G vvdUruiA- 111-] Book , (°,4SSk UNITED STATES DEPARTMENT OF AGRICULTURE BULLETIN No. 380 Contribution from the Bureau of Plant Industry WM. A. TAYLOR, Chief Washington, D. C. PROFESSIONAL PAPER January 15, 1917 ENDOTHIA PARASITICA AND RELATED SPECIES By C. L. SHEAR, Pathologist, and NEIL E. STEVENS, Pathol ogist, Fruit-Disease Investigations, and RUBY J. TILLER, Scientific Assistant, Office of Investigations in Forest Pathology CONTENTS Page Taxonomy 1 Introduction 1 The Genus Endothia 3 The Species of Endothia 13 Morphology and Development .... 22 Mycelium 22 Stromata • . . . . 23 Spore Measurements 30 Physiology 36 Cultural Studies 36 Distribution of the Species of En- dothia 4g Page Physiology— Continued Discovery of Endothia parasitica In China . ^ 54 Discovery of^Endothia parasitica in Japan 58 Present Distribution of Endothia parasitica in America 59 Host Relations of the Species of En- dothia 59 Summary 74 Literature Cited 77 WASHINGTON GOVERNMENT PRINTING OFFICE 1917 " ' BUREAU OF PLANT INDUSTRY. Chief of Bureau, William A. Tailor. Office of Investigations in Forest Pathology. scientific staff. Haven Metcalf, Pathologist in Charge. George G. Hedgcock, Emile P. Meineeke, and Perley Spaulding, Pathologists. J. Franklin Collins, Carl Hartley, William H. Long, and James R. Weir, Forest Patholo- gists. Samuel B. Detwiler, Forest Inspector. G. Flippo Gravatt, C. J. Humphrey, and N. Rex Hunt, Assistant Pathologists. Roy G. Pierce, Forest Assistant. John S. Boyce, Ruth M. Fleming, Ernest E. Hubert, Gilbert B. Posey, Paul V. Siggers, and Ruby J. Tiller, Scientific Assistants. P. of D. MN 25 1917 UNITED STATES DEPARTMENT OF AGRICULTURE J&f'^i. | BULLETIN No. 380 Contribution from the Bureau of Plant Industry WM. A. TAYLOR, Chief Washington, D. C. PROFESSIONAL PAPER January 15, 1917 ENDOTHIA PARASITICA AND RELATED SPECIES. By C. L. Shear, Pathologist, and Neil E. Stevens, 1 Pathologist, Fruit-Disease Investigations, and Ruby J. Tiller, Scientific Assistant, Office of Investiga- tions in Forest Pathology. CONTENTS. Page. Taxonomy 1 Introduction 1 The genus Endothia 3 The species of Endothia 13 Morphology and development 22 Mycelium 22 Stromata 23 Spore measurements 30 Physiology 36 Cultural studies 3<3 Physiology— Continued . Distribution of the species of Endothia. . Discovery of Endothia parasitica in China Discovery of Endothia parasitica in Japan Present distribution of Endothia para- sitica in America Host relations of the species of Endothia . Summary Literature cited TAXONOMY. INTRODUCTION. The discovery of a serious canker of the chestnut in the New York Zoological Park in 1904, by Merkel (49), 2 first attracted the atten- tion of pathologists and foresters to what has proved to be one of the most serious epidemics of a plant disease ever known in this country. The fungus which was found associated with these cankers (PI. I and PL II, fig. 1) and soon demonstrated experimentally to be their cause was described by Murrill (57) in 1906 as a new species of Diaporthe (D. parasitica). Search for the fungus in other places in New York and vicinity soon showed that it was already estab- lished and apparently rapidly spreading. Investigations which have been continued and extended from year to year have shown 1 Formerly Pathologist, Office of Investigations in Forest Pathology 2 Serial numbers in parentheses refer to " Literature cited," at the end of the bulletin. Note.— This ; bulletin is of value to botanists, especially plant pathologists and mvcolo- gists, and to all persons who are interested in the study of chestnut blight. 43737° — Bull. 380 — 17 1 2 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. conclusively that the disease is spreading very rapidly, especially west and south from New York and also north and east. The exact identity and relationships of the fungus causing the disease and the origin of the epidemic soon became the subject of study by various mycologists and pathologists. Different explana- tions were offered for the sudden appearance and behavior of the disease, one view being that the fungus was probably a foreign parasite which had been introduced: another, that the organism was probably a native species which had recently attracted attention, chiefly by reason of the weakened condition of the chestnut trees due to abnormal climatic or other conditions. In attacking the problem of the origin of the parasite and its pos- sible control, it was evidently necessary to secure all the information possible in regard to its life history, identity, distribution, and re- lationships. The senior writer in an unpublished paper prepared in 1908 pointed out the close relationship and possible identity of J>'niju>rflic parasitica with certain species of Endothia. Clinton (16) and Farlow (-28) soon after also made the same suggestion. Two species of Endothia had already been described from this country by Schweinitz (74) under the old generic name, Sphaeria. These, however, had in recent years been regarded as a single species and referred to Endothia gyrosa (Schw.). Owing to a lack of knowledge of the types of these two species and for want of good specimens showing ascospores, it was difficult to determine what species of Endothia were indigenous in the eastern United States. Since it had been suggested that Diaporthe parasitica was either identical with one of Schweinitz's species or a mere variety of it, the present writers undertook a thorough study of the genus Endothia in its taxo- nomic, ecological, and pathological relations. It was first necessary to determine the identity of the two species already described by Schweinitz from America and also to learn their distribution and host relations. As one or both of Schweinitz's species were reported to occur in southern Europe on chestnut, it was important to obtain exact knowledge in regard to the identity and relationships of the European species. The senior writer spent several months in Eu- rope collecting material of Endothia in the field and studying her- barium specimens of types and authentic collections of Schweinitz and other authors. Material was also acquired by collection and exchange with pathologists and mycologists in nearly every region of the world in which Endothia was known to occur. Comparative cultural studies were made of all the living material secured, as well as inoculation experiments on various hosts. The recent discovery of the typical chestnut-blight parasite, Endothia parasitica, by Meyer (27, 76, 78), in China and Japan and the failure to find in Europe or America any native form which w T ould produce the disease appear to settle beyond question its foreign origin. ENDOTHIA PARASITICA AND RELATED SPECIES. 3 The present paper presents the results of several years' field and laboratory study of the species of Endothia. This includes the study of practically all the herbarium material of this genus preserved in the principal herbaria of Europe and America ; also field and lab- oratory studies of over 600 new collections from various localities and hosts in America, Europe, and Asia. Over 4,000 cultures have been studied and about the same number of inoculations made. These studies include the systematic relations of the species of Endothia and their physiological behavior on various culture media and under various conditions of light, moisture, and temperature; also inoculation experiments with the various species on various hosts. The writers wish to record here their- grateful acknowledgment and thanks for opportunities to examine specimens and for assistance rendered by various mycologists and pathologists and directors and curators of botanical gardens and museums, especially the following: Prof. O. Comes, Naples; Prof. Eomualdo Pirotta, Prof. Giuseppi Cuboni, and Drs. E. Pantanelli and L. Petri. Rome; Prof. P. Bac- carini, Florence; Prof. P. A. Saccardo, Padua; Dr. G. Briosi, Pavia; Dr. J. Briquet, Delessert Herbarium, Geneva; M. G. Beauverd, Boissier Herbarium, Geneva; Prof. L. Jost, Strasburg; Prof. W. Pfeffer, Leipzig; Dr. G. Lindau, Berlin; Dr. J. W. C. Goethart, Leiden; Prof. H. O. Juel, Upsala; Dr. P. Hariot, Paris; Sir David Prain, Kew ; Dr. A. B. Eendle, British Museum; Prof. P. B. Balfour, Edinburgh; Prof. T. Petch, Peredeniya, Ceylon; Dr. C. Spegazzini, La Plata, Argentina; Dr. TV. G. Farlow, Harvard University; Dr. W. A. Murrill, New York Botanical Garden ; Mr. Stewardson Brown, Philadelphia Academy of Science; Dr. G. T. Moore, St. Louis Botanical Garden; Prof. E. Bethel, Denver, and Drs. G. P. Clinton, P. J. Anderson, and F. D. Heald. The writers have also received specimens and cultures from numerous other colleagues which have been of great assistance and are duly appreciated. THE GENUS ENDOTHIA. The genus Endothia was established by Elias Fries in 1849 (33, pp. 385-38G), as follows: (X. Endothia. Fr.*) * Colore rubro fulvove, habitu Tuberculariae, peritheciis cellulosis difformi- bus pallidis, ascis diffluentibus, facile distinctum genus, nobis exoticum, sed jam in Europa australi obvium v. c. Sph. gyrosa Schw.— et subgenus, tuber- culo uniloeulari, sistit S. Tnbercularia Dec. Omnium horum generum char- acters proxime plenius exhibeamus, examinatis multis speciebus exoticis. The description of the genus transcribed here was published as a footnote in the work cited and was evidently based on the specimens contained in Fries's herbarium at the time the book was written, 4 BULLETIN 380, U. S. DEPAETMENT OF AGRICULTURE. Fries (31, p. 73) had at that time, according to his own statement, authentic specimens of Sphaeria gyrosa sent him by Schweinitz and also the specimens collected by Guepin and Levieux in France, which he identified as this species. In Fries's herbarium at Upsala at present are found specimens of true S. gyrosa Schw. with Schweinitz's autograph label, but no specimens of S. gyrosa could be found attributed to Guepin or Levieux. There is a small packet marked " Sph. gyrosa" apparently in Fries's handwriting, but there seems to have been some confusion in the labeling or mounting of this specimen, as a small stroma of Ilypoxylon annulatum which does not look at all like Endothia is included. The other piece consists of an irregular pycnidial stroma which may be the southern European specimens referred to in the description quoted. Fries's identification of this European material as E. gyrosa was apparently based chiefly upon its superficial resemblance to the pycnidial stromata of Schweinitz's American specimens. The senior writer has seen and made a careful microscopic examination of a specimen col- lected by Guepin in France and preserved in De Notaris's herbarium at Rome. It is labeled "Sphaeria gyrosa Fries, Guepin, Angers." The specific name " gyrosa " has been crossed out by De Xotaris and " radicalis Schw." written above it and the date " April, 1845," added. This appears to be a part of the same collection that Guepin sent to Fries, as the specimen agrees well with Fries's description and consists chiefly of pycnidial stromata which are rather larger than is usual for Sphaeria radicalis and show con- siderable superficial resemblance to the stromata of Sphaeria gyrosa Schw. A thorough examination of this specimen, however, reveals a few perithecia and ascospores, which leave no doubt that it is S. radicalis of Schweinitz, as indicated by De Notaris on the label. What the plant sent Fries by Levieux was is unknown, as no speci- men so labeled could be found in Fries's herbarium. It appears from all the evidence at hand that Fries was mistaken in his identi- fication of the material from Levieux and Guepin, as no specimens of the true Sphaeria gyrosa Schw. have yet been seen from Europe. There seems to be no doubt, however, that Fries intended the true Sphaeria gyrosa Schw. to represent the type of his genus Endothia, as he had a part of Schweinitz's original collection at the time and never definitely placed any other species in the genus; hence, Sphaeria gyrosa Schw. should be adopted as the nomenclatorial type of the genus. It is clear from Fries's writings and specimens that he knew Sphaeria radicalis Schw., as he had American speci- mens from Schweinitz as well as European collections at the time he founded this genus. He did not, however, apparently regard it as congeneric with S. gyrosa. His specimens of S. radicalis show ENDOTHIA PARASITICA AND RELATED SPECIES. 