyces, the normally single appendage of which is rarely furcate near the base, each branch becoming a functional appendage. The same is true in regard to the production of accessory perithecia. While in many forms more than one is usually produced, in a majority of genera it is typically solitary. In the latter class, however, instances are sometimes met with of the produc- 238 MONOGRAPH OF THE LABOULBENIACEA. tion of accessory perithecia. This abnormal condition has been seen several times in the genus Ceratomyces (Plate XXV, fig. 7), and occurs also in Chetomyces, Corethro- myces, Stigmatomyces, and a few others. In Rhachomyces is found the most frequent and remarkable instance of this nature, the accessory perithecia arising in this in- stance in two distinct ways. In the one case, two may occur side by side as is repre- sented in Plate X, fig. 22, or, through the proliferation of the receptacle below the base of the perithecium first formed, a second may arise a short distance above it (Plate XII, fig. 2). The same process may even be repeated, so that two or even three accessory perithecia may succeed one another, as in fig. 14 of the same plate. In the last mentioned cases the proliferation, usually, if not invariably, follows the abortion of the perithecia first formed, the trichogynes of which have, for some reason, failed to become fertilized. One further instance of the abnormal occurrence of perithecia may be mentioned which has several times been noticed in the genus Peyritschiella. In all but one of the known species of this genus, the receptacle is terminated by a single perithecium (Plate VI) which may exceptionally be replaced by two. In the species referred to (P. geminaia, Plate VI, fig. 7), the receptacle usually bears a terminal pair of perithecia. When, however, the individuals have grown under rather unfavorable conditions, as, for instance, near the extremities of the anterior pair of legs of their host, the lower transverse cell rows of the receptacle may give rise externally to several additional perithecia, which develop normally with the others. To asimilar failure of fertilization is also to be attributed the usually marked increase in the production of antheridia in such cases, which has been previously noted. Such antheridia are, as a rule, formed normally on the appendages; but in certain instances, on the other hand, this increase is effected by an abormal process, as a result of which, accessory antheridial branches take the place of the perithecium, growing from the cells at its base (Plate II, fig. 8). In some cases the branches produced under these circumstances may grow up through and within the atrophied perithecium, emerging between its terminal cells, as is represented in Plate II, fig. 9. A condition similar to this is figured by Peyritsch (18735, Plate II, fig. 11), the protruding filament having been mistaken by him for a trichogyne. A specimen in which a similar growth from within the base of the perithecium has resulted in the production of large numbers of branches is represented in Plate II, fig. 10. In this instance the abnormal growth had apparently followed an injury, by which the upper half of a normally matured perithecium had been destroyed. Such examples well illustrate the fact that the cell series of the perithecium proper are merely eight modified, but independent, filaments MONOGRAPH OF THE LABOULBENIACE. 239 which under abnormal conditions may be replaced by or associated with typically fila- mentous growths similarly derived from the basal cells. In very rare instances individuals are met with in which a typical antheridial ap- pendage, in a normally bisexual form, is substituted for the female organ, the substitution being accompanied by a great increase in the number of antheridia pro- duced. Anabnormal male individual of this kind is represented in Plate II, fig. 7, the two appendages being in general normal even to the formation of the blackened in- sertion cells. We have seen that in general the spores are discharged in pairs, and that, as a rule, and sometimes invariably, they adhere to the host and develop side by side. It is an interesting fact which may possibly have some bearing on the derivation of the dicecious from the moncecious forms, that in certain instances one member of the spore pair may normally, or not infrequently, become atrophied, or produce a smaller and weaker individual than the other. In the case of Laboulbenia inflata the atrophy of one of the spores, after it has reached an inconsiderable development, seems to be an invariable rule, as far as I have had an opportunity for observation, and groups of this species, when detached with a portion of the integument on which they are growing, show the condition of things represented in fig. 5, Plate III, the atrophy being apparent even at an early stage. Normal Variations. Like other groups of plants, the Laboulbeniacee are subject to normal variations in form, size, color, etc., which are partly inherent and partly due to the action of external causes. In individuals growing under identical con- ditions, the variations are comparatively slight, and are expressed by inconsiderable differences in gross size, or variations in the relative development of different parts, often coupled with differences in color, which are, however, in general, due to the varying age of individuals. Forms, for instance, which, when young, even when they are sufficiently advanced to discharge their spores, are hyaline or pale straw-color, may beconie, as their age increases, dark amber-brown or suffused, wholly or in part, with blackish or smoky-brown shades. Among the external causes which influence variation, the most important are associated with the character of the host, its size, and the position in which the para- site grows upon it. The color of the host, for example, often influences that of the parasite, the same species being sometimes very dark or nearly opaque on hosts with a dark or black integument, while they are pellucid, or nearly hyaline, on hosts of a ' lighter color, such differences in color being independent of differences in the age of the individuals in question. 240 MONOGRAPH OF THE LABOULBENIACE. Variations in size among different specimens of a given species are often influ- enced by two factors, namely, the size of the host and the position in which the para- site has become attached. Thus small specimens of a given species of insect will, as a rule, bear smaller parasites than larger ones, and the same is true of smaller species in a varied genus, for example, like Platynus, almost all the members of which are liable to be infested by a single species of Laboulbenia. In regard to differences de- pendent on the position of growth, it is usually true that individuals growing near the circulatory centres of the host, being presamably better nourished, are commonly dis- tinetly larger. The largest individuals that I have observed, for example, have been found growing on the thorax or prothorax about the base of the two anterior pairs of levs, while, on the same insect, those which inhabit the tips of the elytra, or of the legs, include the smallest specimens. It should also be observed that individuals growing in situations in which they are exposed to the most unfavorable conditions are apt to be thick-set, short, and stout, with short appendages. This is true, for example, in specimens of Laboulbenia elongata, L. subterranea, and various other species, when they occur, as they not infrequently do, on the mouth parts or near the tips of the legs of their hosts, the difference in general habit in such cases being often so great that such forms might easily be mistaken for distinct species. The same short, stout habit, it may be mentioned, characterizes species which are found normally in such situations and not elsewhere; as, for instance, in the case of Laboulbenia parvula, Peyritschiella minima, and others, that are, as a rule, found near the extremities of the legs, and only exceptionally in other situations. The rate of growth of the Laboulbeniaceze and the duration of their life period are matters concerning which it is not easy to make exact observations, owing, on the one hand, to the difficulty of obtaining freshly hatched hosts that have not been exposed to infection, and, on the other, to the uncertainties connected with the determination of the exact time at which the infection of the fresh hosts is accomplished. By keep- ing in confinement insects which have been collected with spores upon them just germinating and distributed on definite areas, one may estimate with considerable accuracy the time necessary for the fungus to reach maturity. This period, in the species of Laboulbenia which I have thus cultivated, has proved to be from two to three weeks. It is doubtless variable, however, in different genera; those which are more complicated in structure requiring, for their full development, a period correspondingly longer; as may well be the case, for instance, in Rhachomyces or Zodiomyces. According to Peyritsch, freshly hatched flies confined with others infested by Stigma- tomyces Baeri were found to bear mature individuals of the fungus in from ten to four- MONOGRAPH OF THE LABOULBENIACEZ. 241 teen days, according to the temperature at which they were kept; and it is certainly improbable that many forms attain maturity in a much shorter period than this. In regard to the longevity of individuals, it seems quite certain that those which have been mature in the autumn, may still produce spores during the spring and early summer; for although mature specimens which have hibernated are apt to be much damaged, and are often no longer fertile, fully mature individuals, evidently of consid- erable age, from their dark color, have been found on hosts still hibernating and col- lected by “ sifting” early in the spring before the advent of warm weather. That the germinating spores and young individuals live over winter, attached to their hosts, in a dormant condition, is an undoubted fact; and it is upon the survival of these, rather than on that of the maturer individuals, that the fungus depends for its perpet- uation. ‘That a majority of forms, occurring in temperate climates, live throughout the summer and early autumn cannot be doubted ; since one very rarely finds individ- uals that are old and infertile. When such cases occur, with few exceptions (as in Amorphomyees), the cessation of fertility has evidently resulted from some violent injury which has destroyed the perithecium, or a portion of it. From my own obser- vations in this connection, I think that it may be safely assumed that, in a majority of cases, the life period of the parasite coincides with that of the host. The number of spores formed by a single individual must therefore be often enormous, in view of the fact that the ascogenic cells are continuously active during the growing season. GEOGRAPHICAL DistriputTion. Any intelligent discussion of the distribution of the Laboulbeniacex is hardly possible, since it is in- North America only that their sys- tematic study may properly be said to have commenced; yet, judging from the small number of exotic hosts which have been available for examination in connection with the preparation of the present monograph, it may be inferred that the warmer portions of the earth are quite as rich in representatives of the group as are the more temper- ate regions, while towards the colder northern latitudes they become distinctly less abundant. For, although specimens have been examined from localities as far north as Hudson’s Bay and the Aleutian Islands, a comparatively small percentage of the hosts collected in these regions appear to be infested, while the reverse is true as the southern portion of the United States is approached. Notwithstanding the fact that the Carabide and Staphylinide, families of beetles which furnish by far the majority of the hosts at present known, are relatively much less abundant in tropical than they are in temperate regions, it seems highly probable that, in such localities, these orders are replaced by numerous other insects having suitable life-conditions, among which ' 16 242 MONOGRAPH OF THE LABOULBENIACE. orders other than Coleoptera are doubtless largely represented. This is indicated by the fact that two among the small number of tropical forms at present known occur on such peculiar and unusual hosts as those of Laboulbenia armillaris and L. Hageni, the one a mite, the other a white ant. Southern California also furnishes an instance of the occurrence of a peculiar form on a host, Chilocorus, quite unrelated to any of the usual genera of insects thus parasitized, and it must be remembered that the Diptera have hardly been examined at all in this respect. It thus seems not improbable, in view of the above facts, and of the more favorable conditions for development and perpetuation which exist in the warmer regions of the earth, that we may look to them for a large, if not the largest, future addition to the group; for that the family is des- tined to be greatly augmented can hardly be doubted. Including a small number of forms not enumerated in the succeeding pages, from lack of proper material for description, the total number of Laboulbeniaceze known at present to exist includes one hundred and fifty-eight species, from which five perhaps might well be deducted as synonyms, distributed among thirty genera, three of which are undescribed. Of these one hundred and thirteen are, so far as known, confined to North America, eleven to Europe, eight to Africa, three each to Asia and South America, including Panama, while two are peculiar to Australia. On the other hand, nine are common to North America and Europe ; two to North and South America ; one to North America, Europe, Asia, and probably to Africa; one, the determination - of which is perhaps doubtful, to North America, South America and Africa, including Madagascar ; one to South America and Africa; one to North America, Europe and probably to Asia ; and, lastly, one to Africa and Asia (Japan). Turning to the genera, one finds that of the twenty-five which are known in North America, twenty are not recorded from other continents; that of the six genera known in Europe, two have not been found elsewhere; that of four genera known from Africa, two are indigenous; while of the two Asiatic genera, one is con- fined to that continent. Lastly, South America and Australia are each known to possess but the single genus Laboulbenia, which is universally distributed. It is needless to remark that these statistics have little or no value as showing the actual relative distribution of the species and genera in the several continents, and they are merely given to indicate the possibilities of distribution in so far as our im- perfect knowledge will permit. For convenience of reference a table giving in more detail the facts of relative distribution outlined above is here appended : — MONOGRAPH OF THE LABOULBENIACEZ. 243 Totalmumber or species MMOWM tO-Oxist (si 2 2 ee wee we wwe 158 ‘“ ‘“ he St SPROROUIMER 64"), PRR ACAD ede eel eft. ¢) LOS “ «“ “ North American species Bereribed. «fe ire SPOR NEY Meee eM 2 “ be Notun Amencam@apecics NOW ©. 60. 6 ne) we C80 a tf Higpopesmepemietwiem omar. | eS aE ee eee te | AY “ ¥ A eae ec mea he Sain Ma Py tiie faetie #0 ee chia, «BE ‘“ é i South, serericam BCCICK sip 6 20s oe. lek ee ae pw ek we 6 «“ “ ROI TGNTIET Css 2” oS A Ring ek paar ee ae a 6 ¢ Ms ‘¢ Australian species. . . oy taays 2 Common to North America and Europe. — Gitesncauye es Me tagieus Bi ckamnigeee pallidus, Laboulbenia elongata, L. cristata, L. Gyrinidarum, L. lucurians, L. Nebrie, L. subterranea, L. vulgaris. Common to North America and South America. — Laboulbenia Guerinit, L. variabilis, L. polyphaga. Common to North America and Africa. — Laboulbenia Catascopi, L. polyphaga, L. elongata (?). Common to South America and Africa. — Laboulbenia Pheropsophi, L. polyphaga. Common to Africa and Asia (Japan). — Laboulbenia proliferans. Common to North America, South America, and Africa. — L. polyphaga. Common to North America, Africa, Europe, and Asia. — Laboulbenia elongata. Common to North America, Europe, and Asia (?). — Laboulbenia vulgaris. In regard to the distribution of single species in a given continent no data are available in any case except that of North America, where a wide distribution of given species is indicated. The common Laboulbenia Nebrie, which is also known to occur in Europe, is found in America from Maine to Virginia, and west to Washington, extend- ing northward to Alaska and the Aleutian islands. Laboulbema cristata is another instance of a European form which is found in this country from Maine to Nicaragua, and several examples might be mentioned of species found throughout the United States and Mexico. ‘The writer has unfortunately had no opportunity of col- lecting Laboulbeniacee on their living hosts outside of New England, and it is only in this limited territory that an approximate knowledge of their local distribution has been obtained; yet in general it may be safely assumed that the different species have a range practically coincident with that of their usual hosts. The occurrence of more or less cosmopolitan forms, like some of those mentioned above, which are doubtless more numerous than our present knowledge would indicate, has a certain interest when we consider that the possibilities of dissemination are so restricted as is neces- sarily the case where, as in the present instance, direct transfer from one living host to another is an essential factor in the perpetuation of the species. Of the more local occurrence of the Laboulbeniacexw nothing further need be said at present, since, as has just been mentioned, it coincides with that of the special hosts of the group, which will be presently referred to. 244 MONOGRAPH OF THE LABOULBENIACE. Hosts or LABoULBENIACE®. According to our present knowledge all members of the family are entomogenous, and occur only upon members of the orders Cole- optera, Diptera, and Neuroptera, with a single instance found among the Arachnida. Peyritsch in his third paper refers to an observation recorded by Hagen, where he does not state, of the occurrence of a member of the family on one of the Orthoptera; and although this mus€ be considered of doubtful authenticity, it seems more than probable that others of the insect orders, especially the Hymenoptera, may prove to be similarly parasitized. It will be noticed in the appended host index that coleopterous hosts outnumber very greatly all others; yet it should be borne in mind that this predominance of beetles in the list may be due in part to the fact that no other group has been examined with any care in this connection. The known occur- rence of species on such diverse insects as the delicate fly Drosophila, the larva of a white ant, and the acarid Antennophorus, certainly suggests many possibilities as to the types of hosts likely to be affected. There seems no good reason, then, why a much greater variety of insects than are at present known should not be added to the list, especially from the tropics, provided that they fulfil the apparent requirement of over- lapping generations and moderate longevity, coupled with a cleanly and not too solitary habit. The relative numbers infesting the various orders and families of insects may be seen at a glance from the following table, which includes all forms known to the writer, whether described or otherwise, and is approximately correct, though not abso- lutely so, in so far as regards the number of the undetermined species of hosts. TABLE oF Host Insects or LABOULBENIACER. Total number of species of insects known to become the hosts of members of the Laboul- beniacese 2b wh ee wn opi Ba see og Total number of genera represented by these hosts MEME Sova P oh Ba Mk Total number of hosts belonging to the order Coleoptera . . ... =.=... =. =~. ~. 241i Total numbers of hosts in the different families of this order: — Carabide . sole oe Se ee SS aR ene ta ante neeencas ier Haliplidas. 5 Sees oe ee ee Fe Ue atee 2 Dytiscide : cP pees 0 See va Big. fhe 8 Gyrinidja. < Soiree ak ee eee i 5 ee 15 Hydrophilida: *S Gaeeaeen ance tees a ae 4 Ae oe 9 Staphylinide:- 2a eeata ee Fee eee <6. - Pad Bite eae 50 Coccinellida: .. “2. Sappeeea eke nen 20 ee ae a 13) Sree x Total number of hosts belonging to the order Diptera . . MONOGRAPH OF THE LABOULBENIACE. 245 Total numbers of hosts in the different families of this order : — INV CCORIDICH mete) el ee Unt Ve es. « is 1s) » Genera, ds species, 3 MED RUA meal me te rR Bla hn es a ¢ 1 « 1 Birasophiide: .) case a seer ge sl kk ee SS Mret ls 9) cates wae nerdas” | 2's eyecare Ue gad Ne ue ey oe 1 Total number of hosts belonging to the order Neuroptera . . . .....2.2.42.. di A’single family Termites*: "ss 2 2). 1.) . genus, 1; species, 1 Total number of hosts belonging (omheusden Arachnida 5 /. 5.8 20. 