5 the typical perithelia with necks, whereas no perithecia have been found in any of Schweinitz's specimens of JS-. gyrosa examined by the writers. Fries, in common with Schweinitz, regarded the pyc- nidial cavities of S. gyrosa as perithecia. When the pycnidia of S. gyrosa are mistaken for perithecia and compared with the real perithecia of S. radicalis the differences appear marked. It was therefore quite as natural for Fries to place the two species in different genera as it had been for Schweinitz to place them in dif- ferent tribes of the genus Sphaeria. Fries's mistake in describing as perithecia the pycnidial cavities in the stroma of S. gyrosa ex- plains his reference to the asci as " ascis diffluentibus." Believing that he had perithecia but finding no asci, he interpreted this as indicating that they had disappeared. According to the plan of accepting only names originally applied to the ascospore stage, this name would be invalid, as proposed by Fries, and would be attributed to De Notaris, who placed the peri- thecial form of Sphaeria radicalis Schw. in the genus and described the ascospores. There is not the slightest question, however, in regard to the identity of the different stages of this fungus and their genetic connection, and the name Endothia has been almost invariably applied to these two species in both stages. SYNONYMY. There are only two true generic synonyms of Endothia: En- dothiella Saccardo, 1906 (71, p. 278) and Calopactis H. and P. Sydow, 1913 (81, p. 82). Endothiella was based on Endothiella gyrosa Sacc, which, according to authentic specimens from Saccardo, is undoubtedly the pycnidial form of Endothia -fluens as found in Italy. Calopactis was based on C. singularis, the pycnidial condi- tion of Endothia singularis (H. and P. Syd.) S. and S. Ascospore cultures of this have not yet produced any pycnidia, but the proof of the genetic connection of the two stages appears rather con- clusive from the occurrence of pycnidia and perithecia in the same stroma, as shown in Plate XII. Perithecial stromata and ascospores were also found in the specimen of the Sydow exsiccati in the Patho- logical and Mycological Collections of the Bureau of Plant Industry. Von Hohnel (43, p. 1479-1481) considers Cryphonectria Sacc. as a synonym of Endothia, taking C. gyrosa (B. and Br.) as the type of that genus because it is the first species listed by Saccardo in con- nection with his description of the genus. Saccardo, however, had previously established Cryphonectria as a subgenus, with C. abscon- dita as the type, which is not an Endothia. Valsonectria is also con- sidered by Von Hohnel a synonym of Endothia, but apparently he had not compared specimens of Spegazzini's fungus, which is found upon examination of the type species to be separate from Endothia. The 6 BULLETIN 380, U. S. DEPAKTMENT OF AGRICULTURE. Tulasnes (83, p. 87-89) do not appear to have regarded Endothia as distinct from Melogramma, to which they referred E. gyrosa. The type of Melogramma, however, is M. melogramma (Bull.), which has a somewhat similar stroma, but the ascospores are 3-septate and dark colored and the perithecia not separable from the stroma, while the pycnospores are long, slender, and curved. STUDY OF EARLY COLLECTIONS AND TYPES. There has always been more or less uncertainty in regard to the identity of the older species of this genus of fungi. In order to get more light on this subject, a thorough study of all the available ma- terial in the way of literature, type specimens, and manuscripts was made. The first species to be described in this country was Sphaeria gyrosa Schw. This was collected by Schweinitz at Salem, N. C, and published in 1822 (72, p. 3). 1 Two hosts were given in the original description, Fagus and Juglans. As Schweinitz's description was prepared before the advent of careful microscopical studies and spore measurements, it is impossible to identify the organism satisfactorily from the original description. It was, therefore, important, if possible, to locate the type specimens upon which the description was based. Schweinitz's herbarium was left at his death, in 1834, to the Philadelphia Academy of Science. His specimens of fungi at the time they were transferred to the acad- emy were contained in small, folded paper packets, as shown in Plates V and VI. These packets were then inclosed in other heavy paper wrappers, folded to small quarto size, and three or four of these large packets, each bearing a manuscript list of the species contained, were then inclosed in quarto pasteboard covers, tied with tape. The in- dividual species packets were labeled in Schweinitz's handwriting, with the name of the species and the locality of the collection, as shown in Plate V, figure 2. These species packets frequently bore the names of several locali- ties, but usually two, Salem [N. C] and Bethlehem [Pa.], as most of his collecting was done at these places. This fact, in addition to the evidence afforded by the specimens in the packets, clearly indicates Schweinitz's method of handling his specimens. Frequently some of the specimens in a packet show the remains of a gummed strip. This will be noticed in Plate III, which indicates 1 24. Sphaeria i .■<■ vvw^ 1 /? Cc _ /*// _- *Se/,7t ■ J"r sQ^/j?y _- ,yy/c . ^rt^to L '/, ■* -. , ~t*%£4t&i Fig. 1— Photograph of Schweinitz's Manuscript Notes, with His Description of Sphaeria radicalis.' Fig. 2.— Specimen of S. radicalis in the Mounted Collection of Schweinitz, as Prepared by Michener-, Also Original Packet WITH SCHWEINITZ'S AUTOGRAPH LABEL. Bui. 380, U. S. Dept of Agriculture. Plate VI. / 4£/- %-#<$ -^St/rft . ±Yr , / c e~ Y J ee^^ec ay t r />^ /?&/^<^, •y ~/&-*^* . — *-^-\ /^W^- c. mfti . nmDifWa IBM fefc h i£ $3 i Fig. 1.— Photograph of the Specimen in Schweinitz'S Herbarium Mounted by Michener. Not True Endothia gyrosa but a Nectria. Fig. 2.— Original Paper Packet in which Schweinitz'S Type Material of E. gyrosa was Pre- served, with His Autograph Label. Bui. 380, U. S. Dept of Agriculture. Plate VII. Endothia gyrosa Growing on the Recently Cut End of a Living Branch of Fagus sp. Natural Size. Bui. 380, U. S. Dept. of Agriculture. PLATE VIII. Mycelial Fans of Endothia parasitica under the Bark of Castanea dentata. Illustration from Heald (39), by courtesy of I. C. Williams, Pennsylvania State Forestry .Department. ENDOTHIA PARASITICA AND RELATED SPECIES. 17 Synonyms — Continued. Peritheeia — Continued. Sphaeria radicalis Schw., 1832, Fun. Am. Bor., p. 197. Sphaeria radicalis Schw., Mont., 1834, in Ann. Sci. Nat. Bot, s. 2. t. 1, p. 295. Sphaeria (Diatrype) radicalis Fries, Currey, lS. r >8, in Trans. Linn. Soc. London, v. 22, pt. 3, p. 272, pi. 47, fig. 89. p. p. Valsa radicalis Ces. and De Not., 1863, in Connn. Soc. Crittog. Ital., v. 1, p. 207. Endothia radicalis (Schw.) Ces. and De Not.. 1SG3. in Connn. Soc. Crittog. Ital., v. 1, opp. p. 240. Melogramma gyrosum L. II. and C. Tul., 1863, Selecta Fung. Carpol., t. 2, p. 87. p. p. max. Sphaeria (Diatrype) radicalis Schw., Currey, 1865, in Trans. Linn. Soc. London, v. 25, pt. 2, p. 244. Endothia gyrosa (Schw.) Fckl., Sacc, 1882, Syll. Fung., v. 1, p. 601. p. p. Endothia gijrosa var. rostellata Sacc, 1882, Syll. Fung., v. 1. p. 602. Endothia radicalis (Schw.) Wint, 1887, Pilze, p. 803. Endothia gyrosa Schw., Ell. and Ev.. 1892, No. Amer. Pyren., p. 552. p. p. Endothia virginiana P. J. and H. W. And., 1912, in Phytopathology, v. 2, no. 6, p. 261. Endothia gyrosa (Schw.) Fries, Clint., 1913, in Conn. Agr. Exp. Sta. Rpt., 1911-12, p. 425. Endothia pscudoradicalis Petri, 1913, In Atti R. Accad. Lincei Rend. CI. Sci. Fis., Mat. e Nat., s. 5, v. 22. sem. 1, fasc. 9, p. 654. Endothia gijrosa (Schw.) Fckl.. Hohnel, 1913, in Sitzber. K. Akad. Wiss. [Vienna], Math. Naturw. Kl., Abt. 1, Bd. 122, Heft 2, p. 298. Type specimen. — Sowerby in Herb. Kew. on Castanea sativa, New Forest, England. Coll. C. Lyell, Apr. 15, 1809. Pycxidia. — Stromata corticular or subcorticular, truncate conical to pulvi- nate, usually separate and gregarious, but frequently confluent, 0.75 to 3 mm. in diameter by 0.5 to 2.5 mm. high, compact, varying from light auburn to chestnut on the surface and capucine yellow to cadmium orange within ; pycnidia consisting of simple or more or less complex and irregular chambers in the stroma, opening by an irregular pore or slit at the apex of the stroma ; sporophores usually simple, sometimes branched near the base, cylindric to subclavate, 10 to 13 n long, sometimes 24 to 30; pycnospores oblong to rod- like, pale yellowish in mass, 3 to 5 by 1.5 to 2 n, mostly 3.5 to 4 by 2 fi. Perithecia. — Stromata the same or similar to those producing pycnidia ; peritheeia membranous, few to many, mostly 15 to 25, 300 to 400 n in diameter, usually arising in the lower portion of the stroma, irregularly arranged in one to three layers, prolonged into slender necks which penetrate the stroma above and protrude usually from 300 to 600 p, terminating in conical ostioles ; asci oblong fusoid or subclavate, very short stipitate, 30 to 40 by 6 to S /x, mostly 30 to 35 by 7 p, ascospores irregularly biseriate, oblong fusoid or subellipsoid, not constricted at the septum, hyaline with a thin gelatinous envelope, 6 to 10 by 3 to 4.5 p, mostly 6.5 to 9 by 3 to 4 /x.. Cultural characters. — Cultures one month old on white corn meal show a compact growth with a nearly smooth surface. The color ranges from light cadmium to empire yellow, and the medium becomes perilla purple. Pycnidia and spores usually appear a little later, forming large erumpent stromata which extrude thick masses of pycnospores. The light mycelium with large 43737°— Bull. 380—17 2 18 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. pycnidial stromata and spore masses are distinguishing characters on this medium. Hosts. — America: Exposed roots and branches of Q. alba, Q. coceinca, Q. marylandica, (,>. prinus, Q. rubra, Q. velutina, and Castanea dentata. Europe: Specimens examined, Quercus pedunculata, Castanea sativa, Alnus glutinosa, Ulmus campestris, Carpinus betula, and Corylus sp. Japan: Castanea sp. and Pasania sp. It is also reported on Aesculus, Fagus, and Juglans by Traverso. Type locality. — New Forest, England. Geographical distribution. — America: Southern Pennsylvania and Ohio to South Carolina and northern Mississippi. Europe: Southern England, France, South Germany, and Switzerland to southern Italy and Transcaucasia. Asia: Japan. Illustrations.— Sowerby, 1814, Col. Fig. Engl. Fungi, Sup., pi. 438; Currey, 1858, in Trans. Linn. Soc. London, v. 22, pt. 3, pi. 47, fig. 89 (2 upper spores) ; Ces. and De Not., 1863, in Comm. Soc. Crittog. Ital., pi. 3 ; Sacc, 1873, in Atti Spc. Veneto-Trentina Sci. Nat. Padova, v. 2, fasc. 1, pi. 14, fig. 63-65 ; Sacc, 1883, Gen. Pyren., pi. 6, fig. 6 ; Ruhl., 1900, in Hedwigia, Bd. 39, pi. 2, fig. 10 ; Trav., 1906, in Soc. Bot Ital. Fl. Ital. Cript., pars 1, v. 2, fasc. 1, p. 180, fig. 34; P. J. and H. W. And., 1913, in Penn. Chestnut Tree Blight Com. Bui. 4, p. 22, fig. 2, A and C; Clint., 1913, in Conn. Agr. Exp. Sta. Rpt, 1911-12, pi. 28, fig. b, e, h, and j ; Petri, 1913, in Atti R. Accad. Lincei Rend. CI. Sci. Fis., Mat. e Nat, v. 22, sem. 1, fasc. 9, p. 656, fig. 1-3. Exsiccati. — Pycnidia : Thiim. Myc. Univ., 769, on Castanea ; Sacc. Myc. Yen., 670, on Carpinus betula; Sacc. Myc. Ven., 929, on Castanea. Perithecia : Fckl. Fun. Nass., 640, on Ulmus campestris; Erb. Critt. Ital., 9S6, on Castanea; Rab. Herb. Viv. Myc, 254, on Castanea. Roum. Fun. Sel. Gal., 989, labeled Endothia gyrosa Schw. on beech is appar- ently young Hypoxylon coccineum. The most important synonyms given here have already been dis- cussed. Of the others the writers have examined the types or col- lections upon which the identifications w T ere based. All the material of Endothia in the herbaria of Cesati, De Notaris, Fuckel, and Berkeley, as well as other smaller collections, has been carefully studied. E. virginiana And. and And. has been studied in cultures, as well as typical specimens from the authors of the species, and agrees in every particular with E. fluens. Through the kindness of Dr. Petri a part of the type of his E. pseudoradicalis has been examined, but unfortunately no cultures could be obtained from the specimen. The writers have been unable to distinguish his specimen from forms of E. fluens which appear to show all the intermediate conditions of variation connecting it with typical E. fluens. The ascospores of E. fluens are more variable in size and shape than those of any other species of Endothia studied. After examining many specimens of this species from Europe, it does not seem possible at present to separate any of them. The case of E. pseudoradicalis can not perhaps be regarded as closed until more material of it has been collected and compared in culture. In fact, the slide from the type of Sphaeria radicalis Schw. shows ascospores of both the narrow and broad form. The photomicro- ENDOTHIA PARASITICA AND RELATED SPECIES. 19 graph, Plate XVII, fig. 9, shows an ascospore which agrees with Petri's description and figures. ENDOTHIA FLUENS MISSISSIPPIENSIS S. and S. nov. comb. Synonym : Endothia radicalis mississippiensis Shear and Stevens in U. S. Uept. Agr., Bur. Plant Indus. Cir. 131, p. 4. 1913. Type specimen. — No. 1782, on Castanea dentata, Blue Mountain, Miss., N. E. Stevens, Feb. 13, 1913. Deposited in Pathological and Mycological Collec- tions, Bureau of Plant Industry. Cultural characters. — Cultures one month old on white corn meal show a compact, rather uniform surface, the color of the mycelium varying from cad- mium orange to xanthine orange. This variety is distinguished from the species by the color of its mycelium, by the numerous small pycnidia thickly scattered over the surface of the culture, and by the lack of any purple color in the medium. Hosts. — Castanea dentata, Quercus alba, and Q. velutina. Geographical distribution. — Northern Mississippi, Kentucky, Tennessee. Collections examined. — On Castanea dentata: No. 1706 A. pycnidia, Corinth, Miss., T. E. Snyder ; no. 708, pycnidia, Dumas, Miss., T. E. S. ; no. 1782, ascospores, Blue Mountain, Miss., N. E. S. ; no. 1806, ascospores, Blue Moun- tain, Miss., N. E. S. On Quercus : No. 19S9, pycnidia, Danville, Ky., N. E. S. ; no. 1995, pycnidia, Danville, Ky., N. E. S. ; no. 2032, pycnidia, Lexington, Tenn., N. E. S. ; no. 2255, pycnidia. Sardis, Miss., S. and S. No morphological characters have yet been found to distinguish this variety. It is therefore separated on its cultural characters, which are marked and constant. The plant was first collected by T. E. Snyder, of the Bureau of Ento- mology. ENDOTHIA LONGIROSTRIS Earle, 1900, in Muhlenbergia, v. 1, no. 1, p. 14. Synonym : Perithecia : Diatrype radical is (Schw.) Fries, Mont., 1855, in Ann. Sci. Nat. Bot. 4, t. 3, p. 123. Not Schw. Type specimen. — No. 4340. A. A. Heller, Plants of Porto Rico. In Herb. N. Y. Bot. C4arden. Pycnidia. — Stromata corticular, erumpent, gregarious, sometimes confluent, 1 to 3 mm. in diameter, subcoriaceous, surface orange rufous to chestnut, in- terior zinc orange ; pycnidia consisting of irregular labyrinthiform cavities open- ing by a single large pore or irregular rupture at the apex of the stroma ; sporophores slender, somewhat tapering upward, mostly S to 10 n long ; pyeno- spores oblong elliptic, hyaline or yellowish in mass, when expelled forming a stout spore horn or tendril, colored like the stroma on the outside, 2 to 4 by 1 to 1.5 jit. Perithecia. — Stromata the same as those producing pycnidia, but larger and frequently confluent, forming linear series in crevices in the bark ; perithecia arising usually at the base of the pyenidial stroma, mostly 3 to 10 in the sepa- rate stromata, membranous, 300 to 400 ix in diameter, mostly in a single irregu- lar series, prolonged into long necks, 1.5 to nearly 1 cm. long, sec. Earle, inter- nally black, externally same color and structure as the stroma ; ostiole acute : asci oblong cylindric to fusiform, 25 to 35 by 5 to 7 fi, mostly 30 by 6 /i ; asco- spores overlapping uniseriate to irregularly biseriate, hyaline, ovoid to ovoid elliptical, 6 to 8.5 by 3 to 4 ft, mostly 7 to 7.5 by 3 to 3.5 fi. 20 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. Cultubal chabacteks. — Cultures one month old on white corn meal have a uniform cadmium orange to xanthine orange color. The entire surface is covered uiili a compact growth, irregularly ridged. Tiny mars orange spore masses are scattered irregularly over the surface. Cultures of this species closely' resemble /•:. fluens mississippiehsis on this medium, being distinguished by the smaller and much less numerous spore masses. The medium is changed to amber brown just below the mycelium, shading into mars yellow; whereas, in the case of /;. fluens mississippiensis the color of the medium is very little changed. Type locality. — " Calcareous hills east of Santurce, Porto Rico, altitude 10 ft." Geographical distribution. — Porto Rico and French Guiana. Exsiccati. — Pycnidia and perithecia : Heller, Plants of Porto Rico, no. 4340. This species, which appears to be subtropical or tropical in its range, is known at present from only three collections, the type col- lection from Porto Rico, a collection by Prof. N. Wille, No. 816, Porto Eico, distributed by the New York Botanical Garden, from which the cultures were obtained ; and one made by Leprieur, No. 392, in French Guiana, and determined by Montagne as Diatrype radicalis (Schw.). A specimen of this collection apparently labeled by Mon- tagne and preserved in the Delessert Herbarium at Geneva has been examined and found to agree with the type material of E. longiros- tris. It is readily distinguished from E. tropicaUs by its smaller asco- spores and pycnospores, and from E. fluens by its narrower and more acute ascospores and the long, slender necks of the perithecia. ENDOTHIA TROPICALIS Shear and Stevens sp. nov. Synonyms : Diatrype gyrosa Berk, and Broome, 1875, in Jour. Linn. Soc. [London], v. 14. p. 124. Nectria gyrosa Berk, and Broome, 1877, in Jour. Linn. Soc. [London], v. 15, p. 86. Cryphonectria gyrosa (Berk, and Broome) Sacc, in Syll. Fung., v. 17, p. 784. 1905. Endothia gyrosa (Schw.) Fckl., Hohnel, 1909, in Sitzber. K. Akad. Wiss. [Vienna], Math. Naturw. Kl., Abt. 1, Bel. 118, Heft 9, p. 1480. Type specimen. — No. 2807 S. and S., on Elacocarijus glandulifer, Hakgala, Ceylon, Coll. T. Peteh, August, 1913. Pycnidia. — Stromata corticular, pustular to pulvinate, usually gregarious or scattered, rarely confluent, 1 to 5 mm. in diameter, early becoming friable, orange chrome when fresh to sanford brown when old and weathered ; pycnidia consisting of numerous irregular cavities in the stroma; sporophores mostly simple, clavate, tapering above. G to 10 n long.; pycnospores continuous, oblong in cylindric, very variable in size and shape, pale yellowish in mass, 3.5 to 7 by 1.5 to 2.5 n. Perithecia. — Stromata the same or similar to those bearing pycnidia ; peri- thecia black, membranous, collapsing when dry, 5 to 50 or more in a stroma; 250 t<> 500 ix diameter, irregularly arranged in one to three layers, bearing slender necks which penetrate the stroma and project 0.25 to 1 mm., termi- nating in acute ostioles ; asci oblong or subclavate, nearly sessile, 40 to 50 by 7 fi ; ascospores irregularly biseriate, subelliptical, obtuse, not constricted at ENDOTHIA PARASITICA AND RELATED SPECIES. 21 the septum, hyaline with a gelatinous envelope, 7.5 to 10.5 by 3.5 to 5 fx, mostly 8 to 10 by 4 to 4.5 ft. Cultural characters. — Cultures one month old on white corn meal show small numerous, thickly scattered pycnidia and spore masses very similar to E. parasitica. The mycelium is orange buff to apricot orange. Tins species differs from E. parasitica in culture, chiefly in the brighter color of its mycelium. Host. — Rotten logs and stumps of Elaeocarpus glandulifer. Type locality. — Hakgala, Ceylon. Geographical distribution. — Only known from Ceylon at present. One other collection of this species, No. 290 G. H. K. T. [Thwaite], N. Eliya, Ceylon, 6,000 feet, has been examined in the Kew Herbarium. Through the kindness of Mr. T. Fetch, of Peredeniya, the writers have received two large collections of this fungus. Some of the material was in a living condition and enabled the writers to obtain pure cultures for comparison with the other species of Endothia. This species is closely related to E. parasitica, but is readily sepa- rated by its larger ascospores and larger and more variable pycno- spores and its nonparasitic habit. ENDOTHIA PARASITICA (Murr.) P. J. and H. W. And., 1912, in Phytopathology, v. 2, no. 6, p. 262 Synonyms : Diaporthe parasitica Murrill, 1906, in Torreya, v. 6, no. 9, p. 1S9. Valsoncctria parasitica Rehm, 1907, Asc. Exs., no. 1710. Valsonectria parasitica Rehm, 1907, in Ann. Mycol., v. 5, no. 3, p. 210. Endothia gyrosa var. parasitica Clint. 1912, in Science, n. s., v. 36, no. 939, p. 913. Endothia gyrosa (Schw.) Fckl. Hohnel, 1909, in Sitzber. K. Akad. Wiss. [Vienna], Math. Naturw. Kl.. Abt. 1, Bd. IIS, Heft 9, p. 1480. Type specimen. — Herbarium N. Y. Bot. Garden, on Castanca dentata, Bronx Park, New York City, Nov. 26, 1905, Coll. W. A. Murrill. Pycnidia. — Stromata corticular, slightly erumpent to truncate conical, usually separate and gregarious, frequently confluent in more or less linear series especially in old rimose bark, 0.75 to 3 mm. in diameter by 0.5 to 2.5 mm. high, varying from capucine yellow when young to auburn when old and weathered ; pycnidia consisting of irregular cavities in the stroma, 100 to 300 ,tt in diameter ; sporophores mostly simple, subclavate, acute at the apex, usually 12 to 20 by 1.5 /x, more elongated filaments sometimes reaching 50 fi or more being frequently found among the normal sporophores ; pycnospores, oblong to o CO — X ,_, ; CO^t "? H H ** 1 ; LO 0C CN : .o -t- X r- Oscn o r^ \ ~f \\ '-O^V c coo ? o u o ■* CN r-^a CN CiO *— CO CN CC 1-1 "^ CJ ^ C?r*r-'C O 51 O f"^ CN t^ I t- II cicr- M t~o I so T* tft t}< CS •f ca co ^.u; §c »J5 Wil s o co >h ei MCfH t^ wo CO Nr> CN t^ I OS || C~0 f^ 9i~ US CN ^f CO CN -^C" CM CM ri WJCN •3 CO CN ^ Cv £ COC CO a "H ■ I CD o o OS Ii H - II : CO 1 N N H 00 CO CD c£ .-H 00 t^ COO 3 COC ■S rH iflp C£ H 3 os -. co B 'i B 53 ..834 8 j ► 1 SX-O co w & c- . o i «8-T CO 1 IN X if OS 00 i- OS o OS Os OS OS l~ OS to O OSC OS o •* OS— i ■o os ostr: OSCN 1 ,_ || — ' C* CO 1 O II B9© 01 SiOi-t 1 S"* ,_, HH com -TlO if) CC lS C CO if If CO o "II l- Nrtp* o a ^Hr-I CO ■-< if) OS 1 CI CO r- ■* 1 -<** C*3 t> cor~ ,_, CO C IN o CO Tf 0C CO 50 V — . tr to© c s os^ • Ii o CO* ""* IT - ** fc o 00 .iCO O50Qf -H>-H C > ifJC > 1 1-1 „. OS . CO oo oo wotc CSC £ OS to 50 to O' ^ 50 Lr r~c S -tf e> i toe O *J< -*,-( c "" "■* "~ - o ' c WWMff CO . CO sSiS 00 ^--< NOGC1 ■ceo eeos •-I e»N« c to . to MN 50 Osrjl t£ COtC ,_, CO ,_, ,_( -t< [ | HMOr oi '-'eo 1 ^ f-H IT CM -■ c CO . OS OS 00 lOtDC Osr^ to COCO ~ coo r>- 00 t^ CO ■* OS if) CO -H t~ t~ co — i o -*eo c I^I^I- l- OS I- to r- if to to tc -H00 OSC IO in CO i- rt^H^rfrf _ r ~ i Hr " CC 00 r~ 00 -s a s> c 3 i J s a 03— ' 03 1. a 03 t/3 ii Castanea dentat From Pantanel Gordonsville, V 3 as c d 03 CO- O 2 c 2> - o3 c3 ;_ SI* "3 c a a - w ..C "3 e'e 3 flr* c9 C 1 - i-.Sc £ o E B 1 r T c - J c 03 _o as s "3 ^ o a go =3^ >> o E - as p. o &a _r a. 5 o s :hia parasitica: n Castanea sp., rom Pantanelli ordonsville, Va denton.Md 03 c Eh c as o as On f\ •* ;• * 1 2 3 4 5 < \ \ » i V /■ w # J -67 10 •0 ' 11 12 13 14 15 PHOTOMICROGRAPHS OF PYCNOSPORES AND ASCOSPORES OF ENDOTHIA. Figs. 1 to 6.— Pycnospores: 1, Endothiagyrosa; 2, E. sinqularis; 3, E. fluens; 4, E. longirostris; 5, E. parasitica (American); 6, E. parasitica (Chinese). Figs. 7 to 1 5.— Ascospores: 7, E. gyrosa; 8, E. singularis; 9, Sphaeria radicalis, from Schweinitz'S Specimen in Fries's Herbarium; 10, Endothia pseudorad- icalis; 11, e. fluens; 12, e. fluens mississippiensis; 13, e. longirostris; 14, E. tropicalis; 15, E. parasitica. Bui. 380, U. S. Dept. of Agriculture. PLATE XVIII. Endothia parasitica on Plate Cultures of Corn-Meal Agar 4 Weeks Old. The Upper Plate Was Kept in Total Darkness; the Lower Plate in the Direct Light of a North Window. Bui. 380, U. S. Dept. of Agriculture. Plate XIX. ENDOTHIA SPECIES ON WHITE CORN MEAL (10 GRAMS OF CORN MEAL TO 20 C. C. OF WATER). CULTURES 2 MONTHS OLD. Fig. 1.— Endothia gyrosa; Fig. 2.— E. singularis; Fig 3.— E. fluens; Fig. 4.— E. FLUENS MISSISSIPPIENSIS. Bui. 380, U. S. Dept. of Agriculture. Plate XX. i ENDOTHIA SPECIES ON WHITE CORN MEAL (10 GRAMS OF CORN MEAL TO 20 C. C. OF WATER). CULTURES 2 MONTHS OLD. Fig. 1.