2. ew ee 1 A single family Gasmidm . . - 4... . . . . genus,1; species,1 In comparing the species and genera represented by these hosts it is noticeable that although the Carabidee exceed all other families as regards the number of genera and of species, both of hosts and parasites, the number of genera by which they are infested is comparatively small. Of the seventy-five species of Laboulbeniacese which are found on members of this family, sixty-five belong to the single genus Laboulbenia, while the remaining ten are distributed among but five genera, Dimero- myces, Peyritschiella, Enarthromyces, and Rhachomyces. In the case of the Staphy- linid, however, which, next to the Carabidz, includes more hosts than any other family, the proportions in this respect are quite different. For although the genus Laboulbenia predominates to such a degree in the Carabide and infests all of the orders, and a majority of the families given in the above table; among the fifty species known on members of the Staphylinide, but four belong to this genus, while the remaining forty-two are distributed among no less than eighteen genera. The last mentioned family is thus by far the most interesting in this connection, and its para- sites abound in peculiar forms. It is further observable that the aquatic hosts, with the exception of the Gyrinide, the species of which swim on the surface of the water and carry their hosts at least partly exposed to the air during much of their existence, are infested only by mem- bers of the three genera Zodiomyces, Ceratomyces, Hydraeomyces, and Chitonomyces, of which the two first mentioned are the only representatives of the group of “exogenex,” producing their antherozoids exogenously. In general, it is true that among the Coleoptera infested a majority are either riparian or aquatic in their habits; yet there are a number of forms that occur upon insects which live away from the water and often inhabit rather dry situations. Of these, Chilocorus is perhaps the best example ; while, outside the Coleoptera, the com- mon house-fly affords a conspicuous instance of a host very subject to this parasitism, yet far from riparian in its habit. Peyritsch notes the fact that,in his experience, 246 MONOGRAPH OF THE LABOULBENIACE. hosts inhabiting the margins of running water were much more likely to be parasi- tized than those found along the borders of ponds or of still water generally ; but this has certainly not been my own experience, and the margins of ponds or of bogey places generally have always proved as rich in Laboulbeniaceex as those of brooks or rivers. It is also worthy of note that the statement made by the same author that iridescent hosts are not infected has not proved to be true in my experience. The relation of the parasite to its host as regards its nutrition has already been referred to, and there can be no doubt that the fluids of the insect are absorbed by osmosis through the membrane of the foot or haustorium, whether this be a highly developed penetrating organ, as in Rhizomyces or Moschomyces, or consists merely in a sucker-like external attachment. The relation of the parasite to its host, in position, is a matter of interest for sev- eral reasons; for, although all parts of the latter may be infested, the antenne, the eyes, the mouth parts, the anal plates, — in fact any portion exposed when the insect is at rest, — certain species exhibit a very remarkable definiteness in regard to their point of attachment. In some forms, like Laboulbenia Nebriw, or L. Galerite, the parasite occurs indiscriminately all over the host ; others, like L. ¢erminalis, L. fumosa, or L. lua- wriaus, only at the tips of the elytra and the adjacent region about the extremity of the abdomen. Others, like ZL. parvula and LZ. arcuata, have been found only on the legs, the latter species always in a definite position. ZL. Hurpali, again, occurs on one side only, along the anterior inferior margin of the thorax and the adjacent margin of the prothorax. LL. contorta, L. umbonata, L. elegans, L. lepida and L. texana all occur, with very rare exceptions, on the prothorax just below the external margin, usually on a definite side; and it may be noted in passing that, especially as regards the ex- tremity of the perithecium, all the last-mentioned species have a family resemblance. Instances like those just enumerated might be multiplied; but by far the most remark- able examples occur in the genus Chitonomyces. Species of this genus growing, for example, on Laccophilus maculosus, can be distinguished, as a rule, with certainty merely by noting their relative positions. C. parodorus is invariably fixed to the margin of the left elytron about half-way between its base and apex; C. marginatus occurs just within the margin of the same elytron nearer the apex. C. lichanophorus is found only on one of the inferior median anal plates (always the same plate), and so on through the list of twelve species that infest this host; the only instance which I have noticed of any-deviation in their position being in the case of C. marginatus, which is rarely found attached to the left posterior leg. | Why such constancy in position should exist in some cases, and not in others, is a MONOGRAPH OF THE LABOULBENIACES. 247 matter that I am wholly unable to explain. It is certainly not connected in any way with matters of nutrition, and although it may be in part explained as resulting from the fact that the spores are transferred while the sexes are in coitu, and the points of in- fection are determined by definftely recurring motions of the legs or otherwise, this does not seem sufficient to explain the constancy of the phenomenon. Peyritsch (1875), calls attention to the fact that Sligmatomyces Baeri invariably occurs on the under surface of the male and on the upper surface of the female host. Such a fact, if it were true, would readily be explained in this way. My own observations, however, do not agree with those of Peyritsch in this respect, though this relative position may perhaps be the more common. That certain types of form are associated with certain types of hosts seems to be a fact illustrated in a limited number of instances, —a well-marked case being afforded by the species of Laboulbenia which are found on Clivina and its near allies, and those which occur on Galerita and related forms. The species of Laboulbenia which inhabit water beetles also furnish a case in point. Parasites of Laboulbeniacee. In examining a very large amount of material I have been struck by the general absence of parasites attacking these fungi. A very small number of such parasites do, however, occur, the most striking of which belong to two forms of an apparently new genus of the Chytridinex, which grow externally on species of Ceratomyces, possessing well-developed filaments and large appendicu- late zoosporangia. The only other parasite which seems at all peculiar to the order is a minute organism, perhaps a yeast, that often completely envelopes the appendages, especially of species of Laboulbenia, assuming a characteristic semi-lunar form. The same organism sometimes makes its way into the interior of the perithecium, filling it and destroying its contents. Parasites associated with the Laboulbeniacee, and, like them, attached to living insects, are more or less abundant ; and while the majority are animal in nature (bell animalcules, rotifers, etc.), there is one genus of fungi (?), unknown to me, which, though quite colorless, resembles some of the more simple chlorozoosporic alge. It is attached to the insect (and sometimes by accident to specimens of the Laboulbeniacee) by a slightly blackened base, and consists of about three superposed cells, the upper of which produces a small number of apparently non-motile spores endogenously, that escape through a terminal opening. The genus described as Devoea,' which is evidently not in any way “related to 4 1 Lockwood, 8. Fungi affecting fishes. Jour. N. Y. Microscop. Soc. Vol. VI., p. 67 (1890). 248 MONOGRAPH OF THE LABOULBENIACE, the Saprolegniz,” but seems to be an encysted condition of some protozoan or other animal organism, is also not infrequently found on water beetles. The collection of Laboulbeniacece involves little more than the collection of a sufficient number of the proper hosts, although their presence’on the latter is not always easy to ascertain in case of the smaller forms. In so far as concerns the collection of hosts, my own experience, which is not that of a skilled entomologist, much less of a coleop- terist, indicates that the most favorable localities in which to search for infested beetles is along the margins of small streams or of ponds. In such situations abundance of hosts may usually be found under stones or sticks, or in rubbish, that may be best shaken over a sheet or other white cloth, on which the insects are readily captured. Traps deposited in such situations, and made by raking together a heap of decaying grass, alge, ete., often yield large numbers of interesting specimens when examined in this way. Many forms may also be obtained by leaving bundles of hay or grass in cultivated ground for a few days and examining them over a sheet. Water beetles are in general best obtained by sweeping the margins of ponds or ditches with a dip net; those bearing Zodiomyces occurring in cool gravel along the margins of brooks, or in cold, wet rubbish which has accumulated on rocks or branches in its bed. Staphylinidz which are either mycophagous or fimicolous have not been found to yield any parasites. Forms of this family most likely to be infested are best obtained by using the two kinds of traps above mentioned. The collection of hibernating > specinens by “ sifting” seldom yield parasites in good condition. A few hosts may be obtained on flowers: Harpalus pennsylvanicus, for example, which is very commonly infested, is usually found in abundance climbing up the spikes of Ambrosia arlemisie- folia about dusk. Laboulbeniaceze may be “ cultivated” in the manner above alluded to, by confin- ing infested hosts with such as are free from any parasites; and this is not a matter of any considerable difficulty, provided that surroundings are furnished which are suited to the habits of the insects employed. If pains are taken to keep these surroundings moist and fresh, the parasites will develop normally; but if there is not sufficient moisture to allow of a certain amount of condensation on the surface of the hosts, fertilization is apt to be interfered with, and, as a result, one finds a large number of abnormal and undeveloped forms. As a rule, it is more satisfactory and quite as little trouble to obtain specimens that have grown under natural conditions, and little, if any, advantage is to be obtained from such artificial cultivation. No attempt has been made to cultivate the spores in artificial nutrient media, and the results which might be looked for from such cultures, even if they should prove MONOGRAPH OF THE LABOULBENIACEZ. 249 to be possible, which I very much doubt, are not likely to afford data which could not be as well or better observed by the examination of successive stages in individuals naturally developed; except, perhaps, in so far as concerns the possibilities of mon- strous development which these plants might exhibit under such unnatural conditions. Preparation of Material for Evramination. Having obtained a number of hosts which are liable to be parasitized, it will found that from about five to fifty per cent. will bear parasites. In order to obtain them for examination, the host should be killed and impaled on a fine needle (a No. 12 sewing-needle mounted in a match is the most convenient), care being taken that the surface of the insect remains perfectly clean and dry, and then examined over a dull white, and then over a black surface with a hand lens magnifying about eight or ten diameters, a dissecting microscope being most conveniently used for this manipulation. Every portion of the insect should be examined in different positions, and when the parasites have been dis- covered, they should be removed by means of a dissecting needle like that just men- tioned, the needle having been inserted in the match far enough to give it the requisite stiffness, while its apex should have been ground on a fine oil-stone until a sharp, slightly oblique chisel point has been obtained. With such a point, the individuals are scraped off without much difficulty, and should be transferred to a very small drop of water on the slide. When the desired number have been thus trans- ferred, the individuals may be conveniently arranged by means of a fine hair mounted like the needles. The excess of water should then be drawn away from the speci- mens with a shred of blotting paper, and as soon as the moisture has dried around them, they should be quickly treated with alcohol, and covered with a cover glass which should always be supported by a shred of blotting paper or a chip of cover glass. The alcohol should then be replaced by water, as quickly as possible, when the individualg will be found still living, if the manipulations have been sufficiently rapid, and attached to the surface of the slide as a result of the slight drying just mentioned. Although these directions may seem superfluous, it will be found that if they are accurately followed, much trouble and the loss of many specimens will be avoided. For permanent mounting, I have used only glycerine, to which is best added a small amount of a saturated alcoholic solution of eosin, together with a trace of common salt. This preparation is very satisfactory for general purposes, and should be allowed to run under the cover glass and replace the water slowly, as it evaporates. If the replacement is too rapid, the individuals will become shrunken, but generally regain their normal form in a few days, or in a much shorter time. Dried material for herbarium specimens, which is always sufficiently good for / 250 MONOGRAPH OF THE LABOULBENIACE. purposes of determination, should be mounted in tight pill-boxes on the herbarium sheet, to avoid the depredations of Anthrenus and other pests of insect collections. A little cotton in the box prevents the host, as well as its parasites, from being in- jured by falling about when moved. Care should also be taken that the insect is kept clean and free from dust particles or exudations from its own body. Otherwise, it is often difficult or indeed impossible to detect the parasites when needed for examina- tion. ; Vt ar ary ye “Tv Oe meat Beet, 7 Fig LT i) Or —_ MONOGRAPH OF THE LABOULBENIACE. PART If. NOTE CONCERNING THE SYSTEMATIC POSITION OF THE LABOULBENIACEZ. Tue systematic position of the Laboulbeniaceze has been a matter of much un- certainty, and even in the light of a fuller knowledge, both of the forms and of their development, it still remains undetermined what are their immediate connections within the group of Ascomycetes to which they must undoubtedly be referred. As we have seen, Montagne and Robin (1853), who were the first to describe them as plants, speak of the single genus then known as “e familia Pyrenomycetum novum genus,’ and compare it to Capnodium, although they made no observations on the origin of the spores. Later, Karsten, who first included them in the Mucorini (1869), places them (1895) in a group of “ Stigmatomycetes,” between the Ustilaginee and the Pyrenomycetes; but although this author correctly observed the essential fact of the occurrence of fecundation, he denies the presence of asci, and gives a quite errone- ous account of the spore formation. Peyritsch (1871, 1873), although his observa- tions on the process of fecundation were incorrect, was the first to present any definite evidence of their ascomycetous nature ; yet it seems doubtful whether asci were seen even by him, since his reference to them as “eight to twelve spored”’ indicates the correctness of the criticism made by Karsten, who held that these “asci’’ were merely the ordinary aggregations of spores, coherent in a fusiform mass, as is their wont, and surrounded by their own gelatinous envelopes, which were mistaken for the ascus wall. This element of uncertainty in the observations of Peyritsch led De Bary to place the group among his “ Doubtful Ascomycetes,” a disposition in which he has been followed by most systematists who have alluded to the group at all. In any case, it is at present definitely determined that asci, containing four or very rarely eight spores, always occur; and that they are beyond question the morphological equivalents of the corresponding structures in the Ascomycetes generally. It further seems undeniable that these bodies are of sexual origin, in view of the evidence adduced in the foregoing pages. If, then, we admit both the sexual and the asco- mycetous nature of these plants, their consideration becomes a very important factor 252 MONOGRAPH OF THE LABOULBENIACEA. in any attempt to trace the homologies of the ascus or the derivation of the group of Ascomycetes in general. It is not my intention in this connection to enter into any detailed discussion of the several theories which have been advanced in regard to these matters; yet they cannot be allowed to pass unnoticed. In brief, it may be said that, as regards the primary origin of the Ascomycetes, authorities seem generally agreed in deriving them, in an ascending series, from the Phycomycetes; but in the discussion of the homologies of the reproductive organs in either case, the agreement has not been so striking. De Bary, as is well known, relying in a great measure on his observations in regard to the development of Sphzrotheca, as well as on the account given by EKidam of his genus Eremascus, finds little difficulty in homologizing (with his usual judicious cautiousness of statement) the asci of these genera with the oogonia of the Phycomycetes; while their sexual derivation was further substantiated through ‘the studies of Janezewski and others on Ascobolus, by those of Kihlman and others on Pyronema, by those of Stahl on the Collemacez, as well as by further observations which need not be here enumerated. In later years there has been a reaction from this view, for the most part due to the very important, yet unconvincing, researches of Brefeld. This writer, by the accumulation of a large amount of wholly negative evidence, having discarded as without significance the positive evidence just referred to, presents an argument from which he concludes that although the Ascomycetes have originated from the Phycomycetes, they have lost all traces of sexual organs. According to this view, the ascus is assumed to be merely a modified non-sexual spo- rangium, homologous with the non-sexual sporangia of the Phycomycetes; and an attempt is made to substantiate this assumption by the citation of a series of examples which, in his opinion, illustrate the actual process of evolution by which this transformation has been brought about. In still more recent years, observations made by Dangéard on the phenomena of nuclear fusion in the Ascomycetes and elsewhere, prior to spore formation, have led this writer to believe that oosporic sexuality, thus expressed, is general among the higher fungi, including the group in question; a view which, for reasons that need not here be considered, does not seem to call for serious consideration, It is thus apparent that the question under discussion has resolved itself into the phyllogeny, not of the Ascomycetes, but of the ascus; one “school” asserting its non- sexual character, the other the reverse. Supporters of the former contention, like Van Tieghem, for example, seeing in the trichogyne of Stahl a remarkably developed “ventilating apparatus,” or in the “carpogonium” of Ascobolus or Pyronema, a MONOGRAPH OF THE LABOULBENIACE. 253 group of merely vegetative cells charged with nutriment destined for a special pur- pose. Others, again, believing that the organs described are sexually significant, consider them, in many cases, functional, while in others, they may have lost their sexual character or have disappeared entirely; the presence among the Ascomycetes of purely apogamic forms being as readily reconciled with the coexistence of sexual forms as it is among the Phycomycetes, where a similar degeneration to an apogamous condition is well known to exist in not a few instances. Recent investigations, however, embodied in the very important paper on Sphe- rotheca lately published by Dr. Harper,’ indicate that while De Bary, who may be - considered the chief exponent of the view last mentioned, was correct in his general observation as to the existence of sexual reproduction in connection with the forma- tion of the ascus in this plant, he was misled by his failure to observe the very sig- nificant phenomena exhibited by the changes which take place in the carpogenic cell after its fertilization. These phenomena, which consist in the production of a series of superposed cells only one of which, and that not the terminal one, enlarges or buds out to form the solitary ascus, forbid any such direct comparison as that suggested by De Bary, between this single ascus and the oogonium of the Phycomycetes. It seems not unlikely that further and more exact observations on Eremascus may lead to some similar modification of the course of development described by Eidam ; and in any case, in view of the absence in one or in both of these instances of such evidence as they were thought by De Bary to afford in support of his own views, and the, to myself at least, wholly unconvincing character of the arguments and illustra- tions presented by Brefeld in support of his peculiar theories, one seems justified in suggesting at least the possibility of an origin for the Ascomycetes quite different from that assumed by either of these authorities. In my own opinion, the comparison made by Harper, in the paper cited, between the sexual process therein described and that of Nemalion, though it might seem at first sight hardly warrantable, becomes distinctly justified when one places between these two instances that of the present group. If, on the one hand, we compare the Laboulbeniaceze with the Floridex, a very distinct agreement is apparent between them as regards their mode of growth and general structure ; while this comparison is also suggested by the gelatinous envelope and the conspicuously developed continuity of the protoplasm between adjacent cells within it. The development of the perithecia in the one case finds a parallel in that of certain cystocarps, and the type of sexual reproduction in either group is essen- ' tially identical. 1 Ber. d. Deutsch. Bot. Gesell., Vol. XIII, p. 475 (1895). 