— Endothia tropicalis; Fig. 2.— E. parasitica; Fig. 3.— E. longirostris. 7 M.S.Haifl. CULTURES OF ENDOTHIA SPECIES mont^s> C rowt"of^XL; 5 m0nth ' S ^°^ ° f ***>**» Parasitica; b. corn meal after i Pm?-P^T ° n » P *' iffht ™^ S 0f «r«-meal agar 6 weeks o- ENDOTHIA PARASITICA AND RELATED SPECIES. 41 (PI. XXI, fig. 1) in field work when fructifications of the species are wanting or doubtful. Aside from the differences in color, the most conspicuous and important characteristic of these fungi in corn-meal cultures is found in the fructification. Clinton (18, pi. 26) has already men- tioned and illustrated similar differences in cultures of these organ- isms on agar in Petri dishes. In Endothia parasitica the pycnidia and spore masses are small, numerous, thickly scattered, and em- bedded in the mycelium. E. fuens, on the other hand, forms few, large, erumpent stroma ta, with spores extruding in thick, elongated masses. E. tropicalis closely resembles E. parasitica in number, size, and arrangement of pycnidia and spore masses, but differs in color of mycelium. E. fluens mississippiensis appears somewhat inter- mediate between E. parasitica and E. fluens in regard to the character and abundance of the pycnidia and in color of the growth. These peculiarities have been very uniform and constant in all the cultures on this medium and if they could be coordinated with regular mor- phological differences in nature would justify the separation of this form as a species. (See Pis. XIX and XX.) CULTURES ON LIQUID MEDIA (IN 100 C. C. FLASKS). Some difficulty was experienced at first in growing the species of Endothia satisfactorily on a liquid medium. Abundant growth was obtained on a medium suggested by Dr. Mel. T. Cook. This is a modification of the liquid medium No. II as given by him (19). Cook's liquid medium, No. II, is prepared as follows : Into 500 c. c. of distilled water put 15 grams of glucose and 20 grams of peptone steamed at 100° C. for three-fourths hour ; into another 500 c. e. of distilled water put 0.25 gram of dipotassium phosphate and 0.25 gram of magnesium sulphate, steamed for 20 minutes ; filter both 500 c. c. into same receptacle, steam 10 minutes, put into flasks, about 30 c. c. in each flask, and autoclave. All species grew readily on this medium, Endothia parasitica even producing pycnospores. At the end of one month's growth the sev- eral species were readily distinglished on this medium and may be briefly described as follows: Endothia gyrosa. — Growth scanty ; did not form a continuous mat, but re- mained in small bunches, giving an almost flocculent appearance. The mycelium appeared white when removed from the culture solution, but the solution itself was honey yellow. Endothia singularis. — Growth even less abundant than E. gyrosa; formed small brown knots against the glass. Mycelium buff, and the medium was changed to honey yellow. Endothia fluens. — Growth somewhat more abundant and less closely matted than E. parasitica, entirely submerged ; mycelium white ; liquid unchanged in color. 42 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTUEE. Endothia fluens mississippiensis. — Growth slightly less abundant than in E. parasitica; submerged except at the very edges; much lighter in color, being reddish brown. Endothia tropical is. — Tins differed markedly from either E. parasitica or E. fluens. The mycelium formed a thin felt over the surface, white to salmon orange in color, with no change in the medium. Endothia parasitica. — Mycelial growth very abundant, closely matted, chiefly submerged, but slightly arborescent in one or two small areas, which remained above the surface. Color, dark greenish brown. CULTURES ON STERILIZED TWIGS (IN TUBES). Early in this work it was noted that all the species of Endothia grew readily on sterilized chestnut twigs in test tubes. Later, tests were made with twigs from a number of common, woody plants. Twigs of Acer saccharum, Alnus rugosa, Betula papyri/era and B. lenita, Carpinus caroliniana, Cornus florida, Fagus grandifolia, Fraxinus a?rbericana, Ostrya virginiana, Populus grandidentata, Primus serotina, Rhus glabra, Tilia americana, and Tsuga canadensis were collected in New York State early in June, placed in test tubes with a few 7 cubic centimeters of distilled water and sterilized in an autoclave. All the species of Endothia were tested, and all grew on every species of twig except Tsuga. The difficulty of completely de- scribing this series may readily be seen from the fact that each species of Endothia had a different appearance on every kind of Avood. In general it may be stated that Endothia gyrosa and E. singularis grew more slowly than the other species and produced no spores, while all the other species produced spores on most hosts. The mycelium of E. parasitica was usually white, especially on the bark. E. gyrosa and E. singularis produced various shades of buff, while E. fluens, E. fluens mississippiensis, and E. tropicalis developed a much more brightly colored mycelium, usually showing yellow or orange shades. MOISTURE RELATIONS. In an earlier paper (77, p. 7) the writers reported tests with Endothia fluens and E. parasitica on media containing various per- centages of water. It was observed that pycnospore production began earliest and was most abundant on the media containing the least moisture. Aside from this the writers have thus far been unable to make definite tests as to the moisture relations of these fungi. However, incidental observations in connection with the light tests (p. 43) and temperature tests (p. 45), as well as results of field experiments, particularly those at Woodstock, N. Y., make it apparent that the amount of available moisture is a very important factor in the fructification of the fungus. ENDOTHIA PARASITICA AND RELATED SPECIES. 43 LIGHT RELATIONS. The relation of light to pycnospore production in Endothia para- sitica was first discussed by Anderson (1, p. 20). He says — When plate cultures are grown in total darkness on chestnut-bark agar, no pycnidia are developed, while on plates made at the same time and grown in the light, the usual rings of pycnidia appear (fig. 57). Experiments were also tried in which the plate was left in darkness until about half covered with mycelium and then brought into the light. Circles of pycnidia were developed, beginning with the ring which marked the outermost limit of the colony when removed from the dark chamber. The concentric rings which always appear on agar cultures are due to the alternation of night and day. Later, in a bulletin by Anderson and Rankin (6, p. 592), the same results are attributed to D. C. Babcock. Up to the time the above-mentioned work was published the writers had grown about 3,000 cultures of the several species of Endothia on various media in flasks and tubes. Practically all of these cultures had been kept in dark cases and Endothia parasitica had produced pycnidia abundantly on most of the media used. It seemed desirable, therefore, to determine whether wholly different light relations existed when the fungus was grown on plates. The following series of tests was accordingly made, using E. parasitica only. LIGHT TESTS OF CULTURES ON PLATES. In experimenting with plate cultures in order to check up the results reported by Anderson and Rankin (6, p. 592) it was noted that there was great variation in the rate at which the cultures dried out. There was considerable variation in this respect in dif- ferent plates kept side by side, apparently due to differences in the Petri dishes, and a marked difference between cultures kept in light and those kept in darkness. Since a causal relation between lack of moisture and abundant spore production had already been shown, it seemed probable that this might influence the results of the light tests in plate cultures. In fact, in a few cases the cultures kept in the light did produce spores earlier than those kept in dark- ness. Accordingly, in order to eliminate at least in part this fact which seemed to obscure the possible effect of light, a method was sought of equalizing the loss of moisture. In the following series half the plates were placed under a plain bell jar and the other half under a bell jar of equal size but darkened by being covered inside and out with heavy black paper, such as is used to wrap photographic plates. The two bell jars were then set side by side in front of a north window. By this means the conditions were made much more uniform as to temperature and moisture. There was still a slight difference in the rate of drying and undoubtedly at times a difference 44 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. in the temperature of the light and dark plates, but probably not sufficient to interfere seriously with the experiments. Series 1. On corn-meal agar plates under bell jars. — In nine days there was no distinguishable difference between the plates in light and darkness, a few spore masses occurring near the middle of each. In 18 days most of the light plates showed a central ring of spore masses and a zone of scattered spore masses near the edge. The dark plates showed a few small spore masses near the center, and scat- tered about the outer portion were the small masses of mycelium which usually constitute the early stages of pycnidial formation. In 30 days the number of spore masses had increased somewhat in both sets of plates, but more in the darkened plates, so that the number of spore masses was about equal in all the plates. The two sets of plates were fairly uniform as to the arrangement of the spore masses. Plate XVIII shows a typical example. Series 2. On chestnut-twig agar plates under bell jars. — After nine days the cultures in light and darkness were alike. No spores had yet appeared in either set. In 30 days there were a few spore masses on nearly all of the plates, there being no difference between those in light and those in darkness either in number or distribution. Series 3. On corn-meal agar and chestnut-twig agar under bell jars. — In this test the plates were piled alternately, first a corn-meal and then a chestnut-twig agar plate, so that the two media would be under conditions as nearly identical as possible. The plates were inoculated as before and left untouched for 18 days and after that were examined daily. After 18 days all the corn-meal plates showed spore masses in practically equal numbers, while the chestnut-twig agar plates showed no spore masses whatever. There was no ap- parent difference in the growth on either medium between the plates in light and those in darkness. At the end of 25 days the cultures on chestnut-twig agar plates showed numerous small masses of mycelium, indicating the forma- tion of pycnidia. No difference was perceptible between the dark and light plates. In 28 days, from 100 to 150 of these pycnidia in each plate were extruding spore masses. The light plates showed in general a larger mass of spores than the dark plates, but this was not marked, cer- tainly no greater than was accounted for by the unavoidable dif- ference in radiation and the consequent difference in moisture. This difference in the moisture of the medium was clearly shown each morning by the greater amount of moisture condensed on the covers of the darkened Petri dishes. At this time (after 28 days) four corn-meal agar plates which had been wrapped in four layers of heavy black photographic paper and ENDOTHIA PARASITICA AND RELATED SPECIES. 45 placed on a window sill were opened and examined. In spite of the cold weather prevailing during this test and the consequent low tem- perature of the room at night, these plates contained an average of nearly 200 well-developed spore masses. At the end of 35 days the chestnut-twig agar plates which had been kept in the light showed an average of 160 spore masses, while those kept in darkness showed an average of 130 spore pustules, a comparatively small difference in favor of the light plates. There was, however, a wide difference between the various plates in each series, and it was impossible in most cases to distinguish cultures grown in the light from those grown in darkness either by the num- ber, size, or arrangement of the pycnidia and spore masses. From these experiments it is evident that pycnidia are produced abundantly in total darkness on chestnut-twig agar as well as on other favorable media. There is no perceptible difference in the amount of spore production or in the arrangement of pycnidia be- tween cultures kept in total darkness and those kept in the light during the day if the temperature and evaporation remain the same in both. Continued observation of numerous cultures grown both in daylight and in darkness has convinced the writers that light has no perceptible effect on mycelial growth either in amount, nature, or color production. It seems evident, therefore, that light is a neg- ligible factor in the growth and fructification of these fungi. TEMPERATURE RELATIONS. In an earlier paper (77, p. 9) the writers published the results of three series of tests made to determine the temperature relations of three species of Endothia. Since the publication of that paper cul- tures of other species and additional material of some of the species from widely separated localities have been secured. Four series of temperature tests including this new material were made on solid media. TESTS ON SOLID MEDIA. In these tests cultures of Endothia gyrosa, E. singulars, E. ftuens, E. -ftuens mississippiensis, and E. parasitica were tried on corn-meal agar in slanted tubes, oatmeal in flasks, and potato agar in slanted tubes. The cultures tested were from specimens chosen from the extremes of the known ranges of the fungi and from their different hosts. No difference could be detected in the various cultures of the same species, even in those from widely separated localities and from different hosts. Cultures appeared to have the same temperature relations wdiether made from spores or mycelium. The results may be briefly summarized as follows: At 41° and 39° C. there was no growth in any species. Cultures removed from the incubator at the end of 11 days and kept at room temperature showed no growth. 46 BULLETIN 380, U. S. DEPAETMENT OF AGKICULTURE. At 35° C, Endothia gyrosa, E. singularis, and E. parasitica showed a slight development within 2 days, but at the end of 11 days it was still slight and abnormal in appearance. E. fluens and E. fluens mississippiensis showed no growth at this temperature. At 31° C, Endothia gyrosa, E. singularis, and E. parasitica appeared about the same as at room temperature for the first four days. At the end of six days these species showed somewhat less growth than at room temperature, while at the end of two weeks the growth was less in extent and markedly less freshly colored than that at room temperature. E. fluens and E. fluens mississippiensis showed somewhat less growth than at room temperature even in 4 days, and markedly less at the end of 2 weeks. At room temperature (which at this time varied from 20° to 24° C. ) the growth was much as described in the previous paper. Within 11 days growth was practically complete and in 14 days there was abundant spore production in Endothia parasitica. At 18° and 16° C, all species showed considerably less growth than at room temperature, but there seems to be little difference in the comparative growth of the various species at these temperatures. At 13° the growth was decidedly less than at 16° C. but was fairly normal in appearance in all the species except that Endothia fluens mississippiensis failed to produce the characteristic color at this temperature. At 9° C. there was a very slight growth in all species. At 7°, 5°, and 2° C. there was no growth whatever. Cultures removed to room temperature at the end of 11 days developed normally and at about the same rate as in newly made cultures. These additional tests seemed to confirm the results already pub- lished (77, p. 27) ; that is, growth was best in all species at ordinary room temperature, about 20° to 24° C. The minimum temperature for all was about 9°, and all failed to grow at 7° C. The maximum temperature for Endothia gyrosa, E. singularis, and E. parasitica appeared to be about 35°, while the maximum for E. fluens and its variety E. -fluens mississippiensis was apparently about 32° C. At all the temperatures tried E. singularis grew much more slowly than any of the other species. It was noted that cultures kept at 7°, 5°, and 2° C. showed no growth, but when removed to room temperature developed normally, while cultures kept at 41° and 39° C. failed to grow when removed to room temperature. This seemed to indicate that the fungi are more susceptible to heat than to cold, and such is perhaps the case. There was, however, the additional factor of moisture involved, for while the agar of the cultures kept at 7° and lower w T as in apparently the same condition at the end of 11 days as when first inoculated, the agar of the cultures kept at 41° to 39° C. was considerably dried. This raised the question as to whether the drying out of the agar had not affected the grow T th of the fungi in those cultures kept above room temperature as much as the higher temperatures themselves. The same idea was suggested by the fact that several of the species grew for a few days at 31° C. as well as they did at room temperature, and then fell behind. It seemed possible that this falling off in the ENDOTHIA PARASITICA AND RELATED SPECIES. 47 rate of growth might be due, at least partly, to more rapid drying of the agar at 31° C, or possibly to the more rapid development of some toxin, as was suggested by Balls (7) to explain a similar observation on the " soreshin " fungus. These observations threw doubt upon the accuracy of the writers' previous conclusions, and made it seem possible that the optimum temperature of the species of Endothia might be well above room temperature. This could only be determined accurately by some method which would control tem- perature without altering the supply of moisture. Some months after the above tests were concluded it was discovered that the various species of Endothia would grow readily on several liquid media. Consequently, several series of tests on liquid media were run parallel to those described above, except that the tests were continued for only four days. Experiment showed that at the higher temperatures the medium became considerably reduced by evaporation if left for a longer period. TESTS ON LIQUID MEDIA. In the series of tests on liquid media, all the species of which cul- tures had been obtained were grown on Cook's medium (see p. 41) both in tubes and in flasks, using ten tubes and six flasks at each temperature. The cultures of Endothia gyrosa and E. singularis were made with bits of mycelium from pure cultures. The other species were grown from conidia and the cultures were kept for two days at room temperature, in order to allow the conidia to germinate before being placed in the temperatures to be tested. The following temperatures were used for making the tests: 40°, 37.5°, 35°, 29°, and 27°, and room temperature which was fairly constant at about 22°, 17°, 12°, 9°, 7°, 3°, and 2° C. There was some variation in the temperature of the incubators and refrigerators used, but in most cases they did not vary more than 1 degree above or below the temperature indicated. xVt 40° there were occasional traces of growth, especially in Endothia parasitica, but this may have occurred when the incubator dropped to 39° C. There is no regular and continued growth at this temperature. At 37.5° C. there was perceptible growth in all the species. This is in striking contrast to the results on solid media, as no species grew at a temperature above 35° C. on solid media. At 35° C. Endothia parasitica showed practically the same amount of growth as at 27° and 29° C. for the first three days, but fell behind after that. E. fluen-s showed less growth at 35° than at the lower temperatures. These two species were the only ones tested at 35° C. At 27° and 29° C. growth was markedly more abundant than at 37.5°, and in most of the species was more abundant than at room temperature. In Endothia gyrosa and E. fluens mississippiensis the 48 BULLETIN 380, U. S. DEPARTMENT OF AGEICULTUEE. growth at 27° C. was apparently equal to that at room temperature. At 22° C. (room temperature) all species developed much more rap- idly than at the lower temperatures. At 17°, 12°, and 9° C. there was progressively less and less growth. At 7° C. and lower there was no growth whatever. While these tests are not wholly satisfactory and must be regarded only as approximations, they are of some interest. Below 7° C. there is no growth in any species. It is evident that there is a considerable range of temperature, from below 20° to well above 30° C, within which the species of En- dothia grow readily. Within this range there may be a definite optimum for each species, but this has not yet been determined. For Endothia parasitica the optimum appears to be at 27° C. or above, and the same may be true of the other species. At 40° C. or above no growth occurs. There is considerable evi- dence, however, that Endothia flu-em is less resistant to the higher temperatures than either E. 'parasitica or E. gyrosa. After several of the tests the flasks were kept at room temperature for some days. It was found that all developed normally except those which had been kept at 40° and 37.5° C. These developed more slowly than those which had been kept at lower temperatures. It was particu- larly noticeable also that E. parasitica and E. gyrosa developed prac- tically as well after being kept at 40° as at 37.5° C, while cultures of E. fluens which had been in 37.5° developed fairly well; but if kept at 40° for three days they entirely failed to develop. DISTRIBUTION OF THE SPECIES OF ENDOTHIA. During the past two years the writers have studied over 600 speci- mens of Endothia from various parts of the world. The greater number of these specimens have naturally come from the United States. The maps (figs. 1-4) show the known ranges of the various species in this country. Each dot on a map represents a locality from which the species has been collected. Frequently, of course, many specimens have come from a single locality ; hence the number of dots by no means represents the number of collections. In the case of Endothia parasitica, the dark portion represents the area over which the blight is practically continuous; that is, practically all the stands of chestnut are either diseased or dead. The dots represent known isolated infections and the solid line marks the botanical limit of the chestnut. Endothia gyrosa. is known only from the United States, but has a range in this country wider than that of any other species. As shown in figure 1. it has been found as far north as central Michigan, east to Connecticut, on the Pacific coast near San Francisco, and on ENDOTHIA PARASITICA AND EELATED SPECIES. 49 the Gulf of Mexico. There is, however, a very great difference in the abundance of this species at different points. In the southeastern United States — that is, the region south of central Indiana and southern Virginia and east of central Arkansas and Louisiana — this species occurs in great abundance wherever its hosts are found. Broken branch stubs and exposed roots of Liquidambar, Fagus, and Quercus are covered with fructifications of this fungus. This is especially true of roots exposed by erosion or excavation which have suffered mechanical injury through the tramping of men or cattle. Farther north in Maryland, New Jersey, and Connecticut only an occasional specimen is found. Three days' search in southern Con- h4y^^^ \ J 2 ^-— ^~2^«C7~' s\ — Or *\ jLj ( (/v-J^ I J \ \* i\ Sjw K . \ V \r — /~— \ / • TW^* ^; ^£ .',j* ,— j. ^L^r'f** X l Mki ^£ | ^y .■sjj ... I,.*- »* life Sii i ife- i ^, '- '■v'.-^KS .-tvk^.:^. ■:£.. ^v.- , ; ■ ; Fig. 1.— Japanese Chestnut at Nikko, Japan, from Which the Chestnut Blight Fungus (Endothia parasitica) was Collected by F. N. Meyer, on September 17, 1 915. Fig. 2.— Two Branches of a Japanese Chestnut. The Larger (to the Left) Was Brought to This Country by F. N. Meyer, and from It Endothia parasitica Was Isolated. ENDOTHIA PARASITICA AND RELATED SPECIES. 