254 MONOGRAPH OF THE LABOULBENIACE. If, on the other hand, the group is compared with the ascomycetous fungi, one finds that the differences in general structure are apparent and not real, the vegeta- tive body consisting of a septate filament which elongates and branches in a definite fashion within a general enveloping membrane. The thallus is thus not essentially different in the two cases, and the perithecium of the Laboulbeniacez is exactly com- parable to the corresponding structure in other Ascomycetes, like Spherotheca for example; the process of formation in either case involving the enclosure of a female cell, through the upgrowth around it of filaments originating below its base. It is further unquestionably true that its sexual organs and sexual reproduction are strictly homologous with the corresponding structures and phenomena that have been described in the Collemacex, in Ascobolus, in Spherotheca, and in other instances. Lastly, the products resulting from this sexual process, the asci and ascospores, are identical and homologous with those similarly resulting in the cases mentioned. That the Laboulbeniacez greatly resemble the Florideze may then be inferred from the general structure of its members, its sexual phenomena, and the development of its sporocarps and asci, resemblances which,’ it may be added, are coupled with an aquatic or sub-aquatic habit. That it belongs not among the alge, but among the fungi, is indicated by the absence of chlorophyl and of true cellulose in its cells, coupled with a parasitic habit; while the products of its sexual reproduction, the asci and ascospores, are those of fungi and by no means of alge. In view of these facts the derivation of the Ascomycetes through the Laboul- beniaceze from some branch of the Florideze seems a supposition not unworthy consid- eration, and although personally I regard such a suggestion as a mere speculation, I must confess that if one must have a theory of derivation in this case on a basis of ascertained facts, an origin of the nature above indicated is in my own opinion more probable as well as more logical than that which is usually held: since the latter in- volves the, to my mind, improbable assumption that a parasitically degraded series has attained, by a parallel course of upward development, the same goal which has been reached by the carposporic alge; the alternative, on the other hand, involving the consideration of the fungi as a heterogeneous group derived through degenera- tion at different points from types already elaborated in the algal series. Speculations of this nature are, however, of little value in the absence of facts by which they may be substantiated ; yet if the origin of the family from the Floridex is, to say the least, problematical, it is almost equally a matter of conjecture if one attempts to determine their exact point of union with the Ascomycetes in general. While their color, gelatinous consistency, and entomogenous habit might suggest a MONOGRAPH OF THE LABOULBENIACE. 255 remote relationship to the Hypocreacee, it is worthy of note that the bodies most nearly resembling the characteristic antheridial cells found in the family are the “‘ Hyphopodies mucronées”’ of the Meliolx; but having as yet been unable to ex- amine the latter in a fresh condition, I can at present merely suggest the possibility of a similarity of function. In arranging the genera under which are grouped the species included in the fol- lowing systematic enumeration, the primary divisions have been based on the characters of the male sexual organs. Forms having antherozoids exogenously pro- duced, have been separated in a group of “ exogenz,” comprising but two genera, while the remaining twenty-six genera having antherozoids that are produced endo- genously, are placed in a second group of “ endogene.”’ The two genera of the first-mentioned group are both primarily aquatic, and if we entertain the suggestion that the family has been derived from carposporic algal ancestors, might, from their simple antheridial branches and their aquatic habit, be considered as probably the more primitive of the two groups. Of this group, the genus Ceratomyces may be taken as the type, since it illustrates most clearly not only the exogenous formation of antherozoids, but the indeterminate development of the perithecia and their tendency towards an appendiculate condition. The twenty-four genera of the second group offer many difficulties, if one attempts to arrange them in lineal sequence; but here again the character of the male sexual organs affords a natural means for their general separation into two groups, charac- terized in the one case by the presence of what have been previously described as “compound,” in the other of “simple” antheridia, and these again may be sub- divided according as the sexual organs occur on the same or on different individuals. A further subdivision has been employed in the following synopsis based upon the determinate or indeterminate arrangement of the antheridial cells; but this dispo- sition, while it expresses, in a measure, true relationships, by bringing together such genera as Idiomyces, Stigmatomyces, and Helminthophana, is not wholly satisfactory. Further than this, a definite arrangement into not too numerous sub-groups is hardly possible, although it is evident that genera like Moschomyces, Compsomyces, and per- haps Sphaleromyces, or Teratomyces and Diplomyces should go together. A more definite appreciation of their further relationships will no doubt become possible after the discovery of additional genera; but at present it would be quite superfluous to attempt to represent them graphically. 256 MONOGRAPH OF THE LABOULBENIACE. GENERAL SYNOPSIS OF THE GENERA AND SPECIES WITH THEIR HOSTS. Famity LABOULBENIACE. GROUP I. Endogene. Antherozoids produced endogenously. Orper I. PryrirscHIELLE®. Antheridial cells united to form a compound antheridium. A. Dicecious. Genus I. DrmorpHomyces. Perithecia and appendages borne in pairs to the right and left of the median line. (1) D. denticulatus Thaxter, on Falagria dissecta Kr., N. America, (2) D. muticus Thaxter, on Falagria dissecta Kr., N. America. Genus IJ. Dimeromyces. Perithecia and appendages in a unilateral series. (1) D. africanus Thaxter, on Pachyteles luteus Hope, Africa. B. Moneecious. * Antheridium borne on an appendage free from the receptacle. Genus III. Cantoaromyces. Antheridium lateral below a terminal branch of the append- age. Perithecia free. (1) C. Bledii Thaxter, on Bledius assimilis, N. America. (2) C. occidentalis Thaxter, on Bledius armatus Kr., N. America. (8) C. pusillus Thaxter, on Z'rogophleus sp., N. America. Genus IV. Haputomyces. Antheridium terminal tipped by a spine-like process. Perithecia free. (1) H. californicus Thaxter, on Bledius ornatus Lec., N. America. (2) H. texanus Thaxter, on Bledius rubiginosus Er., N. America. (3) H. virginianus Thaxter, on Bledius emarginatus Say, N. America. Genus V. EKucantHaromyces. Antheridia terminal with a neck-like terminal canal, the antheridial cells in five (?) vertical series. Perithecia free. (1) E. Atrani Thaxter, on Atranus pubescens Dej., N. America. Genus VI. Campromyces. Antheridium terminal with a prominent apical pore. Anthe- ridial cells in two (?) vertical series. Perithecia free. (1) C. melanopus Thaxter, on Sunius prolixus Er., S. longiusculus Mann., N. America. ** Antheridium sessile on the body of the receptacle, sharply pointed. Genus VII. EnartHromyces. Receptacle a single series of superposed cells with one or more lateral antheridia. Perithecia free. (1) E. indicus Thaxter, on Pheropsophus sp., Asia. MONOGRAPH OF THE LABOULBENIACEZ. 257 Genus VIII. Pryrirscurerta. Asymmetrical receptacle consisting of several superposed transverse series of cells above its one or two basal cells; antheridium single, lateral. Peri- thecia free. (1) P. curvata Thaxter, on Platynus cineticollis Say, N. America. (2) P. minima Thaxter, on Platynus cincticollis Say, N. America. (3) P. geminata Thaxter, on Pterostichus luctwosus Dej., P. patruelis De}., P. erythropus Dej., N. America. (4) P. nigrescens Thaxter, on Philonthus debilis Grav., N. America. Genus IX. Dicnhomyces. Symmetrical receptacle of several superposed transverse series of cells, the basal cell single; a pair of antheridia on the subterminal series. Perithecia free. (1) D. furciferus Thaxter, on Philonthus debilis Grav., N. America. (2) D. inequalis Thaxter, on Philonthus debilis Grav., N. America, (83) D. infectus Thaxter, on Xantholinus obsidianus Melsh., N. America. (4) D. princeps Thaxter, on Philonthus sordidus Grav., N. America. Genus X. Hyprmomyces. Receptacle asymmetrical, its distal portion united to the peri- thecium ; its three basal cells superposed. Aquatic. (1) H. Halipli Thaxter, on Haliplus ruficollis De G., Cnemidotus muticus Lec., N. America. Genus XI. Cuxironomycrs. Receptacle asymmetrical, its distal portion united to the peri- thecium ; its two basal cells superposed. Aquatic. (1) C. paradoxus (Peyritsch), on Laccophilus maculosus Germ., Laccophilus sp. indet., N. America; LZ. hyalinus De}j., LZ. minutus Sturm., Europe. (2) C. appendiculatus Thaxter, on Laccophilus maculosus Germ., N. America. (3) C. distortus Thaxter, on Laccophilus maculosus Germ., N. America. (4) C. spinigerus Thaxter, on Laccophilus maculosus Germ., N. America. (5) C. uncigerus Thaxter, on Laccophilus maculosus Germ., N. America. (6) C. melanurus Peyritsch, on Laccophilus hyalinus Dej., L. minutus Sturm, Europe. (7) C. marginatus Thaxter, on Laccophilus maculosus Germ., Laccophilus sp. indect., Hydroporus spurius Lec., N. America. (8) C. rhyncostoma Thaxter, on Hydroporus spurius Lec., Laccophilus maculosus Germ., N. America. (9) ©. lichanophorus Thaxter, on Laccophilus maculosus Germ., N. America. (10) C. uncinatus Thaxter, on Laccophilus maculosus Germ., Hydroporus spurius Lec., N. America. (11) C. affinis Thaxter, on Laccophilus maculosus Germ., Hydroporus sp., N. America. (12) C. hyalinus Thaxter, on Laccophilus maculosus Germ., N. America. (13) C. simplex Thaxter, on Laccophilus maculosus Germ., Hydroporus spurius Lec., and sp. indet., N. America. (14) C. Bidessarius Thaxter, on Bidessus granarfus Aube, N. America. (15) C. borealis Thaxter, on Desmopachria convera Aube, N. America. (16) C. aurantiacus Thaxter, on Desmopachria convexa Aube, N. America. 17 I EE 258 MONOGRAPH OF THE LABOULBENIACE. Orper II. Laboulbeniea. Antheridial cells distinct, discharging independently. A. Dicecious. Genus I, AMORPHOMYCES. (1) A. Falagrie Thaxter, on Falagria dissecta Er., N. America. (2) A. floridanus Thaxter (species pro tem.) on Bledius basalis Lec., N. America. B. Moneecious. * Antheridia borne in definite series on the appendages. x. Antheridia springing directly from successive cells of the appendage. Genus Il. HetMiIntHopHANA. Appendage solitary, bearing the antheridia in four vertical series. (1) H. Nycteribie Peyritsch, on Megistopoda Westwoodii Kolen., Aerocholidia Montaguet Kolen., and Wycteribia Dufourii, Europe. Genus III. Stigmatomyces. Appendage solitary, bearing the antheridia in a single vertical series. (1) S. entomophila (Peck), on Drosophila nigricornis Loew., N. America, Drosophila fune- bris L., Europe. (2) S. Baeri Peyritsch, on Musca domestica L., Europe. (3) S. virescens Thaxter, on Chilocorus bivulnerus Muls., N. America. Genus IV. Ipiomyces. Appendages numerous, bearing the antheridia in three vertical series. (1) I. Peyritschii Thaxter, on Deleaster dichrous Grav., Europe. zx. Antheridia borne on branches of the appendages. Genus V. CoretHromyces. Appendages forming a tuft, the antheridial cells superposed and forming lateral branchlets. (1) C. Cryptobii Thaxter, on Cryptobium pallipes Grav., C. bicolor Grav., N. America. (2) C. setigerus Thaxter, on Lathrobium nitidulum Lec., N. America. (3) C. jacobinus Thaxter, on Lathrobiwm jacobinum Lec., and L. collare Er. Genus VI. Ruapinomyces. Appendage single with terminal sterile branchlets ; antheridia superposed in short series forming branchlets near its base. (1) R. cristatus Thaxter, on Lathrobium nitidulum Lec., L. punctulatum Lec., N. America. (2) R. pallidus Thaxter, on Lathrobium fulvipenne Gray., L. punctulatum Lec., L. angu- lare Lec., N. America. ** Antheridia not arranged in definite series on the appendages. Genus VII. Ratzomyces. Basal cell of the receptacle penetrating the host by rhizoidal outgrowths ; sub-basal cell giving rise to a single simple appendage bearing unilaterally a single series of branches, the basal cells of which bear antheridia. (1) R. ctenophorus Thaxter, on Diopsis thoracica Westw., Africa. Genus VIII Lasouusenta. Multicellular receptacle formed in part by the union of the base of the appendage and the stalk-cells of the perithecium. Appendages arising beside the peri- thecium usually from a black insertion cell. (1) L. anceps Peyritsch, on Platynus viduus Pz., Europe. (2) L. arcuata Thaxter, on Harpalus pennsylvanicus De G., N. America. (8) L. armillaris Berlese, on Antennophorus caput-carabis, 8. America. (4) L. (5) L. (6) L. (DL. (A). E. (9) L. (10) L. (11) L. (12) . (18) L. (14) L. (15) L. (16) L. G7) 1 (18) L. (19) L. (20) L. (21) L. (22) L. (28) L. (24) L. (25) L. (26) L. (27) L. (28) L. (29) L. (30) L. (31) L. MONOGRAPH OF THE LABOULBENIACEZ. 259 Aspidoglosse Thaxter, on Aspidoglossa subangulata Chaud., N. America. australiensis Thaxter, on Acrogenys hirsuta Maclean, Australia. Brachini Thaxter, on Brachinus mexicanus De}. and spp. indet., N. America. brachiata Thaxter, on Patrobus longicornis Say, P. tenuis Say, N. America. Casnonie Thaxter, on Casnonia pennsylvanica Linn., N. America. Catascopi Thaxter, on Catascopus guatemalensis Bates, N. America; Catascopsus two spp., Africa. Clivine Thaxter, on Clivina dentifemorata Putz., C. dentipes Dej., N. America. compacta Thaxter, on Bembidium spp. indet., N. America. compressa Thaxter, on Anisodactylus baltimorensis Say, N. America. conferta Thaxter, on Harpalus pennsylvanicus De G., N. America. confusa Thaxter, on Lembidium sp. indet., N. America. contorta Thaxter, on Platynus extensicollis Say, P. affinis Kirby, N. America. cornuta Thaxter, on Bembidium complanulum Mann., N. America. Coptodere Thaxter, on Coptodera Championi, Bates, N. America. eristata Thaxter, on Pederus littorarius Grav., P. obliteratus Lec., Pederus sp. indet., N. America; P. ruficollis Fabr., Europe. curtipes Thaxter, on Bembidium bimaculatum Kirby, N. America. decipiens Thaxter, on Galerita nigra Chey., G. equinoctialis Chaud., N. America. Diopsis Thaxter, on Diopsis thoracica Westw., Africa. elegans Thaxter, on Harpalus pennsylvanicus De G., N. America. elongata Thaxter, on Platynus cincticollis Say, P. extensicollis Say, P. melanarius Dej., P. ruficornis Lec., P. picticornis Newm., P. bicolor Lec., P. pusillus Lec., P. dissectus Lec., P. brunneomarginatus Mann., P. floridanus Lec., P. sinuatus Dej., P. ovipennis Mann., Anisodactylus baltimorensis Say, Colpodes purpuripennis Chaud., C. ceruleomarginatus Chaud., C. duplex Bates, C. grata Bates, C. petilis Bates, C. incultus Bates, C. sphodroides Chaud., C. cyanonotus Chaud., C. tenuicornis Chaud., N. America. Platynus ruficornis Geeze, Lemosthenes (Pristonychus) cavicola Sch. Europe, Platynus sp., Asia (Japan), Colpodes sp., S. America, Macrochilus biguttatus Goeze, Africa. europea Thaxter, on Chlenius eneocephalus Dej., C. chrysocephalus Rossi, Callis- tus lunatus Fabr., Aptinus mutilatus Fabr., Brachinus explodens Duft., Europe. fasciculata Peyritsch, on Chlenius vestitus F., Omophron limbatum, F., Europe. filifera Thaxter, on Anisodactylus Harrisii Lec., A. nigerrimus De}., A. interpunc- tatus Kirby, Anisodactylus spp. indet., Harpalus erythropus Dej., H. pleuriticus Kirby, N. America. flagellata Peyritsch, on Bembidium lunatwm Duft., Europe. fumosa Thaxter, on Platynus cincticollis Say, N. America. Galerite Thaxter, on Galerita Janus Fabr., G. mexicana Dej., G. atripes Lec., Galerita sp. indet., N. America. gibberosa Thaxter, on Platynus extensicollis Say, N. America. Guerinii Mont. ct Robin, on Gyretes sericeus Lab., S. America; G. compressus Lec., N. America. 260 (58) L. (59) L. MONOGRAPH OF THE LABOULBENIACEA. .Gyrinidarum Thaxter, on Gyrinus fraternus Coup., G. affinis Aub., G. ventralis Kirby, G. analis Say, G. confinis Lec., G. consobrinus Lec., G. plicifer Lec., Gyrinus spp. indet., N. America; G. wrinator Illig., Europe. . Hageni Thaxter, on Termes mozambica Hagen, Africa. s' r} DoD b] . Harpali Thaxter, on Harpalus pennsylvanicus De G., N. America. p ’ ‘1 X . inflata Thaxter, on Bradycellus rupestris Say, N. America. . Kunkelii Giard, on Mormolyce phyllodes Hagenb., Asia (Java). y g _lepida Thaxter, on Anisodactylus nigerrimus Dej., N. America. p , A ‘ J. . longicollis Thaxter, on Galerita leptodera Chaud., Gralerita sp. indet., Africa. . luxurians Peyritsch, on Bembidium spp. indet., N. America, Bembidium varium Oliv., B. bipunctatum Duft., B. flammulatum Clairv., Europe. .macrotheca Thaxter, on Anisodactylus baltimorensis Say, Anisodactylus sp. indet., N. America. .melanotheca Thaxter, on Galerita mexicana Chaud., N. America. L. mexicana Thaxter, on Galerita mexicana Chaud., G. nigra Chev., G. equinoctialis Chaud., N. America. . minima Thaxter, on Callida pallidipennis Chaud., N. America. . Morionis Thaxter, on Morio Georgii Pal., N. America. . Nebrie Peyritsch, on Webria pallipes Say, N. Sahlbergi Fisch, NV. gregaria Fisch, N. America ; WV. brunnea Duft., NV. ville Dej., Europe. . Oberthuri Giard, on Orectogyros heros Reg., Madagascar. . Orectogyri Thaxter, on Orectogyros Bedeli Reg., Africa. . Pachytelis Thaxter, on Pachyteles mexicanus Chaud., N. America. . palmella Thaxter, on Mormolyce phyllodes Hagenb., Asia (Java). . Panagei Thaxter, on Panageus crucigerus Say, P. fasciatus Say, N. America. . parvula Thaxter, on Platynus extensicollis Say, P. eruginosus De}j., Platynus spp. indet., N. America, . paupercula Thaxter, on Platynus melanarius De}j., P. ruficornis Lec., P. extensicol- lis Say, N. America. . pedicillata Thaxter, on Bembidium spp. indet., N. America. . perpendicularis Thaxter, on Bembidium spp. indet., N. America. . Pheropsophi Thaxter, on Pheropsophus cquinoctialis Linn., Pheropsophus spp. indet., S. America; Pheropsophus marginatus De}. (?), Pheropsophus sp. indet., Africa. : . Philonthi Thaxter, on Philonthus debilis Grav., P. ewnctans Horn, P. micans Grav., P. equalis Horn, P. furvus Nord, Philonthus spp. indet., N. America. . polyphaga Thaxter, on Olisthopus parmatus Say, Stenolophus limbalis Lec., S. fuli- ginosus Dej., Badister maculatus Lec., Harpalus pleuriticus Kirby, Agonoderus pallipes Fabr., Loxandrus several spp. (?), N. America; gen. indet.,S. America (?) ; on Amara sp., Africa. proliferans Thaxter, on Hudema tropicum Hope, Eudema sp. indet., Chlenius tenui- collis Fabr., C. velutinus Duft., Africa ; Dolichus sp. (?), Asia (Japan). Pterostichi Thaxter, on Pterostichus adoxus Say, P. luctuosus Dej., P. mancus Lec., P. relictus Newm., Anisodactylus nigerrimus Dej., N. America. MONOGRAPH OF THE LABOULBENIACE&. 261 (60) L. Quedii Thaxter, on Quedius vernix Lec., N. America. (61) L. recta Thaxter, on Platynus extensicollis Say, N. America, (62) L. rigida Thaxter, on Pterostichus patruelis Dej., N. America. (63) L. Rougetii Mont. et Robin, on Brachinus crepitans L., B. explodens Duft., B. scolopeta Fabr., Europe. (64) L. scelophila Thaxter, on Platynus extensicollis Say, N. America. (65) L. Schizogenii Thaxter, on Schizogenius lineolatus Say, S. ferrugineus Putz., Clivina cordata Putz., N. America. (66) L. subterranea Thaxter, on Anophthalmus Menetriesii Motsch, A. pusio Horn, N. America; A. Motschulskyi Schm., Europe. (67) L. terminalis Thaxter, on Pterostichus luctuosus Dej., N. America. (68) L. texana Thaxter, on Brachinus spp. indet., N. America. (69) L. truncata Thaxter, on Bembidium sp. indet., N. America. (70) L. umbonata Thaxter, on Stenolophus ochropezus Say, N. America. (71) L. variabilis Thaxter, on Anomoglossus pusillus Say, Chlenius estivus Say, C. euma- tilis Lec., C. cursor Chev., C. leucoscelis Chaud., C. floridanus Horn, C. pennsylvani- cus Say, C. ruficaudis Chaud., C. sparsus Lec., C. teranus Horn, C. tricolor Dej., C. viridicollis Reiche, Omophron americanum De}., Omophron spp. indet., Patrobus longicornis Say, Platynus extensicollis Say, Pterostichus adoxus Say, P. luctuosus Dej., P. corvinus Dej,, P. caudicalis Say, P. Sayi Brullé, Nebria pallipes Say, Ble- thisa multipunctata Fabr., B. quadricollis Bald., N. America; Pterostichus (?) sp., S. America. (72) L. vulgaris Peyritsch, on Bembidium mexicanum De}., B. levigatum Say, and many spp. indet., Trechus chalybeus Mann., N. America; Bembidium littorale Pz., B. fasciola- tum Duft., B. punectulatum Drap., B. lunatum Duft., B. obsoletum De}j., B. Andree Sch., B. flammulatum Clairv., B. decorum Pz., B. femoratum Sturm., B. bipunctatum Duft., Europe ; on gen. indet., Asia (?). (73) L. zanzibarina Thaxter, on Crepidogaster bimaculata Boh., Africa. Genus IX. Trratomyces. Receptacle of three superposed cells above which a series of smaller cells arranged in a transverse series give rise to numerous appendages which completely surround the bases of the (one or more) perithecia. (1) T. mirificus Thaxter, on Acylophorus pronus Kr., N. America. (2) T. Quedianus Thaxter, on Quedius ferox Lec., N. America. (3) T. brevicaulis Thaxter, on Actobius nanus Horn, N. America. (4) T. Actobii Thaxter, on Actobius nanus Horn, N. America. Genus X. DipLtomyces. Receptacle symmetrical with paired posterior projections, the ap- pendages and perithecia also paired. (1) D. Actobianus Thaxter, on Actobiws nanus Horn, N. America. Genus XI. Ruacuomyces. Receptacle consisting of a main axis of superposed cells from which on one side smaller appendiculate cells are separated, the perithecia sub-terminal. (1) R. speluncalis Thaxter, on Anophthalmus pusio Horn, N. America. (2) R. lasiophorus Thaxter, on Atranus pubescens Dej., Badister micans Lec., Acupalpus carus Lec., and gen indet., N. America. 262 MONOGRAPH OF THE LABOULBENIACE. (3) R. arbusculus Thaxter, on gen. indet., near Lathrobium, Africa. (4) R. longissimus Thaxter, on Colpodes evanescens Bates, N. America. (5) R. furcatus Thaxter, on Othius fulvipennis Fabr., Kurope. (6) R. hypogeus Thaxter, on Anophthalmus Bilimeki Sturm., Europe. (7) R. Lathrobii Thaxter, on Lathrobium longiusculum Gray., Lathrobium sp. indet., N. America. (8) R. pilosellus (Robin), on Lathrobium fulvipenne Grayv., Europe. Genus XII. Cuaromyces. Receptacle a simple series of superposed cells, the appendages and perithecium forming a single vertical series. (1) C. Pinophili Thaxter, on Pinophilus latipes Kr., N. America. Genus XIII. Spuauteromyces. Receptacle two-celled, the single simple appendage bearing a series of antheridial branchlets superposed in a single row. (1) 8. Lathrobii Thaxter, on Lathrobium mnitidulum Lec., L. punetulatum Lec., N. America. (2) S. occidentalis Thaxter, on Pinophiius densus Lec., N. America. Genus XIV. Compsomycers. Receptacle two-celled, the upper cell bearing the appendages and stalked perithecium in a whorl. Perithecium with two stalk-cells, the lower appendiculate. (1) C. verticillatus Thaxter, on Sunius longiusculus Mann., N. America. Genus XV. Moscuomyces. Receptacle two-celled, several arising from a compacted cellular base intruded into the body cavity of the host. Perithecium borne on two stalk-cells, the lower appendiculate. (1) M. insignis Thaxter, on: Sunius prolixus Er., N. America. GROUP II. Exogene. Antherozoids produced exogenously. Typically aquatic. OrDER. Zodiomycetez. Genus I. Crraromyces. Appendage tapering with lateral branches, receptacle few-celled, wall-cells of perithecium seven or more in each row. (1) ©. mirabilis Thaxter, on Tvropisternus glaber Hb., T. nimbatus Say, N. America. (2) C. confusus Thaxter, on Tropisternus glaber Hb., T. nimbatus Say, N. America. (8) C. camptosporus Thaxter, on Tropisternus glaber Hb., N. America. (4) C. minisculus Thaxter, on Tropisternus nimbatus Say, N. America. (5) C. filiformis Thaxter, on Tropisternus glaber Hb., T. nimbatus Say, N. America. (6) C. rostratus Thaxter, on Hydrocombus fimbriatus Melsh., Philhydrus cinctus Say, P. nebulosus Say, N. America. (7) C. terrestris Thaxter, on Lathrobium punctulatum Lec., N. America. (8) C. furcatus Thaxter, on Berosus striatus Say, N. America. (9) C. contortus Thaxter, on Berosus striatus Say, N. America. (10) C. humilis Thaxter, on Berosus striatus Say, N. America. Genus II]. Zopiomyces. Receptacle parenchymatously multicellular, the numerous peri- thecia surrounded by sterile appendages arising from its cup-shaped extremity. (1) Z. vorticellarius Thaxter, on Hydrocombus lacustris Lec., H. fimbriatus Melsh., and sp. indet., N. America. MONOGRAPH OF THE LABOULBENIACE. 263 In addition to the forms above enumerated, which include one hundred and fifty-two species and twenty-eight genera, a certain number of undescribed forms are known, including several new genera. Of these, three species belong to the genus Laboulbenia, two of them North Ameri- can, on Bledius and Anophthalmus, and one European, on Patrobus (Peyritsch) ; one to Cantharo- myces, on Ancyrophorus ; one to Chitonomyces, on an unknown host, together with a new genus, on Bledius, and a doubtful genus on Tachinus. The total number of forms known to exist, excluding certain doubtful species referred to by Peyritsch, is thus one hundred and fifty-eight.? | Although this number may be assumed to illustrate the group in a general way, it is certain that many important additions are inevitable, and from the data available a rough estimate of the numbers of the family existent in all parts of the world would be from five hundred to one thousand. The following descriptions, a majority of which first appeared in the “ Proceedings” of the Academy, have been largely rewritten and revised with additional notes in the light of more abundant material and a more complete knowledge of the group. As will be observed, a limited number are here described for the first time, and include some of the most important additions to the family. 1 To these must be added a new aquatic genus near Chitonomyces, which is alluded to under that genus, and two undescribed species of Ceratomyces. 264 MONOGRAPH OF THE LABOULBENIACE. Famiry LABOULBENIACEAL Pryritscu. DIMORPHOMYCES Thaxter. Plate V, figs. 1-16. Proc. Am. Acad. Arts and Sci., Vol. XXVIII, p. 157. Diccious. Male individual consisting of four superposed cells, the two distal ones sterile, the sub-basal producing a compound antheridium, the six antheridial cells of which are arranged in two antero-posterior rows and discharge into a common cavity lying above them, from which the antherozoids escape through a prominent tubular neck. Female individual consisting of four superposed cells, the two distal ones sterile, the sub-basal cell giving rise to two or more perithecia and sterile appendages which alternate with one another on either side, forming a transverse series. Trichogyne short, radiately branched. Spores once septate. The material available for the illustration of this genus, although abundant, includes, unfor- tunately, no young individuals in which the first stages in the development of the primary peri- thecia are shown. Owing to the small size of the plant and the indistinctness of its septa, it has been difficult to determine with certainty the exact relation which the perithecia and sterile appendages bear to the receptacle. While in most cases the latter has seemed to consist of three superposed cells, as in fig. 1 or 5, in a few specimens I have thought that there was an indi- cation of the presence of septa as is indicated in fig. 4. If the first of these alternatives proves to be correct, the portions of the receptacle which bear the appendages and perithecia must be considered as wing-like outgrowths from the basal cell; while in the second instance they would originate from a sub-basal cell. Although in no other genus, with the exception of Amor- phomyces, is anything approaching a similar condition found, I am inclined to think that the first-mentioned alternative is the correct one. In form and development the species of Dimorphomyces are very peculiar. There is a gen- eral triangular outline to the portion below the sterile cells of the receptacle, a tendency to a posterior convexity and anterior concavity ; and in both species the first organs to be developed are a pair of primary perithecia symmetrically placed and followed immediately by a sterile appendage placed externally on either side. In many individuals there seems to be no further development (fig. 3); but very commonly two or four secondary perithecia are formed, alter- nating with as many sterile appendages. The process by which these organs are formed is unlike any corresponding process in other genera. Assuming that my observations are correct in de- riving them from the sub-basal cell of the receptacle, the first step in their formation would consist in the production of two anterior projections from this cell symmetrically placed on either side of the median line. The tip of each projection is cut off, and the cell thus separated de- velops upward into a perithecium. Meanwhile a lateral proliferation takes place below this MONOGRAPH OF THE LABOULBENIACEZE., 265 septum, and from it arises a second projection, the upper portion of which is cut off a8 before, and develops into a primary appendage, external to each primary perithecium. This condition is represented in figs. 2 and 3. The same process may continue through the production of further successive lateral proliferations, new cells being separated as above described, and develop- ing upward into new perithecia and sterile appendages. As a result of this process a fan-like habit is developed, a series of alternating perithecia and appendages extending obliquely upwards on either side from their common point of origin in the median line of the receptacle, their bases resting upon a unicellular margin (figs. 1 and 5), which results from its repeated pro- liferation. The alternation of perithecia with appendages appears to be invariable ; except in abnormal cases, where, for example, a perithecium has failed to develop, as at the left in fig. 1. The appendages themselves present no special peculiarities, and consist merely of a single series of superposed nearly cylindrical cells. The perithecium originates from a single cell that divides into a lower and upper portion, the former the primordial cell of the perithecium proper, the latter of the procarpe. The development of the former is made out with great difficulty, the septa being very indistinct. It is probable, however, that it corresponds in general with the course of development described as typical in the first part of this memoir; although in the mature perithecium all signs of septa have disappeared, and the cavity of the stalk-cell and of the perithecium are continuous (fig. 3), the single ascogenic cell with its ascus mass floating free within. The spores are of the usual type, and are once-septate. As in the genus Amorphomyces, the members of any given spore pair produce one a male, the other a female individual ; the two sexes being thus invariably associated, as in figs. 2 and 5. The male individuals are very similar in the two known species, and are often indis- tinguishable. Owing to their minute size, I have been unable to obtain material of the youngest stages for figuring, the only early condition observed having been unfortunately lost in an attempt to mount it. In this specimen it was evident, as could be inferred from an examination of the mature individual, that the antheridium was developed as a lateral production from the sub-basal cell of the receptacle. The receptacle, as in the female, consists of four superposed cells, the two terminal ones sterile. The antheridium itself consists of a basal cell, above which are three small cells, of somewhat unequal size, from which the antheridial cells arise. The latter are arranged in three pairs; and all six cells discharge the antherozoids formed within them into a common cavity, that forms the slightly inflated base of the long tubular neck through which they are finally discharged. The formation and discharge of antherozoids continues for a considerable period, so that provision is made for the fertilization of as many secondary perithecia as may be formed upon the female individual ; and the number of anthero- zoids eventually produced by a single individual must be very great. The genus appears to bear no special relation to Amorphomyces, which, in all essential points of structure, with the exception of the remarkable similarity of its trichogyne, is widely different in both sexes. Yet it is singular that these, which are, with one exception, the only two dicecious genera so far discovered, should inhabit the same minute host. Scientifically considered, the three dicecious genera, of which this may be taken as a type, may certainly claim a position first in importance among the Laboulbeniacee, if not among the Ascomycetes as a whole, since their morphology and development would seem to settle beyond 266 MONOGRAPH OF THE LABOULBENIACE. any reasonable doubt the vexed question as to the presence of sexuality in the higher fungi. The immediate relation of the present to other genera, with the exception of Dimeromyces, is not clear ; yet it seems probable that among described forms the species of Dichomyces and Pey- ritschiella approach it more nearly than any others. In both of these instances the same ten- dency towards a bilateral development is combined with a close correspondence in the sexual organs. DIMORPHOMYCES DENTICULATUS Thaxter. Plate V, figs. 11-16. Proc. Am. Acad. Arts and Sci., Vol. XXVIII, p. 157. Male individual. Receptacle of four superposed cells, the distal one tapering upward, and terminated by a more or less distinctly marked, usually slightly blackish, knob. Antheridium arising from the sub-basal cell and partly united to the sub-terminal cell of the receptacle; ex- ternally nearly straight, bulging internally ; its main body about as long as the usually straight terminal neck, the base of which is distinctly inflated to form the cavity into which are dis- charged the antherozoids. Total length to tip of receptacle, 404; to tip of antheridium, 50— dom. Greatest width, 14. Female individual. Sterile portion of the receptacle consisting of a large sub-terminal and much smaller nearly spherical terminal cell, the main portion consisting of two superposed cells forming a more or less three-sided body, the perithecia and sterile appendages arising in the manner characteristic of the genus. Perithecia slightly inflated, rather abruptly contracted to form the stalk portion; the tip oblique; one of the lip-cells projecting as a short but distinct tooth-like prominence; while just below this prominence a conspicuous tooth-like outgrowth arises from one of the anterior sub-terminal wall-cells, and extends obliquely upward and out- ward more than half its length beyond the tip of the perithecium. The stalk of the perithe- cium is usually strongly curved, so that the latter is bent backward beyond the terminal portion of the receptacle. Sterile appendages simple, septate, tapering slightly, usually bent away from the perithecia. Spores 22-25 x 384. Perithecia 65-70 x liu. Appendages about 1104. Receptacle about 40y long. On Falagria dissecta Kr., Waverly, Mass., and Kittery Point, Maine. Better and more abundant material of this species has made necessary important changes in the original diagnosis above cited. The species is a rare one, occurring on the abdomen of its host close to the edges of the last abdominal rings on the upper or under sides, and may be seen projecting beyond their margin in a position indicated in fig. 2. They occur always in pairs, comprising the two sexes ; and numerous pairs are not infrequently found crowded together and producing a little tuft which may be seen without great difficulty under a hand lens. The anterior face is usually turned towards the substratum, and the divergence of the appendages and perithecia gives the plant a characteristic habit, indicated in fig. 11, and makes it almost impossible to obtain a view which is either wholly posterior or wholly anterior, The species seems a very constant one, varying but slightly in size, and usually producing only two peri- thecia. The host, Falagria, may be obtained in great abundance from almost any decaying vegetable matter in fields, and is most readily captured by shaking such material over a white cloth. It is the smallest host known to be infested by any of the Laboulbeniacez. MONOGRAPH OF THE LABOULBENIACE. 267 DimorPHomyces MutTICUS Thaxter. Plate V, figs. 1-10. Proc. Am. Acad. Arts and Sci. Vol. XXTX, p. 104. Male individual, as in D. denticulatus, slightly larger, the basal and sub-basal cells often more or less suffused posteriorly with blackish brown. Female individual, as in D. denticulatus, more rigid and flattened in habit, the perithecia and appendages not bent away from one another. The perithecium larger, longer, proportionately more slender, its apex truncate or but slightly oblique, the sub-terminal wall-cells producing no tooth-like outgrowth. Spores 23-26 x 3u. Perithecia 75-90 x 15u. Total length to tip of perithecium, 90-150. On Falagria dissecta Er., Waverly, Mass., and Kittery Point, Maine. This species occurs sometimes in company with the last on the same host. It is at once distinguished by its larger unarmed perithecia, which never show any indication of the promi- nence so characteristic of D. denticulatus. Its general habit, though it develops under identical conditions, is also quite different from that of the preceding species, and no tendency to an an- tero-posterior divergence is observable between the perithecia and appendages. The conforma- tion of the tips of the perithecia in either case, as shown in figs. 7 and 13, is also very different. Four or even more perithecia are not rarely developed in this species, and in fig. 1 an instance is shown of an unusually highly developed individual in which the single primary perithecium that remains is bent to the left, its fellow having been broken off or destroyed, while five new secondary perithecia are developing on either side; the youngest, at the extreme right, consisting of a mere prominence not yet separated from the proliferating marginal cell described in the preceding account of the genus. The same figure shows the remarkable phenomenon of an almost complete development of two male spores within the old primary perithecium. The species is more abundant than the preceding and more readily seen, from the greater size of its projecting perithecia. DIMEROMYCES noy. gen. Plate IV, figs. 12-17. Dicecious. Male individual consisting of a series of superposed cells from which are pro- duced, laterally, sterile appendages and antheridia in a single series. The antheridium com- pound, consisting of a stalk-cell followed by four basal cells, above which are six antheridial cells arranged symmetrically in the same plane, and discharging the antherozoids into a common cavity, whence they make their escape through a terminal orifice, at the tip of a long, slender, tubular, terminal prolongation. Female individual like the male, the antheridia being replaced by perithecia. The latter stalked, the cavity of the stalk-cell, basal cells and perithecium proper, eventually continuous, through the absorption of all the septa. This is in some respects among the most interesting of all the genera of Laboulbeniacee, since it combines with a dicecious habit a more complicated development of the male individual than is found in any other instance. In Dimorphomyces, to which it is more closely allied than to any other genus, the antheridtum is nearly, if not quite, as highly developed ; but it is always 268 MONOGRAPH OF THE LABOULBENIACE®. solitary, and the receptacle is very greatly reduced, being quite different from that of the female ; while in the present genus it differs only in its slightly smaller size. The antheridium is almost identical in structure with that of Dimorphomyces, from which it differs only in its more slender and elongate form, and in its free stalk-cell. The perithecium also, when mature, shows the same remarkable absorption of its basal septa which one finds in the last-mentioned genus, its whole cavity, from the apex to the insertion of the stalk-cell, becoming continuous about the time that the spores begin to mature. The trichogyne, as far as can be determined from a somewhat im- perfect specimen (fig. 17), is small and irregularly inflated. The receptacle is quite unique in structure, and apparently in development. The material available does not, unfortunately, illustrate the complete development, there being no very young stages; but it is evident that the young plant ends with a single primary appendage, which is the upper one of the series in the mature individual. Then, by successive divisions of the basal cell of the receptacle, new cells appear to be cut off from its distal end; each of which, in its turn, cuts off a small cell, always on the same side, from which the secondary appendages, the perithecia, and the antheridia are directly developed. A somewhat similar arrangement of organs is found in the female individual of Dimorphomyces ; but in this case the series are twofold and the proliferations terminal from wing-like lateral outgrowths. DIMEROMYCES AFRICANUS nov. sp. Plate IV, figs. 12-18. Male individual brownish. Receptacle consisting of usually seven very obliquely superposed cells, from all of which, except the basal, may be developed on the side which is uppermost, a sterile appendage or an antheridium in no regular order except that the terminal cell always bears an appendage. Antheridia rarely more than three, usually two, somewhat flattened, borne on a short free stalk-cell, the basal cells small, the six antheridial cells in two transverse rows of three each, the neck long and slender, slightly curved, its base distinctly inflated. The appendages simple, rigid, septate, tapering, becoming blackish brown ; the sub-basal cell somewhat constricted and deeply suffused with blackish brown. Antherozoids about 2.5 x .75y, rod-like. Antheridia, including stalk-cell, 60 x 104, the neck, including its inflated base, about 388. Receptacle, 125-150 x 35. Female individual like the male, but larger; the receptacle usually consisting of eleven cells, the antheridia replaced by banana-shaped perithecia, one to four in number, short-stalked, brownish; the distal end more deeply suffused, and tapering somewhat abruptly to the broadly truncate apex. Spores once septate 75 x 5.5. Perithecia, 140-175 x 26-35, including stalk- cell. Appendages (longer), 175-260. Receptacle, 120-150 x 35-50. On Pachyteles luteus Hope, Mt. Coffee, Liberia (O. F. Cooke). The types of this interesting form were found 6n two specimens of the host which were among the insects kindly loaned me for examination by Professor Cooke, and occurred on the inferior surface of the abdomen near the base of the two posterior pairs of legs. I was not able to determine from the material whether the sexes always grow in pairs, since from their position of growth it was not possible to remove a portion of the chitin bearing them without injuring the insect. It is altogether probable, however, that the development of the spore pairs corresponds to that of the other dicecious forms (Amorphomyces and Dimorphomyces). MONOGRAPH OF THE LABOULBENIACEZ. 269 HAPLOMYCES. Plate VII, figs. 1-10. Proc. Am. Acad. Arts and Sci. Vol. XXVIII, p. 159. Receptacle consisting of two small superposed cells from which arise the single perithecium and the single antheridial appendage. VPerithecium large, pointed, borne on a single stalk-cell surmounted by three basal cells. Antheridial appendage consisting of a basal cell surmounted by a terminal body, the antheridium, entirely divided by anastomosing septa into numerous small cells, and furnished with a short lateral projection, together with a sub-terminal short spine- like process arising from a rounded base. Asci four-spored, arising from eight ascogenic cells. Spores once septate. Owing to the fact that no fresh material of this genus, or any of its younger stages have been examined, there are numerous points connected with it which need to be made clear. Unlike the succeeding genus, Cantharomyces, its nearest ally, the antheridium appears to be ter- minal, not lateral. No discharge of antherozoids was observed in any of the specimens examined, and it is uncertain how and where such discharge takes place. Certainly not through the ter- minal spine-like process, so characteristic in the genus, which is undoubtedly a peculiarly modified sterile cell, perhaps the remains of the original terminal spore segment. It seems more probable that a lateral projection prominent in some specimens (at the right in fig. 3) rep- resents the point of discharge; but what relation the irregularly honeycomb-like mass of cells composing the body of the antheridium bears to this projection, or how many of these cells are really spermatic, it is impossible to say. Examined with an immersion, indications of a central cavity, containing small roundish bodies, probably antherozoids, and extending upward and out- ward to the external projection previously mentioned, may be made out, though not with suffi- cient definiteness to enable one to figure these structures. From analogy with related genera, however, there can be little doubt that some such arrangement of the spermatic cells about a common cavity must exist. The perithecium is remarkable from the fact that it contains eight ascogenic cells arranged symmetrically in four pairs, a condition only occurring in this and, perhaps, in the succeeding genus. 