57 As Tachingko is 300 miles south of Changli, where E. parasitica was first collected by Meyer, and Yatyeko is 500 miles west of Tachingko, it seems highly probable from the collections that E. parasitica is widely distributed in China (fig. 5). Meyer, writing from Hangchow, July 1, 1915, refers as follows to the condition of the chestnuts in that locality : Well, I have a few interesting discoveries to report. First, there are many specimens of Castanea mollissima scattered at the bases and on the lower slopes of the hills around here, and these chestnuts are seriously attacked by the bark fungus, and in my estimation are going to succumb to it these coming years. The chinquapins (Castanea spp. ), however, which are very abundant on the higher and more sterile hill slopes, seem to be immune ; Fig. 5. — Outline map of China and Japan, showing the localities in which Endotliia parasitica has been found. at least I did not see any evidences of damage or even of attacks. This brings another interesting point to my mind. I was told in Nanking that various missionaries at Ruling, the great summer resort in central China for missionaries, were cutting down their chestnuts, as the tops w T ere all dying, due to borers working underneath the bark. Meyer has since stated to the writers that he believes the de- struction of the chestnut at Ruling is due to Endothia parasitica rather than to borers. In the writers' earlier publication the following statement was made (76, p. 297) : The Chinese organism has thus been shown to be practically identical with the American in all its morphological and physiological characters and in the production of the typical chestnut blight and the pyenidial fructifications 58 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. of the fungus. There is apparently but one other requirement that could be made according to the strictest pathological canons to perfect the proof in this case, and that is the production of typical ascospores of E. parasitica on the lesions produced by the inoculations. The last requirement has now been fulfilled. Specimens collected February 15, 1915, from inoculations made September 20, 1913, on chestnuts in Virginia, near Point of Rocks, Md., with Chinese ma- terial, show perithecial stromata with typical ascospores of E. 'par- asitica, thus completing the evidence. DISCOVERY OF ENDOTHIA PARASITICA IN JAPAN. More than two years after his original discovery of Endothia parasitica in China (June 3, 1913), Meyer also discovered the fungus in Japan. A brief account of his discovery has already been pub- lished by the writers (78). It may be sufficient here to state that fol- lowing the discovery of Endothia parasitica in China the writers endeavored by correspondence to obtain the fungus from Japan. While not successful in obtaining Endothia parasitica, the writers did receive several specimens of fungi, including species of Endothia on species of Castanea. These, together with several specimens of fungi found on chestnut nursery stock from Japan, make it clear that there are in that country several Pyrenomycetes other than Endothia parasitica more or less parasitic on Castanea. Meyer first discovered the chestnut-blight fungus in Japan at Nikko, September 17, 1915, on wild trees of Castanea crenata Sieb. and Zucc. A photograph of the trees from which he collected speci- mens of Endothia parasitica is shown in Plate XXIII, figure 1, and a branch from which the diseased material brought to the United Sates was taken is shown in Plate XXIII, figure 2. Shortly after Meyer's arrival in Washington in December, 1915, the specimens collected at Nikko were turned over to the writers for study. Examination at once showed cankers and mycelial fans typical of Endothia parasitica. The material also contained typical pycnospores and ascospores of the fungus. Cultures made from single ascospores on various culture media proved to be identical with those of Endothia parasitica found in this country and in China, thus establishing beyond question the identity of the fungus. Meyer's observations as to the resistance of the Japanese chestnuts to this disease are of great interest. He states that the trees vary considerably as regards their power of resistance, but that in general the Japanese chestnut is even more resistant to Endothia parasitica than is the Chinese chestnut (Castanea rnollissima) . As announced in the same publication (78), Endothia parasitica was collected by Dr. Gentaro Yamada at Morioka, northern Japan. These specimens, which show typical cankers as well as ascospores of the fungus, were received by the writers on January 8, 1916. ENDOTHIA PARASITICA AND RELATED SPECIES. 59 PRESENT DISTRIBUTION OF ENDOTHIA PARASITICA IN AMERICA. The present range of Endothia parasitica in America, as shown by the map (fig. 4), is probably merely the extent to which it has been able to spread in the time since it was first introduced. Whether Endothia parasitica was introduced into one locality or several is uncertain, but the studies of Heald (40, 41) and others have shown clearly that the spores of E. parasitica are carried by the wind, by insects and birds, and on nursery stock, which would account for its wide distribution and for its occurrence in isolated localities, long distances away from the main body of the disease. It also makes it seem probable that the fungus will continue to spread with some rapidity. Certainly, there is no evidence that any factor, climatic or other- wise, is likely to prevent the spread of this fungus into the large area of chestnut south of its present range. On the contrary, the duplicate inoculations made by the writers show clearly that the fungus grows more rapidly at the southern limit of its present range than farther north, where it is much more common. The longer growing season in the South is also no doubt an important factor. In this connection, it may be noted that Koppen (46), in his map of the vegetation regions of the earth, places the portion of China where Endothia parasitica has been found indigenous in the same climatic region as that portion of the United States where it is now doing such destructive work. He designates this region as the " Hickory " division of the mesotherms. HOST RELATIONS OF THE SPECIES OF ENDOTHIA. ENDOTHIA GYROSA. Endothia gyrosa occurs commonly on Liquidambar, Fagus. and Quercus, occasionally on Castanea, and has been found on Vitis in Alabama, but the writers were unable to obtain fresh material from this host. While Fagus and Quercus are, of course, closely related, it seems remarkable that a fungus should be abundant on hosts so different as Liquidambar and Quercus, yet so rare on any other host as to be only once reported. It seemed possible, indeed, that the fungus on Liquidambar. while morphologically and culturally identical with that on the various other hosts, might prove to be physiologically dif- ferent. In order to obtain more definite information on this point, several series of cross inoculations were made. It had been observed that Endothia gyrosa was found most fre- quently on the cut or broken ends of branches or on exposed, bruised, or broken roots. In making inoculations, therefore, a small branch, 60 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. 1 inch or less in diameter, was cut off about 6 inches from the main trunk. Mycelium from corn meal in flasks was placed on the cut end of the stub and covered with wet cotton, over which oiled paper was tied. In about two weeks the paper and cotton were re- moved. In all cases, branches similar to those inoculated were cut as checks. Tablk III. — Inoculations with Endothia gyrosa. Source of fungus and date. Host inocu- lated. Number of inocu- lations. Number success- ful. Remarks. Fagus: May 8, 1913 . . , May 29, 1913 . Sept. 15, 1913. Do Apr. 2, 1914 . Do Do May 23, 1914 . Do Do Do Quercus: May 29, 1913. Do Sept. 15, 1913. Do Apr. 2, 1914 . Do Do May 23, 1914 . Do Do Do Castanea: May 29, 1913 . Do Apr. 2, 1914 . Do Do. May 23, 1914 . Do Do Liquidambar: May 29, 1913 . . Do Sept. 15, 1913. Do. Do Apr. 2, 1914 . , Do Do May 23, 1914 . Do Do Do Castanea . Fagus — Liquidambar Quercus Fagus — Quercus.. Castanea . ....do Liquidambar Quercus x Fagus ....do Liquidambar ....do... Castanea . Fagus — Quercus l Castanea . ....do... Liquidambar Quercus ' Fagus ....do Liquidambar Fagus... Quercus ' Castanea . .do. Liquidambar Quercus > Fagus Liquidambar Fagus Castanea . Quercus Fagus Quercus ' Castanea do Liquidambar Quercus ' Fagus Pycnospores first observed on Oct. 16. Pycnospores first observed on Aug. 29 for two and on Oct. 10 for the third. No growth until the spring of 1914; pyc- nidia scattered and small on Oct. 13. No growth until spring; well developed on Oct. 13, 1914. Pycnidial stromata well developed on Oct. 13, 1914. Do. Do. Pycnospores first observed on Aug. 29, 1913. Very slight indications of growth on Aug. 29, 1913; a few pycnidia with spores on Oct. 16. Large well-developed pycnidia on Oct. 13, 1914. Large abundant pycnidial stromata on Oct. 13, 1914. Abundant well-developed pycnidial stro- mata on Oct. 13, 1914. Pycnospores first observed on Aug. 29, 1913. Slight indications of pycnidial formation on Aug. 29, 1913; pycnospores on all on Nov. 17, 1913. Large well-developed pycnidial stromata on Oct. 13, 1914. Scattered, fairlv well-developed pycnidia on Oct. 13, 1914. Abundant well-developed pycnidia on Oct. 13, 1914. Pycnospores first observed on Aug. 29. No evidence of growth until the spring of 1914; pycnidia few T and small on Oct. 13. No growth until the spring of 1914; pyc- nidia small on Oct. 13. Abundant pycnidia on Oct. 13, 1914. i The species used in this case was Quercus prinus, which proved to be an exceedingly unfavorable host for Endothia gyrosa. ENDOTHIA PARASITICA AND RELATED SPECIES. l 61 Inoculations with Endothia gyrosa were also made on numerous hosts from which it had never been reported. Six or more inocu- lations were made on each host, in the manner described above, ex- cept that a part of each series was left unwrapped. The following inoculations showed no growth whatever: Those made in Virginia, April 4, 1914, on Cornus florida, Fraxinus americana, Juglans cinerea, Ilex opaca, Sassafras variifolium; in Maryland, April 17 and 22, 1914, on Carya glabra, Comus florida, Liriodendron tulipifera, Nyssa sylvatica, Sassafras variifolium, and Quereus alba; and in New York, July 11, 1914, on Betula alba, Primus serotina, Populus trem- uloides, Rhus glabra, Salix sp., and Sassafras variifolium. On Acer pennsylvanicum and Carya two out of the six inoculations developed a few stromata. These were found only on the tissue injured by the cut and there was no evidence of parasitism. On Castanea, Fagus, Quereus, and Liquidambar, however, a branch inoculated as described above dies back rather faster than the checks. This would indicate, as suggested by Clinton (18, p. 419), that E. gyrosa is a weak parasite ; that is, that it is able to invade injured and dying tissue. It is evident from Table III that Endothia gyrosa coming from any of the four hosts named will, under favorable circumstances, grow on any of the others. Several other interesting facts are brought out by the table. Inoculations made with material from Liquidambar grew in general more rapidly on Liquidambar than on any of the other hosts. In many cases, material from Liquidam- bar failed to grow on Castanea, Fagus, and Quereus, and even when inoculations were successful growth was somewhat slower and pyc- nidial production less abundant. On the other hand, inoculations from Fagus, Quereus, and Castanea usually grew less rapidly on Liquidambar than on any of the other three hosts. This is, of course, what would be expected from the systematic relationships of the host species, and while the inoculations made are too few to permit any definite conclusions they are nevertheless suggestive. As shown by Table III, Quereus prinus proved a very unfavorable host for Endothia gyrosa. In all cases inoculations made in the fall (Sept. 15) failed to show any growth until the following spring. This corresponds with the results in inoculations of Endothia parasitica, but it is, of course, impossible to determine whether this failure to grow is due to the dormant condition of the host or to unfavorable weather con- ditions. Perhaps correlated with the results just noted are the unusually poor results obtained from inoculations made in the early spring. It will be noted that inoculations made on April 2, 1914, were in general much less successful than those made on May 23, 1914, in exactly the same locality and in many cases on the same hosts. 