4 The hosts of these curious little forms all belong to the staphylinid genus Bledius, common further south along the sandy or gravelly margins of streams, especially in shady places, where they may be found under stones or burrowing in the sand. The only material examined has been that contained in the collection of the Museum of Comparative Zoology at Cambridge. HAPLOMYCES CALIFORNICUS Thaxter. Plate V, figs. 1-4. Proc. Am. Acad. Arts and Sci. Vol. XXVIII, p. 159. Perithecia olive brown, tapering abruptly to the small blunt apex, greatly inflated exter- nally, internally nearly straight; a distinct prominence, the base of the old trichogyne, usually visible on its anterior face below the apex, its basal cells short, wider than long, more or less suffused with olive brown, the stalk-cell stout, sub-triangular, hyaline. Receptacle small; its basal cell nearly hyaline, more than twice as large as the sub-basal, which is intensely blackened 270 MONOGRAPH OF THE LABOULBENIACE. and quite opaque, as is the basal cell of the appendage, except along its inner margin. Anthe- ridium twice as long as broad, slightly convex inwardly and concave outwardly ; its thorn-like appendage sharp, prominent. Spores, 837-40 x 3y. Perithecia, 130-145 ~« 65 py, the stalk-cell 45-50 X 38 4. Receptacle, 45-55 x 15-18 w. Antheridium, 83 X19. Total length of appen- dage, 48-554. ‘Total length to tip of perithecium, 240-260 pu. On Bledius ornatus Lec., California. This species was found growing on the abdomen of a single specimen of its host, examined in the collection of the Museum of Comparative Zoology. It seems to be sufficiently dis- tinguished from the succeeding species by its olive brown color and very differently shaped peri- thecium. The fourteen type specimens show no variation from the form represented in the plate. HAPLOMYCES viRGINIANUS Thaxter. Plate V, figs. 7-10. Proe. Am. Acad. Arts and Sci. Vol. XXVIII, p. 160. Perithecium short, stout, straw-yellow, outwardly inflated, the inner margin nearly straight, the distal portion conical, its apex bluntly pointed; stalk-cell long, nearly cylindrical, distally expanded slightly, the basal cells of perithecium very small, almost obsolete. Basal cell of receptacle large, the distal cell very small, and several times as broad as long, the stalk-cell of the perithecium arising from it, but also connected with the distal portion of the basal cell- Basal cell of antheridial appendage squarish, slightly broader than long, the antheridium rounded, its reticulations coarse, the thorn-like apiculus very fine (about 5 w long) and abruptly distinguished from its flattened base. Spores, 83 < 3.7. Perithecia, 110-130 x 55-60 pw. Stalk-cell of perithecium, 75-110 « 19-25 w. Antheridium, 18 « long, 22-234 wide. Receptacle, basal cell, 45-50 ~ 18-19 yw, distal cell about 18.5 6y. Total length to tip of perithecium, 220- 275. Total length of appendage, 30-33 pu. On Bledius emarginatus Say, Virginia. This curious little species was found growing on the abdomen of its host. It is very dis- tinct from either of the remaining forms, and although approaching H. californicus more closely in the shape of its perithecium, is at once distinguished by the great elongation of the latter’s stalk-cell, which is associated with a remarkable reduction in the size of its basal cells. The sub-basal cell of the receptacle is so reduced that the stalk-cell of the perithecium appears at first to arise from the basal cell. The antheridial appendage, though almost identical in the other two species, is here distinctly different in form and general appearance. HAPLOMYCES TEXANUS Thaxter. Plate V, figs. 5-6a. é Proc. Am. Acad. Arts and Sci. Vol. XXVIII, p. 160. Perithecia almost symmetrically conical, straw-yellow, tapering to the rather blunt apex ; the stalk-cell nearly hyaline, more than twice as long as broad, more slender at the base; the basal cells elongate, nearly equal. Receptacle small, the basal cell nearly hyaline; the distal cell very small, slightly broader than long, sometimes blackened and opaque, the opacity including the lower outer portion of the basal cell of the appendage, the unblackened portion of which, MONOGRAPH OF THE LABOULBENIACE®. 11 together with the antheridium, becomes suffused with brown. Spores, 40-45 x 3.7. Perithecia, 165-185 x 50-55 w. Stalk-cell of perithecium, 65-90 x 26-33 4. Antheridium, 82x 18 yp. Re- ceptacle, 87-45 x 184. Total length to tip of perithecium, 315-3870 uw, Total length of anthe- ridial appendage 35-40 yp. On Bledius rubiginosus Er., Texas. Distinguished from the preceding species by its pale yellowish color, conical, nearly straight perithecium, and by the elongated basal cells of the latter. The two varieties of this form represented in figs. 5 and 6 occurred together on the abdomen of the same individual; and though the blackened form may be taken as the type, it is improb- able that the two are distinct, since similar variations in color are common in other cases. CANTHAROMYCES Thaxter. Plate V, figs. 11-24. Proc. Am. Acad. Arts and Sci. Vol. XXIV, p. 9; lc. Vol. XXVIII, p. 161. Receptacle consisting of two superposed cells, the distal producing one or more stalked peri- thecia, and one or more antheridial appendages. Perithecia sub-conical, borne on a single stalk-cell surmounted by three basal cells. Antheridial appendages consisting of two superposed cells, terminated by one or two cells which may bear several branches, the sub-basal cell divided into two parts longitudinally or obliquely, one of which (the antheridium) is subdivided by anastomosing septa into numerous small cells. Spores once septate. Trichogyne filamentous. This genus is very closely allied to the preceding, differing only in the fact that the anthe- ridium is not terminal, but lateral, on an appendage terminated by sterile branches and destitute of the spine-like process characteristic of Haplomyces. The general relations of its parts are otherwise similar, and its antheridium possesses the same characteristic irregularly honey- combed structure. In C. Bledii, the only species abundant material of which has been examined, two or even three perithecia may arise from the same receptacle, and very rarely two append- ages. In the absence of fresh material, except in the case of the very minute C. puszllus, the exact nature and relations of the spermatic cells remains a matter of doubt. In one specimen of C. Bledii, a well-developed trichogyne is present, rather copiously branched and bearing no resemblance to the greatly reduced type characteristic of Peyritschiella and its allies. As in Haplomyces, there appear to be eight ascogenic cells, but this point has not been definitely determined. The species all occur on beetles belonging to the Staphylinidiz. CANTHAROMYCES BuiepII Thaxter. Plate VI, figs. 17-24. Proc. Am. Acad. Arts and Sci. Vol. XXIV, p. 10. Color pale yellowish. Perithecia sub-conical, nearly symmetrical, basally inflated, borne on a well-developed stalk-cell surmounted by three smaller basal cells, one to three arising from the sub-basal cell of the receptacle. Receptacle consisting of a basal and somewhat larger sub- basal cell; often deeply and broadly blackened externally and inferiorly, in which case it pro- jects laterally. Appendage usually single, rarely two arising from the same receptacle; 272 MONOGRAPH OF THE LABOULBENIACEZ. consisting of a squarish basal cell followed by a somewhat larger sub-basal cell, almost the whole of which becomes divided into numerous small cells to form the antheridium proper, which is bordered internally by a narrow sterile cell above which the appendage becomes repeatedly. and variously branched; the branches seldom extending beyond the middle of the perithecium. Spores, 25 x 3.5 u. Perithecia, 90-130 x 33-55 w, average, 114 x 42. Total length of appendage, 90-180 w, average, 150. Total length to tip of perithecia, 200-870 pu, average, 280 pw. On Bledius assimilis, Champaign, Ill. I am indebted to the courtesy of Prof. S. A. Forbes for abundant material of this species, kindly sent me from the locality mentioned. A single undetermined species of Bledius, from northern Illinois in the Museum of Comparative Zoology, was also found infested by the same parasite; but no fresh material has been available for examination. The species varies very considerably in size ; and forms in which the sub-basal cell is blackened might be mistaken for a distinct species, owing to the slight distortion which accompanies the discoloration. It occurs on all parts of the host, even the legs and antenne. A species bearing some resem- blance to the present form was also found on a species of Ancyrophorus from Lake Superior ; but the two specimens examined are not sufficiently perfect to warrant their description. CANTHAROMYCES OCCIDENTALIS Thaxter. Plate VII, figs. 15-16. Proc. Am. Acad. Arts and Sci. Vol. XXVIII, p. 161. More or less suffused with brown. Perithecium rather short, sub-conical, slightly inflated toward the base; the distal portion very slightly curved outward, the apex bluntly pointed ; stalk-cell large, cylindrical, not exceeding the antheridium; basal cells each several times as long as broad. Basal cell of receptacle very small, the sub-basal cell much larger, inflated without blackening, or externally and inferiorly deeply blackened and slightly incurved. Basal cell of the appendage similar to the distal cell of the receptacle, larger and unmodified, or similarly blackened ; the sub-basal cell large, its upper inner portion obliquely cut off and divided into numerous small cells to form the antheridium, which bulges slightly on the inner side; the ap- pendage terminated by several superposed cells. Perithecium, 96-100 x 60 w. Stalk-cell of per- ithecium, 120-140 x 264. Length of appendage to tip of antheridium, 110-150». Total length to tip of perithecium, 280-335 pu. On Bledius armatus Er., Utah. Distinguished from C. Bledii by its brown color, the greater elongation of the basal cells of the perithecium, and the relatively small antheridium, which in C. Bledzi is external and com- prises five-sixths to three-quarters of the sub-basal cell. The two specimens examined have each a single perithecium, and but one appendage, the tip of which is somewhat broken. A number of young specimens apparently belonging to this species, and occurring on B. jacobinus Lec., have the terminal portion of the appendage unbranched and consisting of a short tapering series of superposed cells. The figures given represent the two type specimens which were found on the abdomen of the host. MONOGRAPH OF THE LABOULBENIACE. 2738 CANTHAROMYCES PUSILLUS Thaxter. Plate VII, figs. 11-14. Proc. Am. Acad. Arts and Sci. Vol. XXIX, p. 100. Perithecium becoming reddish brown, inflated just above the base, the distal portion conical, tapering to a blunt symmetrical apex, borne on a rather short narrow stalk-cell bent towards the appendage and separated from the perithecium by three small sub-triangular basal cells, Receptacle consisting of a very small basal and a much larger rounded sub-basal cell, more or less suffused with brown, which gives rise to the stalk-cell of the perithecium and the appendage. Antheridial appendage consisting of a large squarish basal cell followed by the antheridium proper, which is primarily a large squarish cell, its outer half, or more, becoming divided by anastomosing septa into numerous small cells, the inner portion also showing a division into two or three larger cells ; the whole bearing terminally a series of usually three superposed flattened cells, strongly constricted at the septa, and giving rise distally to from one to three simple cylin- drical, nearly hyaline, sparingly septate branches, usually exceeding the perithecium in length. Spores, 18 x 2. Perithecia, 22-26 x 30-55 uw. Total length to tip of perithecia, 80-85 y, to tip of appendages, 90-120 w. On Trogophleus sp. York, Maine; Waverly, Mass. This species is perhaps the smallest of the known forms of Laboulbeniacex, and is some- what difficult to discover and remove from the legs or elytra of its host, where, however, it is not rarely found. Owing to its minute size, the detailed structure of the antheridium was not plainly made out, neither was any discharge of antherozoids noticed. Its structure corresponds so closely, however, to that characteristic of the genus as emended, that there can be little doubt of the correctness of its generic reference. It occurs more commonly on the legs or near the tips of the elytra of its host, a small blackish staphylinid common on wet logs along the margins of brooks or in wet rubbish caught in similar situations. EUCANTHAROMYCES, Thaxter. Plate V, figs. 25-27, Receptacle consisting of two superposed cells, giving rise on one side to a free stalked peri- thecium, on the other to a free appendage. The appendage consisting of a basal and sub-basal cell terminated by a compound antheridium. The antheridium formed from numerous small cells, obliquely superposed in three rows, bordered externally by a sterile cell, and terminated by a cavity from which the antherozoids are discharged through a short, irregular, finger-like projection. This genus is based upon the peculiar structure of its antheridium, which appears to be dis- tinctly different from the honey-comb-like body which occurs in Haplomyces and Cantharomyces, as well as from the more simple form illustrated by Camptomyces, which is, probably, its nearest ally. The material examined consists of but two type specimens, in one of which the antheridium was in perfect condition, the terminal cavity being filled with antherozoids. It is impossible to determine in this specimen whether all or only a few of: the small cells which compose the antheridium are antheridial cells; but, from analogy with Dimorphomyces and Pey- ritschiella, it seems probable that the majority of them are; and that the cells represented in 18 274 MONOGRAPH OF THE LABOULBENIACES. the figure (Plate VII, fig. 27) merely represent the bases of such antheridial cells; the terminal portions extending upward and inward, and discharging into the common cavity. This matter cannot, however, be determined without an examination of fresh material of immature specimens. EvcANTHAROMYCES ATRANI Thaxter. Plate V, figs. 25-27. Proc. Am. Acad. Arts and Sci. Vol. XXX, p. 480. Pale straw-colored. Perithecium rather long, slightly inflated, tapering to a blunt apex with rounded lips, its stalk consisting of a single large free basal cell surmounted by three smaller cells. Basal and sub-basal cells of the receptacle long and very obliquely superposed, lying almost side by side. The appendage consisting of a basal cell not wholly free, but partially con- nected with the stalk-cell of the perithecium at its base, followed by a second sub-triangular cell, the oblique upper walls of which separate it on the inside from the body of the antheridium proper, and on the outside from the narrow cell which forms the sterile margin of the latter. Antheridium sub-cylindrical with rounded apex, consisting of three series of obliquely super- posed cells, decreasing in size from below upward, and running obliquely upward and outward, the lower series of six cells, the middle of four, and the upper of two; the three series terminating in a common cavity filled with antherozoids, which are discharged through a terminal irregular, finger-like projection, which is bent strongly outward. Perithecium, 135 x 35 p. Length to tip of perithecium, 2604. To tip of antheridium, 150 p. On Atranus pubescens De}j., Virginia. Two specimens of this form were found in company with Rhachomyces lasiophorus, on an example of Atranus kindly sent me by Mr. Pergande, and, although carefully sought for on all the available material of its rather rare host, has not been again observed. Neither of the speci- mens, which are both figured in the plate, are fully mature, and there is doubtless some slight further development of the perithecium. It is unlikely, however, that any difficulty will be encountered in its subsequent identification. CAMPTOMYCES Thaxter. Plate VI, figs. 1-6. Proc. Am. Acad. Arts and Sci. Vol. X XIX, p. 109. Receptacle consisting of two superposed cells, the upper bearing the short-stalked perithe- cium laterally and the antheridial appendage terminally. Perithecium narrow, with coarse- lipped asymmetrical apex. Appendage consisting of a single large basal cell bearing the antheridium terminally. Antheridium multicellular, sub-conical, with a prominent terminal pore for the discharge of the numerous roundish antherozoids. Trichogyne developed as a small vesicular prominence from a permanent ear-like appendage which arises laterally from the young perithecium. Ascogenic cells two in number. Spores once septate. It is much to be regretted that in this genus, as in all the genera closely allied to it in which the antheridium appears to be a terminal organ, no material is available which illustrates the first steps in its production. In the youngest conditions examined, the antheridium has, in all instances, attained nearly complete development, even when the perithecium is a mere bud from MONOGRAPH OF THE LABOULBENIACE. 275 the sub-basal cell of the receptacle. Whether the antheridium as a whole is truly terminal, and developed from the smaller of the two primary spore segments, is uncertain. When mature, it is a somewhat complicated organ ; but in a general way corresponds to that of Dimorphomyces, except in the number and arrangement of its antheridial cells. The latter appear to be placed in probably two, possibly three, vertical rows, each cell extending obliquely inward and upward towards a general lateral and partly central cayity into which they discharge, and which is usu- ally well filled with irregularly squarish or roundish antherozoids. I have been unable to make out to my own satisfaction whether all the smaller cells of the antheridium are spermatic, and what their exact disposition is. The figures given (5 and 6) represent only in a general way the relations of the cells to the cavity for the reception of antherozoids; the view in the one case being sectional in a plane which does not show the openings through which the antheridial cells discharge their contents ; in the other (fig. 6), showing the appearances visible on the anterior (inner) swrface of the organ. The antherozoids are produced in very large numbers, and no form that I have seen is so well adapted to illustrate their discharge, which almost invariably occurs when the specimen is placed in water. This is perhaps connected with the fact that the trichogyne is very rudimentary, consisting of a small vesicular prominence, with one or two short projections, and is not in close proximity to the antheridium, so that the production of a large number of male bodies is necessary to insure fertilization. The carpogenic cell is very large, and the trichophoric cell bears the trichogyne laterally and anteriorly, the latter subtended by a peculiar and well-marked prominence. This prominence, at the left in fig. 4, constitutes the basal portion of the trichogyne, the receptive portion of which is seen in the angle between it and the projecting apex of the trichophoric cell. The latter is separated from the base of the trichogyne by a septum which was accidentally omitted from the figure. The cavity of this basal portion, which becomes indurated and persists as a slight projection on the mature perithecium (fig. 2), is not, as represented, continuous with that of the trichophoric cell. The ascogenic cells are two in number, and the asci do not seem strictly biseriate. CAMPTOMYCES MELANOPUS Thaxter. Plate VI, figs. 1-6. Proc. Am. Acad. Arts and Sci. Vol. XXIX, p. 101. Perithecium tinged with brownish, slightly inflated towards the base, its distal half narrow, tapering gradually to the rounded attenuated apex, below which on one side is a rounded pro- jection ; borne on a large sub-triangular stalk-cell, surmounted by three smaller basal cells. Receptacle narrowly funnel-shaped, tapering to a pointed base, and consisting of a large basal cell, slightly translucent near its lower extremity, but otherwise becoming wholly opaque, followed by a flattened sub-basal cell, from which the mature perithecium with its stalk project _nearly at right angles to the long axis of the receptacle, while distally it bears the appendage. Appendage consisting of a single squarish cell, followed by a flattened cell which bears terminally the sub-conical, slightly asymmetrical antheridium. Spores about 30x 3.5y. Perithecium, 130-150 x 30-33 w. Total length to tip of antheridium, 110-125. Greatest width, 25 pw. Antheridium, 25 x 16 uw. On Sunius prolixus Er., Waverly, Mass., and York, Maine. On S. longiusculus Mann., Kittery Point, Maine. 276 . MONOGRAPH OF THE LABOULBENIACEZ. This curious species occurs rather rarely, more often on the upper than on the inferior sur- face of the abdomen of the two hosts mentioned, the receptacle lying close against the surface, while the perithecia project outward, often backward. Specimens on SS. longiusculus are somewhat smaller and much paler than those on the larger host, the basal cell of the receptacle being more or less translucent. The form is not easily overlooked when mature, occurring usu- ally in groups of half-a-dozen or more. The hosts may be found in dry hay or other rubbish in cultivated land. ENARTHROMYCES nov. gen. Plates III-IV. Receptacle consisting of a simple series of superposed cells, the distal ones bearing sterile appendages, those below giving rise to antheridia or perithecia or remaining sterile. Anthe- ridia compound, terminally pointed and perforate, the six antheridial cells converging upward to a general cavity, into which they empty through narrow necks. Trichogyne simple, two- celled. Perithecia stalked, one or more formed by dire¢t budding from the cells of the receptacle. In the form and structure of its antheridium, and to some extent of its perithecium, this re- markable genus recalls Peyritschiella, to which it is undoubtedly more closely related than to any other. The series of specimens which constitute the types of the single species contain no very young individuals; yet it is apparent that, after the axis is formed and most of the sterile appendages have been produced, an antheridium arises, first originating as a lateral bud, in such a position that the first trichogyne when mature lies beside its apex. The antheridium is the largest thus far observed, of the compound type, and its general structure may be made out with no great difficulty. The antheridial cells are irregularly flask-shaped, and occupy the somewhat inflated basal portion of the organ, the distal part being more or less conical, and apparently originating from a single cell, the remains of a septum being usually visible (Plate III, fig. 19) just above the openings of the canals through which the contents of the antheridial cells are discharged. The cavity which it contains is of considerable size, and is often filled with free antherozoids. The latter are larger than other known forms of endogenous origin, and seem to be surrounded by a thin wall at the time of their final discharge. The trichogyne is quite peculiar, and invariable in form, being rather short and inflated, septate near the middle, and prolonged into a short terminal projection (Plate III, fig. 17, ¢7), which seems to be the only portion that is receptive, the remainder being distinctly thick walled. Indeed, it is only the terminal part of this projection that is very thin walled ; so that, after fertilization, when this receptive portion has withered, the trichogyne appears to end in a short, cylindrical, and distally truncate and perforate outgrowth (fig. 18, ¢7). The young perithecia arise as lateral buds from the cells of the axis, becoming cut off from it in the form of free clavate cells (fig. 13), which become once septate (fig. 14) and follow the course of development described in Part I (p. 228). ENARTHROMYCES INDICUS nov. sp. Plate III, figs. 13-19; Plate IV, figs. 8-11. Pale amber or straw colored. Perithecia large, sub-cylindrical, terminally rather abruptly and symmetrically rounded below the short, clearly distinguished, broadly truncate apex, the lip- MONOGRAPH OF THE LABOULBENIACEZ. 217 cells of which end in papillate enlargements, while one of them produces posteriorly a clavate outgrowth bent abruptly upward near its base and extending free above the apex. Below the base of this projection, and on the same side, the perithecium is prominently hunched. Basal cells of the perithecium often rounded and bulging. Receptacle consisting of from fifteen to twenty superposed cells, somewhat longer than broad, the two lowest always sterile, the third always producing a perithecium, the sixth, more commonly the seventh, producing an anthe- ridium; the cells immediately above the third producing perithecia or remaining sterile ; those immediately above the sixth or seventh producing perithecia or antheridia or remaining sterile, the total number of antheridia rarely exceeding three ; the perithecia rarely produced two from a single cell. The terminal cells of the receptacle, usually five or six in number, bearing septate, simple, lateral, sterile appendages. The latter straight, usually two borne on op- posite sides of a given cell, but sometimes three or four from the same cell, deeply blackened and somewhat constricted in the region of their basal septum, consisting more commonly of seven cells, the fourth and sixth, and notably the fifth, broader and shorter than the others, the terminal cell longer than the rest and tapering to a blunt tip. Similar appendages rarely produced even below the uppermost perithecium. Spores, very long and slender, 120-130 x 5m. Perithecia, 140-160 x 50-60 uw; the outgrowth, 85x10; the stalk-cell, 50-85 x 25-30 pu. Receptacle, 400-680 x 25-35 wu; average, 500 X 28 w. Appendages, 140-200 x 20 mw. On Pheropsophus sp., Booloo Valley, banks of the Beeas River, Northwest India. Sufficiently abundant material of this fine species was obtained from the inferior surface of the thorax and abdomen of an undetermined Pheropsophus from the above-mentioned locality in the collection of the Museum of Comparative Zoology. Although varying somewhat in the number of perithecia developed, it seems to be an unusually constant form. Even the sterile appendages, although they vary in number in different individuals, are remarkably uniform, and rarely vary in respect to the number of cells which compose them. In many cases the first, and sometimes others of the perithecia become aborted, as a result of the non-fertilization of their trichogynes ; and there may be three or even more such undeveloped perithecia on the same in- dividual, with from one to three or even four which have reached maturity (Plate IV, fig. 8), so that although there are more commonly not more than two perithecia in a given individual, there may be six to eight, developed and undeveloped, in exceptional cases. In their color, habit, and peculiarly blackened bases, the sterile appendages, curiously enough, recall those of the Laboul- benia (L. Pheropsophi), which infests a similar host in Africa and South America; so much so that a young specimen, which was the first examined, was for the moment mistaken for an abnormal condition of this species. Owing to the large size of the sexual organs, this form offers unusually good opportunities for a closer study of the sexual processes, and it is to be regretted that it should occur in so remote a locality. It seems doubtful whether the lower half of the two-celled body described as the trichogyne is not morphologically a portion of the trichophoric cell, since, though it is separated from the latter by a constriction, no septum is visible between them. In one instance, a second trichogyne was seen developed from the base of this basal half, the first trichogyne having failed to become fertilized. 278 MONOGRAPH OF THE LABOULBENIACEZ. PEYRITSCHIELLA Thaxter. Plate VI, figs. 7-24; Plate II, fig. 12. Proc. Am. Acad. Arts and Sci. Vol. XXIV, p. 8, 1890. Receptacle composed of four superposed areas: a basal consisting of a single cell; a sub- basal consisting of a single cell or of several cells transversely and asymmetrically disposed in a single row, one or more of which may be appendiculate ; a sub-terminal area consisting of a similar series of several cells appendiculate at either side, and, on one side only, producing the single antheridium ; and, lastly, a terminal series like the sub-terminal, but bearing one, rarely two, perithecia centrally placed; the cells of the series external to the perithecia, and the cell be- tween them, if there are two, appendiculate. Antheridium conical, compound, forming a sharp, tooth-like projection on one side. Appendages simple, continuous, separated from the cell which bears them by a prominently constricted, usually blackened septum. Spores once septate. Peri- thecia symmetrical or nearly so, the tips ending in four papille. This, together with the following and very nearly related genus Dichomyces, constitutes a peculiarly well marked type among the Laboulbeniace, and no other form possessing a com- pound antheridium, with the exception of Dimorphomyces, presents the same tendency towards the production of a bilaterally symmetrical receptacle, which is so perfectly developed in Dichomyces. In the present genus, there is always an irregularity in the form, and an asym- metry in the disposition of the cells, especially in the two lower transverse series; and though in P. geminata a greater degree of symmetry is attained than in any other species, the lower series is always asymmetrical, and in all cases only a single antheridium is ever found. The development of the receptacle is comparatively simple. The germinating spore, as in fig. 11, develops a terminal appendage which corresponds to the appendage lying at the left of the perithecium in figs. 8 and 10, or between the two perithecia in fig. 7. The larger, lower segment of the spore then divides into four cells, by the formation of three septa (fig. 12). The lower of these, the basal cell of the receptacle, undergoes no further modification, and in the single species P. curvata, the sub-basal cell remains unchanged. In all the other species, how- ever, the sub-basal cell and the two cells above it become divided in a characteristic fashion. This division is effected by the formation of oblique septa cutting off the two upper angles of the cell, as in fig. 13. The small triangular cells thus formed begin to grow obliqyely outward and upward, as seen at the right in fig. 138, and through the formation of further oblique divisions in these cells the characteristic transverse series result (fig. 14). The number and position of the appendages varies in different species and in different individuals, as indicated by the figures, and are quite peculiar in structure and appearance. They are simple and con- tinuous, or sometimes pseudoseptate, and always connected with the cell that bears them by a characteristically constricted, usually blackened septum. They vary in shape from a mere blad- der-like oval cell (fig. 20, Plate VI) to a more highly developed form, such as occurs in P. cur- vata (fig. 9). The appendages of the following genus, Dichomyces, are similar in all respects ; bat in no other case, with the exception of the species of Chitonomyces and its nearest allies do appendages occur having a similar appearance. A single antheridium only is produced on one side from the sub-terminal member of the superposed cell-series. Their structure corresponds very closely to that of Dimorphomyces, ex- MONOGRAPH OF THE LABOULBENIACE. 279 cept that there appear to be not more than four antheridial cells (Plate I, fig. 12), which dis- charge into a common cavity above. The latter forms the slightly inflated base of the terminal conical portion of the organ, the antherozoids, which are comparatively minute, making their way through a pore at its apex. The perithecia, except in P. geminata, are usually solitary, and arise normally from the dis- tal cell-series ; but in abnormal cases they may be produced externally from the sub-distal or even the lower series (P. geminata). In structure, the perithecia present no great peculiarities ; the cell-rows containing each four cells, the lip-cells being usually characteristically papillate, and symmetrical or nearly so. The ascogenic cell is commonly solitary (Plate 1, figs. 27-31); but in some instances there appear to be two. The trichogyne, in the few cases in which it has been observed, is very rudimentary, consisting of a vesicular prominence, sometimes showing a tendency to produce slight protuberances, as in the case of Camptomyces. The four known species all occur on terrestrial beetles belonging to the Carabide and Staphylinidx, and on account of their small size and appressed habit are usually detected with difficulty. In the following descriptions the side which bears the antheridia is spoken of as the anterior, the opposite as the posterior face of the individual. The antheridium may thus be described as occurring on the left side. PEYRITSCHIELLA CURVATA Thaxter. Plate II, fig. 12; Plate IV, figs. 11-18. Proc. Am. Acad. Arts and Sci. Vol. XXIV, p. 8. Perithecia hyaline, nearly straight, tapering to the symmetrically four-lobed apex, bent at an angle to the axis of the receptacle, the curvature of which it continues. Receptacle consisting of two superposed cells, followed by two transverse series of cells of variable numbers (four to seven) ; the terminal bearing one, rarely two, perithecia, with a group of sterile appendages on either side, the sub-terminal producing on one side, the sharply pointed antheridium with or without one or more adjacent sterile appendages; on the other side a group of from one to three similar appendages. Appendages simple, cylindrical, continuous, or pseudoseptate, hya- line, sometimes becoming tinged with brownish. Spores, 26 x 3.5-4y. Perithecia, 90-100 x 22-29 u. Appendages (longer), 604. Total length to tip of perithecium, 280-300 pu. On Platynus cincticollis Say. Vicinity of New Haven, Conn., of Cambridge, Mass., and of Kittery Point, Maine. The more common species of the genus occurring on the right shoulder of its host, and sometimes on the adjacent edge of the elytra. It is distinguished from the three remaining species by the absence of the lower transverse series of cells, the sub-basal cell of its receptacle remaining undivided at maturity. In very rare instances, two perithecia may be borne on a single individual; but I have seen but two cases of this kind among some dozens of specimens. The curved habit of the species is doubtless due to its position of growth upon the host, and is a character of little importance. The beetle on which it is found is common in the localities mentioned, along small brooks, where it may be found, concealed beneath stones, sticks, and rub- bish generally, in shady places. 280 MONOGRAPH OF THE LABOULBENIACE. PEYRITSCHIELLA GEMINATA Thaxter. Plate IV, figs. 7-8 and 24. Proc. Am. Acad. Arts and Sci. Vol. XXIX, p. 101. Hyaline. Receptacle asymmetrical, consisting of a single basal cell followed by three succes- sive, more or less definite, transverse rows of cells. The lowest of these rows is the most variable and irregular, consisting of from two to four cells, one of which, larger than the rest, is an axile cell continuing the basal cell directly, while the remaining one to three cells are cut off from it on one side, each successive cell smaller and placed higher in the series, the outermost and uppermost bearing one of the sterile appendages characteristic of the genus. The second row also consists of a larger median cell, which is free for a short distance on one side, and on both sides of which are cut off, as in the first row, from two to four cells, the smaller uppermost ones on both sides giving rise to from one to three appendages, according to the number of the cells. The upper row is either symmetrical or asymmetrical, according as it gives rise to two or to only one perithecium. In the first instance it consists of a median cell, above which are produced two sterile appendages, and a variable number of cells cut off laterally as in the lower rows, but nearly symmetrical in size and shape, the outermost bearing sterile appendages. If one perithe- cium only is produced, the row is asymmetrical, and a greater number of appendages appear on one side of the perithecium than on the other. Perithecia very slightly inflated near the base, tapering abruptly but slightly to the spreading apex, which is four-lobed, the lobes rounded, large and prominent. Spores about 37 x 3.7m. Perithecia, 756-80 x 18-22. Total length to tip of perithecia, 220-260 wu (150 u in specimens from the smaller host). On Pterostichus luctuosus De}j., P. erythropus Dej., and P. patruelis Dej., Kittery Point, Maine; Arlington, Mass. Unlike the other species, this form not infrequently produces two perithecia, the arrangement of its distal cells in such cases closely resembling that of Dichomyeces. The main body of the receptacle is, however, asymmetrical, and the antheridium occurs only on one side. It is distinguished from the preceding species by its more slender form, smaller and shorter appendages and perithecia, and especially by the division of the sub-basal cell of its receptacle to form an incomplete transverse series, or rather half-series, of cells. Its antheridium is small and inconspicuous, occurring on one side as in the other species. It may be found growing on the inferior surface of the prothorax near its anterior margin, and on the adjacent posterior margin of the prothorax, as well as on the anterior legs. In the last position the individuals be- come very considerably distorted in form, and in such specimens several instances have been observed in which perithecia were developed from the lower and middle transverse series of cells on the right side (opposite the antheridium). The species is a rare one, though the hosts which it infests are not uncommon under stones or in rubbish near the margins of ponds or streams, PE£YRITSCHIELLA MINIMA Thaxter. Plate VI, figs. 19-21. Proc. Am. Acad. Arts and Sci. Vol. XXIV, p. 266. Hyaline or slightly yellowish. Perithecia somewhat asymmetrical, large, stout, tapering rather abruptly to the blunt apex, which terminates in four not very prominent, nearly symmetri- MONOGRAPH OF THE LABOULBENIACE. 281 cal papilla. Receptacle stout, sub-triangular in form, consisting of a rather small, often slender, basal cell, followed by the usual three transverse series of cells. The lower of these consists of the median or primary cell, on one side of which a single small cell is cut off, while on the opposite side a series of four or more cells, mostly appendiculate, extend obliquely upward and outward, overlapping one another to some extent. The middle series consists of a large median cell, on either side of which numerous (four or more) appendiculate cells extend obliquely upward ,and outward, and produce, on one side, the large pointed antheridium, which does not usually project beyond the margin of the receptacle. The distal series, like the sub-distal, composed of fewer cells and bearing the single perithecium. Appendages oval or short cylin- drical, becoming brownish; their basal septa large, but not conspicuously blackened. Spores, 37-40 x 4. Perithecia, 100 x33. Appendages (longest), 383. Total length to tip of perithecium, 190-220 4, Receptacle, 90-110 x 50-58 yw. On Platynus cincticollis Say. Vicinity of New Haven, Conn.; Weston, Mass.; York, Maine. A rare species occurring near the extremity of the anterior legs of its host, where, owing to its compact form and small size, it is easily overlooked. It is very distinct from the two pre- ceding species, and peculiar from its sub-triangular form, from the presence of cells cut off on both sides of the primary or median cell of the lower transverse series, and from the numerous short, inflated, bladder-like appendages, The latter are thickly clustered on the left margin of the receptacle, while the right presents an unbroken line from the foot nearly to its summit. The habitat of the host has been already mentioned under P. curvata, and both species have been found on a single individual. PEYRITSCHIELLA NIGRESCENS Thaxter. Plate VI, figs. 22, 23. Proc. Am. Acad. Arts and Sci. Vol. XXVIII, p. 184. Receptacle sub-triangular in form, consisting of a single basal nearly hyaline cell followed by the usual three series, the lower broadly edged with black on either side, consisting of a long median cell from which three cells are cut off on the right and two on the left; the middle series of about ten cells, which form slight blunt appendiculate projections on either side, extending much higher on the left than at the right, where the single antheridium is situated; the upper series shorter than the middle, bearing the single stout, asymmetrical perithecium, on either side of which are several short appendages asymmetrically placed. Perithecia, 65 x 19 uw. Receptacle, 70 x 37 w. Total length, 130 pu. On Philonthus debilis Grav. Waverly, Mass. The single type from which the above description is drawn was found on one of the posterior legs of its host, the figures given representing an anterior and posterior view of the same speci- men. The compact sub-triangular form of the receptacle and the stout perithecium connect it more closely with P. minima than with either of the other species ; while the external blackening of its lower cell series suggests some of the species of Dichomyces. In its asymmetrical development and single antheridium, however, it corresponds exactly to the type structure of Peyritschiella, The perithecium of the unique type is not quite mature, and more fully developed material may show that its apex becomes characteristically papillate as in the other species. 