62 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. ENDOTHIA SINGULARIS. The material of Endothia singularis distributed by Sydow as Calo- pactis singularis was on Quercus gambellii Nutt. The writers have seen abundant material on this species as well as specimens on Q. utahensis (A. DC.) Rydb., Q. leptophylla Rydb., and Q. nitescens Rydb. Specimens on the latter two hosts were sent by Bethel, who, in a letter, reports finding this species also on Q. pungens Liehm. All of these species except Quercus leptophylla are chaparral- forming shrubs growing at an elevation of 4,000 feet or more. There is at present no evidence that the fungus is parasitic on any of the species. Inoculations with the mycelium of Endothia singularis were made on Fagus and on Quercus alba, Q. velutina, Q. rubra, and Q. palustris, as well as on Q. ilicifolia on Overlook Mountain in the Catskills. No growth has, however, been noted in any case. ENDOTHIA FLUENS. When these investigations were commenced, the writers thought that the Endothia found in Europe might be the same as Endothia parasitica found in America. Inoculations were accordingly made in Maryland during October, 1912, with cultures from material col- lected on the chestnut by the senior writer at Stresa, Italy, and Etrembieres, Switzerland, using material of E. fluens sent by P. J. Anderson from Pennsylvania ; also material of that species and of E. parasitica collected in Virginia as checks. In this case, as in all others where no special mention is made of the method, inoculations were made by cutting through the bark to the wood with a sharp knife. The inoculating material was then inserted with a freshly cut twig and the wound tied up either with cord or rubber bands. If cord was used it was cut away within two to four weeks. The rub- ber bands became loosened by exposure to the weather within about the same time. Inoculations were made with all the above material on sprouts of Castanea dentata and Quercus prinus. The results are summarized in Table IV. Table IV. — Inoculations of Endothia in Maryland in October, 1912. Fungus. Host inoculated. Number of inocu- lations. Number showing growth. Endothia parasitica. Do E. fluens: European American Do Castanea dentata. Quercus prinus. . . Castanea dentata. do Quercus prinus •28 14 L'3 9 ENDOTHIA PARASITICA AND RELATED SPECIES. 63 The inoculations were examined every 10 days until December 1 and monthly thereafter throughout the winter. There was no per- ceptible growth until the last of April, when several of the inocula- tions of Endothia parasitica showed slight sunken areas. By May 20 all inoculations checked as showing growth (last column of table) showed the slight yellowish elevations of the bark which indicate the beginnings of pycnidia. On August 30 all the inoculations of E. parasitica checked as showing growth had spread rapidly and attacked the living tissues of the host, producing typical cankers with mycelial fans and abundant pycnidia. No signs of growth were noted in the inoculations of Endothia fiuens until about the middle of May, 1913, when most of them showed signs of pycnidium formation. By August 30 all those marked as showing growth had produced characteristic pycnidia with spores, which when cultured proved to be typical E. fiuens. In no case, however, did this fungus spread for any appreciable distance beyond the injured portion or show signs of active parasitism. These results agree with those given by Anderson and Anderson (2, p. 206) with American material of E. fiuens, and have since been fully con- firmed by further observation. During the summer of 1914 about 1,100 inoculations of Endothia fiuens from both European and American sources and of E. fiuens mississippiensis were made on Castanea sprouts. In no case was there any evidence of active parasitism, as in E. parasitica. Although Endothia fiuens has been found in Europe on a con- siderable number of deciduous host plants (as recorded on p. 18), the writers have thus far failed to find it in this country on any except Castanea and Quercus. It seemed possible that the European strain of the fungus might be somewhat more plurivorous 1 in its habits than the American. In order to throw some light on this point, the following inoculations were made : On March 31, 1914, 10 inoculations were made, half of European and half of American material, at Francis, Md., on the following hosts: Alnus rugosa, Betula nigra, Carpinus caroliniana, Carya glabra, Fagus grandifolia, Lirio- dendron tulipifera, and Liquidambar styraciflua. Pycnidia appeared only on Carya glabra and Carpinus caroliniana. Of the inoculations which actually produced pycnidia, four on Carpinus and three on Carya, one of each was the European strain. On April 22 inoculations were made with American material of E. fiuens at Kensington, Md., on Acer rubrum, Carya glabra, Cornus florida, Fagus grandi- folia, Prunus serotina, Quercus prinus, Sassafras variifolium, Vaccinium sp., 1 This term is proposed to apply to fungi occurring on two or more hosts or substrata and may be applied to all fungi except true parasites. It is derived from plus (plur-), more, and vorare, to devour. Compare omnivorous already in use for fungi. The term pleioxonous might be derived from De Bary's proposed word pleioxony and applied to true parasites having the power to invade more than one species of host plant, and the term plurivorous restricted to nonparasitic organisms. 64 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. and Titis sp. Of these, Acer rubrum and Carya glabra gave numerous small pycnidia. On July 10 the following hosts were inoculated at Woodstock, N. Y., with JJ. flu ens from Europe: Acer rubrum, A. pewnsylvanicum, Carya ovata, Corylus americana, Fraxinus americana, Hamamelis virginiana, Kalmia latifolia, Popu- lus grandidentata, Prunus serotina, Rhus glabra, Salix sp., Sassafras variifo- liion, and Syringa vulgaris. Each host was inoculated in six or seven places, hut all failed to develop except two inoculations on Acer pennsylvanicum and one on Corylus americana. The results cited above are so largely negative that they prove very little except that the European strain shows no special affinitj' for these hosts in America. ENDOTHIA FLUENS MISSISSIPPIENSIS. Only five collections of Endothia fiuens mississippiensis have thus far been made, three on Castanea dentata and two on Quercus sp. From the results of the inoculations its host relations appear very similar to those of E. -fiuens. The results are; shown in Table V. Table V. — Inoculations with Endothia fiuens mississippiensis on Castanea and Quercus. Source of culture. Host inoculated. Date. Number of inocu- lations. Number showing pycnidia. Jan. 20,1912 May 8, 1913 .....do... 8 4 4 9 12 12 8 Do do 4 Do do 4 Do do 7 Do Apr. 18,1914 do. 10 do 10 The inoculations of January 20, 1912, showed no signs of growth until early in May, when the first signs of pyenidium formation were - observed. The inoculations with Endothia fiuens mississippiensis made May 8, 1913, showed within three weeks discolored areas near the cut which were larger than those about the check cuts. On July 25, 1913, all of the inoculations of E. fiuens mississippiensis marked " successful " showed the beginnings of pyenidium formation. By August 30, 1913, they were producing pyenospores, which when cul- tured proved to be E. fiuens mississippiensis. Inoculations were made in April, 1914, for the purpose of com- paring the material collected on oak with that collected on chest- nut. No difference was detected, and there was no indication of active parasitism. This form behaved in this respect exactly as did the E. fiuens from Virginia both on Castanea dentata and Quercus prinus. A series of inoculations parallel to that made with E. fiuens was made with E. fiuens mississippiensis. The same hosts were used, and in most cases the dates and places of the inoculation were the same. The results of all that showed any growth are given in Table VI. ENDOTHIA PABASITICA AND RELATED SPECIES. 65 Table VI. — Inoculations with Endothia fluens mississippiensis on Acer and Carya. Location. Host. Number of inocu- lations. Number showing pycnidia. Woodstock, N. Y 6 6 6 1 2 Do „ 2 Francis, Md 1 Kensington, Md 1 Do o As in Endothia fluens the growth was confined to the injured tis- sues, and there was no evidence of parasitism. ENDOTHIA TROPICALIS. The material of Endothia tropicalis from which the writers se- cured their cultures, was collected by T. Petch in Ceylon. As the species of Endothia in the Northern Hemisphere are chiefly on members of the Fagacese, Petch's statements with regard to hosts are of considerable interest. In a letter of March 6, 1914, he writes: We have no Fagacere native in the island. We have introduced various species of Quercus and Castanea, but subsequent to Thwaite's discovery of this fungus. I do not think there can be any doubt that the fungus is native to Ceylon * * * Of the specinients now sent * * * those in the packet * * * are from a tree which was producing shoots from the base. This tree is Elaeocar- pus glanduiifcr Mast. From the bark and habit, I believe that all my " finds " of Endothia have been on this species. In the accounts of the American chestnut disease, I notice that several authors speak of " cankers," and give their rate of growth. I never see "cankers" (Krebs) on the Ceylon trees. The bark appears to die regu- larly and smoothly from above downward, and is quite unbroken except for the minute cracks through which the stromata emerge. Inoculations. — As already noted, ascospores of Endothia tropi- calis resemble those of E. parasitica even more closely than do those of E. fluens. This fact, together with its similarity on culture media and its oriental origin, led the writers to fear possible parasitic tendencies. Inoculation experiments were accordingly made only on the chest- nut and under carefully guarded conditions. In all, about 30 inoculations were made on 2-inch chestnut sprouts, using the methods described for other species. Of 25 inoculations made in May and June, practically all had de- veloped a few pycnidial stromata by October 20. These stromata were a somewhat brighter orange than those of E. fluens or E. fluens mississippiensis, and the spores when cultured produced typical E. tropicalis. In no case, however, was there any evidence of parasitism. 43737°— Bull. 3S0— 17 5 66 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. ENDOTHIA PARASITICA ON HOSTS OTHER THAN CASTANEA. The first collection of Endothia parasitica on a host other than Castanea of which the writers have any knowledge is that made by J. Franklin Collins at Martic Forge, Pa., June 30, 1909. As an- nounced by Dr. Metcalf at the Boston (December, 1909) meeting of the American Phytopathological Society, the specimen consisted of a small dead branch of Quercus velutina with several spore tendrils typical of E. parasitica. This material, which consisted of a terminal branch with leaves still retained, was at once sent to the laboratory at Washington, and cultures obtained from it were sub- sequently used in making numerous inoculations on Castanea dentata on Long Island, N. Y., in July, 1909. On November 17 of the same year, Metcalf reported that the inoculations were entirely successful and had produced typical lesions, thus establishing without question the identity of the fungus. Fulton (37, p. 53) reports E. parasitica on the dead bark of Quer- cus alba and Quercus velutina, but found no evidence that the fungus produces in any sense a disease of such trees. Clinton (18, p. 428) mentions cultures from three different species of Quercus and (p. 376) reports specimens on Quercus alba, Q. rubra, and Q. velutina. Anderson and Babcock, as quoted by Anderson and Rankin (6, p. 564), found Endothia parasitica on Quercus velutina, Q. alba, Q. prinus, Rhus typhina, Acer rubrum, and Gary a ovata, but it seemed parasitic only on Quercus alba. They made inoculations with mate- rials isolated from Castanea on Quercus prinus, Q. velutina, Q. alba, Q. coccinea, Rhus typhina, Acer rubrum, Liriodendrontulipij 'era, and Carya ovata. Two trees of Rhus were girdled and killed by the growth of the fungus. On Quercus alba the fungus seemed slightly parasitic, but none of the trees were killed. The fungus grew and produced spore horns on the wounded tissue near the point of inocu- lation on all the hosts except Acer and Liriodendron. Rankin (62, p. 238) also made inoculations with Endothia para- sitica from Castanea on Quercus prinus, Q. rubra, Q. alba, and Q. coccinea. He found that the mycelium advanced into the living tissues for a short distance in a few cases, but that in no case were typical cankers formed. Pycnidia were produced abundantly on the injured tissues of all the hosts. During the course of this work only four specimens of Endothia parasitica on hosts other than Castanea have come to the writers. One was on chestnut oak (Quercus prinus) collected by F. W. Besley, at Towson, Md., December 26, 1911; one from Quercus velutina, at Germantown, Pa., as well as one from white oak (Quercus alba), at Kennett Square, Pa., were collected by S. B. Detwiler; and one from dead maple, Acer sp., at Florence, Mass., by Roy G. Pierce. ENDOTHIA PARASITICA AND RELATED SPECIES. 