282 MONOGRAPH OF THE LABOULBENIACES. DICHOMYCES Thaxter. Plate VI, figs. 25-36; Plate VIII, figs. 11-14. Proc. Am. Acad. Arts and Sci. Vol. XXVIII, p. 183. Receptacle flattened, sub-triangular ; consisting of a single basal cell followed by three trans- verse series of cells symmetrically placed ; the distal row bearing a pair of perithecia symmetri- eally placed; or a single perithecium, the fellow of which is undeveloped, together with certain sterile appendages; the sub-distal row producing a pair of compound antheridia symmetrically placed and subtended by one or more sterile appendages. Perithecia symmetrical. Spores once septate. Appendages and antheridia as in Peyritschiella. This genus, although very closely related to Peyritschiella, is distinguished by the bilateral symmetry which characterizes not only the arrangement of the cells of the receptacle itself, but of the organs produced from it, namely, the antheridia and perithecia; although in a single species (D. inaequalis) one perithecium only is developed, which is, however, placed symmetri- cally in relation to the cell from which, in the other species, a second perithecium arises. Apart from this bilateral symmetry, the structure and development of the members of the two genera are identical, and were it not for the fact that of the eight species included by them, four correspond exactly to one, and four to the other type, the two genera might readily be united. The four species of Dichomyces are among the most striking and curious in form of all the Laboulbeniacew. They have been found thus far only on beetles belonging to the Staphylinide, such as inhabit rather dryer situations, in most cases common among rubbish in cultivated fields. In the following descriptions the face bearing the antheridia is spoken of as anterior. DICHOMYCES FURCIFERUS Thaxter. Plate VI, figs. 25-29. ‘Proc. Am. Acad. Arts and Sci. Vol. XXVIII, p. 184. Receptacle consisting of a small basal cell, which is nearly hyaline distally and suffused with _ brown basally, the receptacle above gradually expanding into a more or less fan-like form, the basal portion wholly black and opaque, the blackened area extending upward externally and including a prong-like projection with one or two appendages at its base, which extends above | the base of the perithecium and forms the terminal portion of the sub-distal transverse series of cells on either side; the latter, seven in number, becoming generally suffused with blackish brown, the long rectangular median cell usually more or less distinct, the remainder partly or wholly opaque and indistinguishable; the antheridia lighter brownish. The distal row of cells seven in number, their septa straight, thin, and clearly defined; the middle cell of the series bearing distally two appendages, placed antero-posteriorly, the terminal ‘cells of the series on either side also bearing two appendages, the inner slightly anterior. Perithecia bent slightly forward, tapering very slightly to the blunt apex, which bears a short, recurved, tooth-like pro- jection on either side from each anterior lip-cell. The perithecia and distal row of cells faintly tinged with reddish purple. Appendages short, simple, hyaline, cylindrical, with a constricted blackish base. Perithecia, 63 x 16-184. Receptacle: length to base of perithecia about 904; to tips of external projections, 100-120 uw; greatest width, 55-60 pu. On Philonthus debilis Gray., Waverly, Mass., and Kittery Point, Maine. MONOGRAPil CF THE LABOULBENIACE. 283 A most singular plant looking like a two-pronged fork or a pair of inverted pincers, between the black arms of which arise the two perithecia. The lower portion of the receptacle is so opaque that the cell structure is indistinguishable; but, as far as it can be made out, the lower trans- verse series of cells consists of a long median cell, on either side of which two, perhaps three, cells are cut off in a fashion similar to that represented in the figure of D. princeps (Plate VIII, figs. 11-12). In the present, as well as in the remaining species, the appendages vary somewhat in number according to the number of terminal cells in the distal and sub-distal series, which is not quite constant; and, in the majority of individuals, but few of the appendages remain unbroken at maturity. Of the two appendages situated between the perithecia, one represents the original terminal appendage of the germinating spore. The general form of the receptacle is quite characteristic, and in the region of the distal series of cells it is distinctly concave on the anterior side, the posterior face being somewhat convex. From the fact that its host is dark-colored, while the parasite itself is partly blackened, it is often very difficult to detect, lying, as it generally does, appressed against the surface of the former. It sometimes occurs, however, in such large numbers that it is seen without any diffi- culty, and in such cases, although it is usually found on the under surface of the abdomen, it may extend to the legs, thorax and all parts of the body of its host. Its natural position seems to be with the tip of the perithecium turned toward the extremity of the insect’s body, and the anterior face turned inward, the perithecia being turned slightly backward (upward). Although the host is very common, this species, like its relatives in the genus, seems to be decidedly rare. DICHOMYCES INZQUALIS Thaxter. Plate VI, figs. 30-34. Proce. Am. Acad. Arts and Sci. Vol. X XIX, p. 103. Receptacle as in D. furciferus, its fork-like projections prominent and indistinctly septate, the distal row of cells bearing a single perithecium ; otherwise symmetrical, except that the sub- median cell, above which a second perithecium arises in D. furciferus, is much reduced in size. Appendages ten to twelve, one external to and near the base of each antheridium, two above the median cell of the distal row, and three to four borne one from each of the three to four cells of the distal row external to the sub-median cells, all arising as in D. furciferus. Perithecium large, slightly inflated towards the base ; or sub-cylindrical, tapering abruptly at the extremity to a sub- truncate apex destitute of appendages. Spores, 26 x 3.54. Perithecia, 100 x 25. Receptacle, length to base of perithecium, 924; length to tips of lateral forks, 110-130; greatest breadth, 50-60 w. Total length to tip of perithecium, 180-190 pu. On Philonthus debilis Grav. Kittery Point, Maine, and Waverly, Mass. This species occurs, sometimes in company with D. furciferus, on the abdomen, more rarely on the legs and thorax of its host. It is at once distinguished by its solitary perithecium, which is destitute of the terminal outgrowths peculiar to the last named species. It is much more common in the localities mentioned than its ally, and in none of the numerous specimens examined does there appear to be any variation in the points of structure which dis- tinguish it. In no instance was any attempt observed to produce the usual pair of perithecia, and the single perithecium, by its larger size, somewhat different form, and abruptly truncate apex, serves readily to distinguish it. 284 MONOGRAPH OF THE LABOULBENIACE. DicHOMYCES INFECTUS Thaxter. Plate VI, figs. 35-36. Proc. Am. Acad. Arts and Sci. Vol. X XIX, p. 102. Receptacle consisting of a short basal cell succeeded by an axile cell placed vertically, on either side of which a series of three obliquely superposed cells forms a blackened border. The remainder of the receptacle consists of two successive transverse symmetrical rows of cells, the lower row made up of three central and several smaller external cells, terminating on either side in a short, blunt projection below the prominent antheridia; beside which arise externally single sterile appendages. The distal row is composed of seven cells, the external cells on either side not extending beyond the bases of the perithecia and destitute of appendages ; two appendages arise between the perithecia, one on either side. Perithecia two, closely approxi- mated, arising from single broad, flattened cells, short and stout, tapering slightly toward the sub-truncate apex, which is destitute of papilla or appendages. Perithecia, 66 x 22. Recep- tacle, 60 x 40 pw. On Xantholinus obsidianus Melsh. Waverly, Mass. Two specimens of this small form were found at the tip of the abdomen of its host, only one of which is preserved as the unique type of this well-marked species. Its broad, stout perithecia, and the absence of any external appendages on the distal series of cells, distinguish it readily from other species. Although sought for diligently, it has been found but once, notwithstanding the fact that its host is a beetle common in cultivated fields, where it may be found running over freshly turned soil or concealed in decaying vegetable matter. DicHomyces princers Thaxter. Plate VIII, figs. 11-14; Plate II, fig. 11. Proc. Am. Acad. Arts and Sci. Vol. XXX, p. 479. Nearly hyaline, becoming slightly and uniformly tinged with pale reddish brown, sometimes narrowly edged with blackish near the base. Receptacle large, consisting of a single small squarish basal cell, above which are three successive transverse rows of cells placed side by side, the upper margin of each series convex ; the lower series consisting of a long narrow axial cell, with three or four more or less obliquely superposed cells on either side; the middle series consisting also of an axial cell, with five to eight cells on either side, which extend obliquely upward and outward to form a free rounded projection, each cell of which bears a short append- age, the antheridia prominent at the base of each projection; the third or distal transverse series like the second, the cells often slightly more numerous, forming projections in a similar fashion on either side which bear the same short appendages. The axial cell of the terminal series is followed by two small cells, each bearing a short appendage, on either side of which a large somewhat flattened cell forms the base of the perithecium. Perithecia two, more or less divergent, elongate, slightly inflated and tapering rather gradually to the blunt apex. Perithecia, 110-165 x 22-30 pw. Spores, 38 x4 y. Receptacle, 150-180 x 70-75 pu. On Philonthus sordidus Grav. Waverly, Mass. This fine species is by far the largest and most conspicuous member of the genus, and, owing to the absence of any deeply blackened portions, illustrates the typical form more clearly MONOGRAPH OF THE LABOULBENIACEZ. 285 than any of its allies. I have met with it on one occasion only, when several hosts were collected from a heap of rubbish near cultivated land, on which the parasite occurred in great numbers, thickly covering the lower surface of the abdomen and extending to the legs and thorax. Unfortunately, all but one of these hosts made their escape in transit, but the single one remaining has furnished abundant material of the mature form. Owing to its light color, large size, and more than usually divergent, though not erect, habit of growth, it is not readily overlooked on the black abdomen of its host. CHITONOMYCES Peyritscu, Plates VIII and XXVI. Sitz. d. k. Acad. d. Wiss. LX VIII, p. 250 (1873) ; Hetmaromycers Peynitscn, 1. c, p. 251; Hermaromyces EMEND., Thaxter. P Proc. Am. Acad. Arts and Sci. Vol. XX VII, p. 30. Receptacle consisting of a basal and a terminal portion; the former consisting of two superposed cells, and three or four upper smaller cells which form the base of the perithecium ; the latter consisting of four cells lying beside the perithecium, the terminal cell always free, originally more or less dome- or bell-shaped, bearing a single terminal appendage, and sometimes becoming otherwise modified at maturity ; the sub-terminal cell connected on its inner side with, or rarely free from, the two cells below it, from the upper of which is separated a small cell that bears terminally, in the angle formed by the perithecium and the receptacle, one or two appendages, beside which is situated the usually small antheridium (?). Appendages slender, filamentous, simple, aseptate or spuriously septate, hyaline, evanescent, the base blackened and slightly constricted. Perithecium more or less completely united to the distal portion of the receptacle, each series of wall-cells containing not more than six cells, some of which may be appendiculate ; the apex often variously modified. Spores fusiform, once septate. Since the description of numerous species under the name Heimatomyces, I have had an opportunity of examining specimens of Chitonomyces melanurus Peyritsch, and, as I formerly suggested, the two genera prove identical. I have therefore concluded to use the latter name to distinguish the genus ; since, although I very much dislike to regard the rules of priority at the expense of those of common sense, it seems desirable that the nomenclature adopted in the present monograph should be, in so far as possible, a fixed one, and the name Chitonomyces undoubtedly has precedence in the present instance to the extent of nearly half a page. The genus is one the position of which has been, until the present paper was in press, quite uncertain ; the character of the sexual organs not having been ascertained with sufficient exact- ness to warrant any definite statement concerning them. The very recent discovery, however, of an undescribed and very closely allied aquatic genus, in which the antheridium is so placed that it can be readily seen, renders its position no longer doubtful, and confirms my first impres- sion, which was based on the general resemblance of its appendages and the structure of its receptacle to those of Peyritschiella. The new genus, a description of which is necessarily reserved for a succeeding supplement, is characterized by a general structure curiously inter- mediate between that of Chitonomyces and the genus last mentioned, the small but very definite compound antheridium occupying a position on the anterior margin of the plant just below the base of the perithecium. 286 MONOGRAPH OF THE LABOULBENIACE. In the present genus, the antheridium, or what I have taken to be that organ, lies just beside the bases of the lower appendages, and has not been satisfactorily made out, owing to the small size and lack of definition in the cells lying in this region. In C. rhyncostoma, what appears to be the neck of the antheridium becomes enlarged, as the individual matures, and projects as a conspicuous hook-like prominence from the angles between the perithecium and the appendage (Plate XXVI, fig. 18). The character of the trichogyne is also a matter of uncertainty, and although I have definitely made out the carpogenic and trichophorie cells, the latter terminating close beside the supposed antheridium, I have been quite unable to determine whether one of the two “appendages” was in reality a trichogyne or whether, as is more probable, the latter is reduced merely to a slightly inflated prominence. Owing to the minute size of a majority of the species, and the difficulty in obtaining material of young individuals, the determination of these matters is by no means easy, and involves an expenditure of time that I have been unable to afford. In other respects the structure of the members of the genus is well defined and remarkably constant ; although the perithecium is subject to curious variations of form, and the terminal portion of the receptacle, especially its terminal cell, is often so modified as to obscure its true structure. In C. melanurus, for example, the terminal and the sub-terminal cells are, at maturity, opaque and indistinguishable, the former becoming proliferous below its original apex and developing a hook-like extremity, the insertion of the originally terminal appendage being turned to one side, and visible only as a slight prominence from its inner margin (Plate XXVI, fig. 19). A somewhat similar modification is seen in C. marginatus, in which the whole distal portion of the receptacle becomes blackened, and the terminal cell is similarly proliferous (Plate XXVI, figs. 20 and 21; Plate VIII, fig. 27). The curious outgrowths from the wall-cells of the perithecium are remarkable from the fact that they are not confined to any special cell or even cell-series, being quite variable in position, as in C. paradoxus, C. appendiculatus, C. spinigerus, C. uncinatus and others. The appendages closely resemble those of Peyritschiella, and are very slender and evanescent, being usually completely broken off in mature specimens, the blackened bases alone remaining. The more or less dome-like cell which bears the terminal appendage is identical with the similar cell which bears this appendage in the last-mentioned genus. Reference has already been made to the form of the foot, which is peculiar to this and the succeeding genus, and seems undoubtedly designed to allow a greater freedom of motion necessitated by the exigencies of life in the water on a rapidly moving host. The base of the foot, by which alone it is attached to the host, is rather narrow and pointed, while the upper por- tion is more or less rounded, and serves as a fulerum on which the individual may roll from side to side through a considerable arc. The remarkable constancy with which the different species occur in definite positions has also been previously alluded to (p. 246); and, as will be seen, is practically invariable in every instance. All the species are aquatic, and are, so far as known, confined to hosts belonging to the family Dytiscide. As in previous instances, the side bearing the perithecium is considered anterior, the lower appendages are thus described as arising from the left side. It may be remarked in passing that species credited to Connecticut only are undoubtedly as widely distributed as the others, this being the only locality in which they have been carefully sought for. MONOGRAPH OF THE LABOULBENIACE. 287 CHITONOMYCES PARADOXUS (Peyritsch). Plate VIII, figs. 17-21. HrmMatomycres PARADOXUS Peyritsch, Sitz. d. k. Acad. d. Wissen. Bd. LX VII, p. 251; Wien, 1873, Plate ILI, figs. 35-39 ; Winter, Die Pilze Deutschl. II, p. 525, p. 920, fig. 5; Sorokin, Veg. Par. of Man, ete., p. 425, Plate XXXII, figs. 763, 770; Saccardo Sylloge, Vol. VIII, p. 915; Berlese, Malpighia, Vol. III, p.60; Thaxter, Proc. Am. Acad. Arts and Sci. Vol. XX VII, p. 32. Pale straw color becoming tinged with amber brown. Perithecium asymmetrical, the irregu- lar lip-cells lateral and internal, the sub-terminal wall-cells of one of the external series produced to form a pointed terminally curved outward horn-like projection; a prominent ridge extending transversely on the left side from about the centre of the perithecium across the sub- terminal cell of the receptacle. Receptacle of the usual form, the basal cell large, the sub-basal cell flattened, the terminal and sub-terminal cells projecting almost wholly free beside the peri- thecium. Spores strongly curved, fusiform, 65 x Ty. Total length to tip of perithecial horn about 225 w; to tip of receptacle, 150-185 uw. Greatest width, 60-75 yp. On Laccophilus maculosus Germ., New England; on Laccophilus sp., Kansas (Barber); on LL. minutus Sturm., and L. hyalinus Dej., Europe (Peyritsch). This species, which is the largest of the genus, occurs invariably on the outer margin of the left elytron, somewhat beyond the middle, where it grows in groups of from two to half-a-dozen, and is very easily scen from its large size. It is not uncommon, and is not to be confused with any other species from the terminal, horn-like outgrowth of its perithecium. 2 CHITONOMYCES APPENDICULATUS Thaxter. Plate XXVI, figs. 25-26. HeEIMATOMYCES APPENDICULATUS: Proc. Am. Acad. Arts and Sci. Vol. XXVII, p. 31. Becoming faintly brownish. Perithecium tapering to a rather sharp apex, curved strongly outward, hunched externally and bearing a straight, sub-clavate, one-celled, brownish appendage arising externally some distance below the apex, projecting outward and upward at an angle of about 45°. Basal cell of the receptacle rather narrow, twice as long as the squarish sub-basal cell. Spores, 32X83 y. Perithecia, 55 x 15m to 75 x22. Basal and sub-basal cells, 30-45 uw in length. Perithecial appendage, 30-33 x4 y. Total length to tip of perithecium, 100-130 yp. Greatest breadth, 30-36 w. On Laccophilus maculosus Germ., Connecticut. A rare species, confined to the anterior pair of legs of its host, and distinguished at once by its clavate perithecial appendage, which corresponds to the similar horn-like projection from the perithecinm of C. paradoxus, but is very different in form and in its relation to the other parts of the perithecium. CHITONOMYCES DIsToRTUS Thaxter. Plate XXVI, figs. 29-30. HeEIMATOMYCES DisTORTUS: Proc. Am. Acad. Arts and Sci. Vol. XXX, p. 477. Pale yellowish, more or less clavate in general form. Perithecium inflated, its external mar- gin strongly curved, becoming abruptly constricted below a long, slender, tubular, terminal 288 MONOGRAPH OF THE LABOULBENIACEZ. mouth, which is usually, but not always, bent abruptly outward almost at right angles to the nearly straight inner margin of the perithecium. A short, straight, bluntly-pointed, rather stout appendage arises on one side only of the perithecium, just below this tubular apex beyond which it projects. The basal and sub-basal cells of the receptacle about equal in length, the lat- ter broader: distal portion of the receptacle composed of the usual four cells, the sub-termi- nal cell forming a distinct external prominence below the terminal cell, which is bent toward and partly overlaps the perithecium. Perithecium (main body), 60 x 18 y, its tubular apex, 18-25 x 6. Spores, 20x 3y. Length of receptacle, 110 pu. On Laccophilus maculosus Germ., Connecticut. A singular Species appearing at first sight malformed or abnormal. It occurs in company with C. appendiculatus on the anterior legs of its host, and is at once recognized by the tubular prolongation of the apex of its perithecium. CHITONOMYCES SPINIGERUS Thaxter. Plate VIII, figs. 15-16. HEIMATOMYCES SPINIGERUS: Proc. Am. Acad. Arts and Sci. Vol. XXX, p. 478. Brownish yellow. Perithecium small, its tip slightly exceeding that of the receptacle ; its extremity blunt, lobed, curved outward, and bearing two projections unequal in size just below the tip, one of which extends outward beyond the perithecial margin as a blunt prominence. Basal cell of the receptacle often bent, expanding distally, much longer than the flattened sub- basal cell: the distal portion of the receptacle with greatly thickened external walls, and consisting of the usual four cells, the terminal one short, with a broad base and bent towards the apex of the perithecium. Three small cells are distinct below the perithecial cavity, from the outer of which is produced externally a prominent spur-like process. Perithecia, 55 x 15 pw. Total length to tip of perithecium, 88-90. Spur-like process, 12-30 w long. On Laccophilus maculosus Germ., Connecticut. Distinguished from all other species by the spur-like process from the base of the peri- thecium. The septa are all defined with unusual clearness, the externa! walls being greatly thickened. Apparently among the rarest of the twelve species inhabiting this host. CHITONOMYCES UNCIGERUS Thaxter. Plate XXVI, figs. 27-28. HEIMATOMYCES UNCIGERUS: Proc. Am. Acad Arts and Sci. Vol. XXX, p. 478. Pale yellowish. Perithecium moderate, rather broad, its outer edge straight, its upper fourth free from the receptacle, its prominent bluntly-tipped extremity bent abruptly outward at right angles: a slender hooked appendage arises from a point close to the receptacle about two-thirds of the distance from the base to the apex of the perithecium, projecting from it obliquely out- ward. Basal cell of the receptacle large and long, the sub-basal cell small, sub-rectangular, flattened: the distal portion composed of the usual four cells, the terminal one not very promi- nent, and bent strongly toward the perithecium. Perithecia, 80 x 22-25 u. Spores, 45 x 4u. Perithecial appendage about 22m long. Receptacle, 132 long. Total length to tip of perithecium, 135 p. MONOGRAPH OF THE LABOULBENIACE. 289 On Laccophilus maculosus Germ., Connecticut. The more or less wedge-shaped apex of the perithecium of this very distinct form projects outward abruptly at right angles to the straight outer perithecial margin. The hook-like appendage is quite unlike that of any other species in form and position, and, occurring only on one side, is not seen unless the perithecium lies at the right. It occurs with C. spinigerus, C. hyalinus, and rarely C. marginatus, on the posterior legs of its host. CHITONOMYCES MELANURUS Peyritsch. Plate XXVI, fig. 19. Sitz. d’k. Acad. d. Wissen., Bd. LX VII, p. 251, Plate III, figs. 30-34; Wien, 1873; Winter Die Pilze Deutschl., Bd. II, p. 924, p. 920, fig. 4; Sorokin, Veg. Par. of Man, etc. Vol, I, p. 424, Plate XXXII, fig. 771; Saccardo Sylloge, Vol. VIL, p. 914; Berlese, Malpighia, Vol. ILI, p. 59. Straw colored, becoming faintly brownish. Perithecium slightly inflated near its base, its sub-terminal wall-cells expanded to form a projection from the outer and inner margin, each of which extends a little beyond the small hyaline apex which lies between them; the outer projec- tion smaller and distinctly brownish, the inner nearly twice as broad and hyaline. The two lower cells of the receptacle nearly equal, forming a rather slender basal portion, above which it expands somewhat abruptly: the terminal and sub-terminal cells deeply blackened, the latter proliferous externally below the insertion of the appendage, which is thus turned inward, and becomes lateral in position, the proliferation extending some distance beyond it, and forming a terminal, translucent, outwardly turned hook. Length of perithecium, 95-100. Total length to tip of perithecium about 150-160; to tip of receptacle, 145-150. Greatest width, 30-35 pm. On Laccophilus minutus Sturm. and L. hyalinus Dej., Europe. This peculiar species, which is the type of the genus, does not appear to occur in this country, its place being taken curiously enough by the succeeding species, which occurs in exactly the same position often in company with C. paradozxus, and exhibits a somewhat similar proliferation of the terminal cell of the receptacle. The materialfrom which the accompanying figure and description were derived was found on a specimen of Laccophilus from Germany in the collection of the Museum at Cambridge. Although Peyritsch placed it in a genus distinct from C. paradorus, there is no doubt whatever as to the generic identity of the two forms. CHITONOMYCES MARGINATUS Thaxter. Plate XXVI, figs. 20-22; Plate VII, figs. 25-27. HEIMATOMYCES MARGINATUS: Proc. Am. Acad. Arts and Sci. Vol. XXVII, p. 34. Long and slender, at first nearly hyaline, then yellowish. Perithecium straight, then suddenly bent inward below the hyaline, neck-like, strongly curved tip. Basal cells of the receptacle sub- triangular, the sub-basal half as large as the basal, the three cells at the base of the perithecium more than usually developed: the terminal cells all becoming black and opaque at maturity ; the terminal one continued by a squarish outgrowth basally hyaline, at first lateral and external, becoming terminal (the true apex of the cell being pushed inward and becoming lateral), hardly exceeding the tip of the perithecium, which it conceals. Spores, 30x 3m. Perithecia, 95-110 x 22. Total length to tip of perithecium, 140-1604. Basal cells of receptacle, 25 yp. 19 290 MONOGRAPH OF THE LABOULBENIACEZ. On Laccophilus maculosus and Hydroporus spurius, Connecticut, Maine; Laccophilus sp., Kansas, Barber. A form peculiar for the modification of the terminal cells of the receptacle, which makes the perithecium appear as if bordered by a black band. It is found in company with C. paradorus, and recalls in some respects the preceding species, which is similarly associated on the left elytron of its host. In one instance a few specimens were found on the left posterior leg, but such a position is exceptional. CHITONOMYCES RHYNCOSTOMA Thaxter. Plate XXVI, figs. 17-18. HEIMATOMYCES RHYNCOSTOMA: Proc. Am. Acad, Arts and Sci. Vol. XXVII, p. 33. Evenly suffused with dull amber-brown. Perithecium relatively large, the broad extremity abruptly hooked inward, so that the papillate apex becomes lateral and internal; the external series of wall-cells prominent with conspicuous septa. Basal cell of the receptacle rather short, somewhat inflated ; the sub-basal cell much flattened; the terminal and sub-terminal cells about equal in length, forming together an outwardly-curved, finger-like projection, exceeding the peri- thecium: a short but rather conspicuous hook-like structure (antheridium?) arises from the angle between the perithecium and the receptacle on the left side. Spores,26 x3. Perithecia, 75 X 22. Total length to tip of perithecium, 100 4. Basal portion of receptacle, 25-30 u long. On Laccophilus maculosus Germ., and Hydroporus spurius Lec., Connecticut. This species occurs rather rarely on the margin of the right elytron in a position correspond- ing to that occupied by C. paradoxus, but nearer to the middle. The broad hooked apex of the perithecium, coupled with the finger-like development of the distal portion of the receptacle, serve at once to distinguish it from known species. The hook-like projection, which arises near the base of the lower appendages, has already been alluded to in the previous description of the genus, and may be the antheridium. CHITONOMYCES LICHANOPHORUS Thaxter. Plate XXVI, figs. 15-16. HeEIMATOMYCES LICHANOPHORUS: Proc. Am. Acad. Arts and Sci. Vol. XX VII, p. 32. Hyaline except for the suffused basal cell. Perithecium bent outward at an angle from the basal part of the receptacle, tapering slightly to the papillate apex. Basal cell of receptacle enlarged and greatly elongated, more or less intensely blackened above its hyaline base ; sub- basal cell flat and almost obsolete. Terminal and sub-terminal cells together forming a straight, finger-like projection as long as or longer than the perithecium, on which the appendage is terminal. Spores, 33-37 x 2.5-3 yw. Perithecia, 65-90 x 30 w. Total length to tip of perithe- cium, 150-180 uw. Basal cell, 90-110 p. On Laccophilus maculosus Germ., Connecticut. This species is confined to the median inferior anal plate of its host, and has only been observed upon males. It is not to be confused with any other species, being distinguished by its elongated basal and apical cells, almost black and white color, and papillate divergent perithe- cium. C!. rhyncostoma, which has a very similar finger-like prolongation of the terminal portion of its receptacle, is at once separable by the basal cell, and the form of its perithecium. MONOGRAPH OF THE LABOULBENIACE. 291 CHITONOMYCES UNCINATUS Thaxter. Plate XX VI, fig. 25-24. HeimaTomycers uncinatus: Proc. Am. Acad, Arts and Sci. Vol. XX VII, p. 33. Evenly suffused with pale amber-brown, Perithecium large, curving evenly inward to the somewhat pointed apex. Basal cells of the receptacle rather slender, the terminal cell pushed to one side and bent past the apex of the perithecium by a somewhat indurated, blunt outgrowth from the sub-terminal cell, the tip of which it nearly equals. Perithecia, 75-85 x 20 u. Total length, 110-130 yw. Basal cells of receptacle, 37 pw. On Laccophilus maculosus Germ., and Hydroporus spurius Lec., Connecticut. A rather rare species, occurring in groups on the inferior surface of the abdomen of its host, and distinguished by the peculiar development of the sub-terminal cell of the receptacle, which gives its extremity a somewhat hooked appearance. CHITONOMYCES AFFINIS Thaxter. Plate XXVI, figs. 12-14; Plate VIII, figs. 29-30. HEIMATOMYCES AFFINIS: Proc. Am. Acad. Arts and Sci. Vol. XX VII, p. 31. Rather strongly suffused with amber-brown. Perithecium commonly slightly curved inward, or nearly straight, the tip often slightly bent outward. Basal cell of receptacle large, sub-tri- angular, suffused laterally and terminally with deep black-brown: sub-basal cell very flat ; terminal cell small, its axis bent strongly inward. Spores, 50-55 x 3m. Perithecia, 100-110 x 30 4. Basal and sub-basal cells of receptacle, 40-45. Total length to tip of perithecium, 150-170 yp. On Laccophilus maculosus Germ., and Hydroporus sp., Connecticut. This species occurs near the margin of the right elytron toward its tip, and is distinguished from other species of the more simple forms by its large, blackened basal cell and the position of the terminal cell of the receptacle, which is bent against the perithecium on the left side, and ° is partly concealed by it when viewed from the right. It is one of the commonest species, and is easily detected from its dark color. CHITONOMYCES HYALINUS Thaxter. Plate XXVI, figs. 10-11. HEIMATOMYCES HYALINUS: Proc. Am. Acad. Arts and Sci. Vol. XXVII, p. 31. Hyaline or very faintly tinged with yellowish brown. Perithecium large; at first hunched externally and bent inward near the apex; at maturity becoming nearly straight, tapering slightly to the rather narrow apex, which is bent somewhat abruptly outward; one or two oblique ridges more or less distinct below the apex on the inner margin, the walls of the cells composing the outer margin greatly and often very abruptly thickened below the tip. Basal portion of the receptacle rather short and stout, the two lower cells nearly equal. Spores, 30 x 2.5. Perithecia, 75-90 x 204. Total length to tip of perithecium, 110-120 w. On Laccophilus maculosus Germ., Connecticut, This species, although it possesses few striking characteristics, is yet quite distinct from any of the others, its chief peculiarity consisting in the fact that the walls of the cells composing the 292 MONOGRAPH OF THE LABOULBENIACEZ. outer margin become abruptly thickened below the tip. The oblique ridges, which are often very distinct below the apex on the inner side, and the abrupt outward curvature of the latter at maturity serve also to separate it from other known species, It is rather rare and detected with some difficulty. CHITONOMYCES SIMPLEX Thaxter. Plate X XVI, figs. 1-3. HEIMATOMYCES SIMPLEX: Proc. Am. Acad. Arts and Sci. Vol. XX VII, p. 30. Pale yellowish or faintly brownish. Perithecium rather slender, continuing the strong curve of the receptacle evenly outward to its rather coarse blunt straight tip. Basal cell of the recep- tacle much longer than the flattened sub-basal cell: terminal cell bell-shaped, small, the whole distal portion of the receptacle evenly curved parallel to the perithecium. Spores, 26 x 11 p. Perithecia, 55-60 x 11-12 4. Length of receptacle, 75. Basal cell, 15 x 7.5 u. Total length, 90-100 pu. On Laccophilus maculosus Germ., Hydroporus spurius Lec., Connecticut. A common species, occurring in considerable numbers on the elytra of the host, near the middle of the distal portion, possessing the simple typical structure of the genus, and dis- tinguished by its evenly curved habit, and the prominent blunt almost cylindrical terminal por- tion of the perithecium. CuitonoMyces Bipessarius Thaxter. Plate XXVI, figs. 4-5. Heimatomyces Brpessarius: Proc. Am. Acad, Arts and Sci. Vol. XXVIII, p. 185. Hyaline, becoming faintly tinged with blackish. Perithecium small, stout, its upper third or fifth free from the receptacle, the apex bent outward, the basal portion straight, the tip broad, with large, prominent lips. Receptacle stout, the two basal cells more nearly equal, the two cells above these longer than broad and nearly equal; distal portion nearly as in @. borealis, the base of the short terminal cell horizontal. Perithecia, 40-48 x 15. Receptacle, 65 » long. Total length to tip of perithecium, 80 y. On Bidessus granarius Aube, Kittery Point, Maine. This is among the smallest of all the Laboulbeniacex, and though presenting no great pecu- liarities, is distinguished by its stout form as well as the short and very broad but not abruptly distinguished tip. It occurs in various positions on the upper surface of the elytra of its minute host. CHITONOMYCES BOREALIS Thaxter. Plate XXVI, figs. 8-9. HEIMATOMYCES BOREALIS: Proc. Am. Acad. Arts and Sci. Vol. XXVIII, p. 185. Hyaline or slightly yellowish. Perithecium large and stout, its distal half or more free from the receptacle, tapering slightly to the large blunt apex. Sub-basal cell of the receptacle small ‘and flattened: the distal portion of the receptacle composed of only three cells (the fourth obso- lete or very minute), the distal one large, longer than broad, its base very oblique, the two others very long and sub-triangular, the septum between them running obliquely from the insertion of MONOGRAPH OF THE LABOULBENIACE. 293 the “trichogyne” nearly to the base of the inner cell. Perithecia, 80-90 x 22 4. Receptacle, total length, 75-80 «4; length to tip of perithecium, 110-120 p. On Desmopachria convexa Aube, Kittery Point, Maine, and New Haven, Connecticut. This species differs from those previously described by the apparent absence of the fourth cell in the distal portion of the receptacle. It is related to the preceding species in its cell arrangement, but is easily separated by its greater size, the relative position and shape of the perithecia, etc. The ascogenic cell faces outward, a position which seems exceptional. It is apparently rare, and occurs in small numbers on the anterior legs of its host. CHITONOMYCES AURANTIACUS Thaxter. Plate XXVI, figs. 6-7. HeEIMATOMYCES AURANTIACUS: Proc. Am. Acad. Arts and Sci. Vol. XXIX, p. 103. Pale straw-colored, the cell contents including numerous rather bright orange granules or oil globules. Perithecium exceeding the tip of the receptacle by from one-fifth to one-quarter of its length, small, slender, the tip usually curved outward, the lips rather prominent, its external basal cell marked by more or less distinct and very fine transverse ridges. Receptacle slender, the basal cell suffused with brown below, the sub-basal cell small and flat, the two succeeding cells elongate, the outer shorter and continued above by an unusually large basal cell of the peri- thecium: distal portion of the receptacle, as in C. borealis, composed of three cells, the two lower very long and narrow, sub-triangular, obliquely superposed, their lower extremities nearly touching the sub-basal cell. Perithecium, 50 x 14-15. ‘Total length to tip of receptacle, 85- 90 «; to tip of perithecium, 100-110 w. Greatest breadth, 22 p. » On Desmopachria convera Aube, Kittery Point, Maine. A rare species occurring on the right elytron near the middle of its distal half. Dis- tinguished from C. borealis, which occurs also very rarely on the same host, and C. Bidessarius, by its slender form and orange color as well as the details of its structure. It is a very delicate species, seldom found in good condition. The striation of the outer basal cell of the perithecium is very characteristic, but not always distinct. HYDR/AZOMYCES nov. gen. Plates XXVI and VIII. Receptacle consisting of a basal and terminal portion, the latter united to the perithecium along its inner margin and terminating in a sub-conical, terminally appendiculate free cell, the cell below the sub-terminal cell producing from its left face three outgrowths, which extend upward, and are separated as cells, ending like the terminal cell in a sub-conical body bearing a single terminal appendage and almost completely concealing the face of the sub-terminal cell : the basal portion consisting of three superposed cells, above which three or perhaps four cells form the base of the perithecium. The wall-cells of the latter arranged in four longitudinal series, each of which contains more than six (eight) cells. Spores fusiform, once-septate. In view of the invariable character of the fundamental cell structure and arrangement in the sixteen species composing the genus Chitonomyces, to which the single form included by the present genus was first referred, and the very distinct variations from this type presented by 294 MONOGRAPH OF THE LABOULBENIACE. the latter, I have concluded to separate it, basing the new genus on the greater number and different arrangement of the small appendiculate cells bearing the lower appendages, the different number of wall-cells in the perithecium and the fact that three of the lower cells of the recep- tacle are superposed. ‘The genus Chitonomyces, although so closely allied, shows no variation in these respects. The relation of the mature perithecium to the insertion of the trichogyne, and the considerable growth of the former beyond it, recalls the similar relation existent in Gera- tomyces, while the same is true to a less degree in Chitonomyces. Although these three genera are all aquatic, it does not seem probable that the similarity just referred to indicates any close relationship with Ceratomyces in view of the very different character of the male organs in either case. As in Chitonomyces, material has been lacking for a proper study of the younger stages of development, and the character of the trichogyne and antheridium has not been made out. In one specimen the procarpe was distinctly seen, and an attempt was made to represent it in fig. 24, Plate VUI. The trichophoric cell here was large and slightly inflated, lying close beside the inner appendiculate cell. The septation in this figure is somewhat misleading from the fact that it does not represent what is visible in a single plane. There seem to be at least two if not more ascogenic cells in the mature perithecium, but their structure and early development have not been seen. It is needless to say that both these genera are in a most unsatisfactory condition in so far as concerns their development, and were it not for the discovery of the new genus above mentioned as intermediate between Chit- onomyces and Peyritschiella, their position would be entirely uncertain. The single form inhabits species of two genera of aquatic beetles belonging to a family (Haliplide) no other members of which are known to be attacked by Laboulbeniacez. Hypraomyces Haipitr Thaxter. Plate XXVI, figs. 31-33; Plate VIII, figs. 22-24. Hermmatomyces Harieri: Proc. Am. Acad. Arts and Sci. Vol. XXVIL p. 32. Strongly suffused with dull amber-brown. Perithecia rather evenly inflated, the extremity evenly rounded, the lip-cells forming a flat, hyaline, abruptly projecting terminal papilla. Basal cell of the receptacle rather long, tapering below, its base suffused with blackish, the two cells superposed above it broader than long, nearly equal: the terminal cell more or less conical, the three lower appendiculate cells nearly as large as the terminal one, and almost completely hiding the sub-terminal cell on the left side. Spores, 30x3. Perithecia, 100 x 35-404. Total length to tip of perithecium about 150 pu. On Haliplus ruficollis DeG., and Cnemidotus muticus Lec., New Haven, Connecticut, and Kit- tery Point, Maine. ; This species seems by no means common, and is found usually on the right elytron of its host or on the legs. A single specimen only has been found on Cnemidotus. MONOGRAPH OF THE LABOULBENIACEA. 295 AMORPHOMYCES Thaxter. Plate V, figs. 17-29. Proc, Am. Acad. Arts and Sci. Vol. XXVIII, p. 158. Sexual organs borne on separate individuals. Male individual. Receptacle consisting of two superposed cells, the upper bearing terminally a single simple antheridium. Female individual. Receptacle consisting of a single cell, above which four (7) small cells form the base of the solitary terminal perithecium. Spores continuous. Asci four spored. Ascogenic cell solitary. Trichogyne a nearly sessile vescicle with short radiate branches. The above description is based on the characters of the single species A. Falagric, since it is quite certain that the second supposed species, A. floridanus, is generically distinct and not even nearly related to it. The genus is among the most interesting of the whole group, not only from the fact that the sexes are separated on different individuals, but from the peculiari- ties of its development. It is the only form in which the perithecium is a strictly terminal structure, and, unless it be Dimorphomyces, no other genus has a unicellular receptacle. The general development of the female is indicated by figures 20-28, the latter showing the first two septa which form in the germinating spore, dividing it into three superposed cells, Of these the lower, which may become partly divided, forms the unicellular receptacle; while of the two others the middle one (¢) constitutes the primordial cell of the perithecium proper and the upper gives rise to the procarpe. The development of the male individual is extremely simple. As in the case of the female the spore first divides (fig. 23 at the right) into three superposed cells. Of these three cells the two lower remain unchanged, and may be called the receptacle; although, from analogy with the female, one might infer that the middle cell was morphologically a basal cell of the antheridium. The terminal cell then elongates, producing a long tubular neck through which the antherozoids formed in its basal portion make their escape (fig. 20, at the right). The spores are unique from the fact that even when fully mature they show no signs of septa > and their contents are exceptional in containing a number of large oil globules (fig. 29). They are always discharged in pairs, the members of which thus develop side by side. It is also an invariable rule, the existence of which has been established by very numerous observations, that, of the members of any such spore pair, one always produces a male and the other a female. Kyen at the time of discharge there often is observable a slight difference in the size of the two spores (fig. 29), the spore at the left in the figure being somewhat smaller than its fellow.