67 The specimen collected by Besley on Quercus prinus showed the fan- shaped mats of mycelium typical of E. parasitica on Castanea spe- cies. The fungus had apparently girdled the tree. The specimen on Quercus alba, collected by Detwiler, was similar to one on Quercus prinus in appearance and came from a dead tree which had appar- ently been killed by the growth of the fungus. The specimens on Acer sp. and on Quercus alba were received in the spring of 1914, and cultures isolated from them were used in making inoculations for the purpose of determining whether the fungus had either lost or gained in virulence by passing through other hosts. INOCULATION EXPERIMENTS. The cultures secured from Acer and Quercus, together with one made from Castanea at about the same time, were inoculated into three separate sprouts of Acer rubrum, Castanea dentata, and Quer- cus prinus. The sprouts chosen were of nearly the same size, 2 inches in diameter, and similarly situated, and each was inoculated in five places, with two check cuts above. The inoculations were made the usual way on March 31, 1914, and were examined at least once a month during the summer. None of the inoculations on Quercus produced any growth what- ever. On Acer the inoculations with the culture from Quercus all failed to develop; one of the inoculations with the culture from Acer showed a few pycnidia, while four of the inoculations with material from the chestnut developed a few pycnidia. On Castanea the three series of inoculations were almost identical, every inocula- tion producing a typical canker. Of course, these inoculations are too few to be conclusive, but it is evident that there was no decrease in virulence on the chestnut in passing through Acer or Quercus and that no particular affinity for either Acer or Quercus was gained. On the maple, in fact, the culture direct from chestnut produced the most growth. In addition to those listed above, numerous inoculations were made in order to determine whether Endothia parasitica had any parasitic tendencies on other deciduous hosts. These inoculations were all made during the spring of 1914 by the usual method of cutting well through the bark and inserting mycelium and spores from a pure culture, usually on corn meal. The wounds were then wet, some bound with wet cotton, others with paraffin paper, and about half were left unwrapped. Seven or more inoculations were made on April 4 in Maryland on Alnus rugosa, Betula nigra, Carpinus caroliniana, Fagus grandifolia, Kalmia latifolia, Liriodendron tulipifera, and Liquidambar styraci- fiua, none of which developed. Inoculations were also made on April 68 BULLETIN 3S0, U. S. DEPARTMENT OF AGRICULTURE. 22 in this locality on Acer rubrum, Carya glabra., Cornus forida, Fagus grandifolia, Liriodendron tulipifera, Quercus primes, Sas- safras variifolium, Vaccinium sp., and Vitis sp. without success. On April 18, the following hosts were inoculated in Virginia : Acer rubrum, Betula nigra, Benzoin aestivale, Carpinus caroliniana, Carya glabra, Cornus florida, Fagus grandifolia, Liriodendron tulipifera., Primus serotina, Quercus alba, Ulmus americana, and Vitis sp. Each host was inoculated in from four to six places. Of these, pyenidia were produced only on Acer rubrum, Carpinus, and Lirio- dendron. A similar series was made on the same hosts in the same place on May 27. Inoculations on one tree of Quercus alba showed undoubted evidence of parasitism and is described below. On July 9 and 11 from five to fourteen inoculations were made on each of the following hosts at Woodstock, N. Y. : Acer rubrum, Betula alba, Carya ovata, Fagus grandifolia, Frascinus americana, Hamamelis virginiana, Juglans cinerea, Kalmia latifolia, Nyssa syl- vatica, Ostrya virginiana, Populus grandidentata, Prunus serotina, Rhus typhina, Quercus rubra, Salix sp., Sambucus canadensis, and Sassafras variifolium. Pyenidia appeared on Acer rubrum and Ostrya only. The fungus made considerable growth on two plants of Rhus typhina, partly girdling branches one-half inch in diameter and producing distinct fans. The fans were, however, much smaller than those usually found in Castanea. Inoculations were made at Avon, Conn., July 15, on Acer saccharum, Betula alba, Carya glabra, Cornus florida, and Ostrya virginiana. Pyenidia developed only on Ostrya. The successful inoculations with Endothia parasitica are shown in Table VII. Table VII.- -Successful inocttlations in 191% with Endothia parasitica on hosts other than Castanea. Locality. Date. Host. Number of inocu- lations. Number success- ful.! Virginia Do Do Do New York... Do Connecticut.. Apr. 18 ..do ...do May 27 July 11 ..do July 15 Acer rubrum Carpinus caroliniana . . . Liriodendron tulipifera . Quercus alba Acer pennsylvanicum. . Ostrya virginiana do 1 Inoculations producing pyenidia are classed as successful. It must be noted that while pyenidia were produced in the cases listed as successful, there was no indication of parasitism, nor did the growth extend beyond the tissue injured by the cut except in Quercus and Rhus. Out of about 400 inoculations with Endothia parasitica on hosts other than Castanea, about TO of which were made on different I ENDOTHIA PARASITICA AND RELATED SPECIES. 69 species of Quercus, chiefly Q. prinus and Q. alba*, only one case has been noted in which the fungus assumed a typically parasitic role. The data in this case may be summed up as follows : Four inocula- tions were made May 27, 1914, on a small tree of Qmrcus alba. This tree was suppressed, and although when cut down it showed about 30 annual rings it was only 16 feet high and about 2 inches in diam- eter. It was in a moist, shady locality close beside a stream, and in spite of its small size was apparently healthy. The inoculations were made in the usual way from a culture of E. parasitica on corn meal. On August 1 it was noted that all four inoculations were pro- ducing pycnidia, and in at least one case typical fans had been developed. On October 15 all four cankers had more than half girdled the seedling. No observations were made during the winter, but at the time the leaves had reached half the normal size, in the spring of 1915, the tree was completely girdled. On July 1 this tree presented an appearance closely similar to that of a small chestnut tree girdled by Endothla parasitica. All the leaves above the point of inoculation were dead and remained attached to the branches. Below the girdled portion, water sprouts had developed, as has been frequently described for chestnut trees affected with E. parasitica. Cultures made from this tree showed the fungus to be typical of E. parasitica. Whether this case of parasitism was due to unusual virulence on the part of the fungus or to unusual susceptibility on the part of the host is, of course, merely a matter of conjecture ; the latter alternative seems, however, somewhat more probable, as other inoculations with this strain of the fungus on Q. prinus and Q. alba failed to show similar results. In addition to the above, a somewhat similar observation has been made by the writers near Amherst, Mass. In connection with other work, a sprout of Quercus prinus about an inch in diameter was inoculated with Endothla gyrosa on July 15, 1911. When this inocu- lation was made the tree was partly (about one-fourth) girdled. E. gyrosa developed normally and by October 1, 1914, had produced several pycnidial stromata. No change was apparent when the inocu- lations were examined in May, 1915. E. parasitica was abundant in the region, however, and apparently gained entrance through the cuts originally made, for when the plat was next visited, August IT, 1915, the sprout appeared quite dead, though still retaining its full-sized dead leaves. Further examina- tion showed numerous pycnidia of E. parasitica in addition to those of E. gyrosa near the region of the original inoculation. The pycnidia of E. parasitica were on all sides of the stem, while those of E. gyrosa were confined to the portion above the cuts made in inoculating. The mycelial fans typical of E. parasitica were abundant also. These 70 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. observations leave no doubt that the tree was girdled and killed by E. parasitica, Endothia parasitica in exceptional cases undoubtedly attacks other hosts than Castanea, producing cankers and sometimes causing the death of the host. The results of the inoculations just recorded appear to indicate that some unusual conditions of host or parasite must obtain in such cases. Whether such a combination of conditions or factors will ever become sufficiently frequent to lead to serious destruction of Quercus or other forest trees remains to be determined. ENDOTHIA PARASITICA ON CASTANEA SPP. Although found occasionally on species of other genera, Endothia parasitica is dangerously pathogenic only on members of the genus Castanea. The parasitism of this fungus on the American chestnut {Castanea dentata) was first proved by Murrill (57) and has since been demonstrated by numerous investigators. When Endothia parasitica was discovered in the United States it was considered by some investigators to be a native fungus which had suddenly become parasitic, and various theories were advanced to account for the supposed unusual susceptibility of the host. As enumerated by Clinton (18, p. 391), the factors suggested include winter injury, drought injury, fire injury, weakened condition due to continued coppicing, and reduced amounts of tannic acid due perhaps to weather conditions. Continued study by many investigators in different localities has, however, fully confirmed the observation originally made by Met- calf and Collins in 1910 (53) that " a debilitated tree is no more subject to attack than a healthy one " and that Endothia parasitica is actively parasitic on the healthiest specimen of Castanea dentata in case there is opportunity for wound infection. The writers have personally made over 1,200 inoculations of E. parasitica on Castanea dentata without finding a single individual that showed any re- sistance. CASTANEA ON LIMESTONE SOILS. Not only are all trees susceptible, but so far as is known no con- dition of soil, altitude, or moisture renders them more resistant to the disease. The idea has been held by some writers that chestnuts grown on limestone soils were immune to the disease, and the plant- ing of chestnut orchards on such soils was advocated. This view is held by Gulliver (38, p. 53), who sums up his observations in two regions in Pennsylvania as follows: In every series of tracts taken from limestone to overlying shale soils, the percentage of blight is least at a comparatively short distance * * * from the edge of the limestone. Tracts on soils derived from limestone which show the highest percentage of blight seem to be those where the soil has ENDOTHIA PARASITICA AND RELATED SPECIES. 71 become acid from underground drainage. Chestnut trees on soils derived from other alkaline rocks show less blight than is found in the trees on shale soils with limestone underneath. On the other hand, Detwiler (24, p. 67) reports observations in the Lizard Creek valley which seem to show that these relations do not always occur. He says — A belt of limestone borders Lizard Creek valley on the south, and the per cent of infection is as high in that region as elsewhere. Infection centers have been found near limestone quarries, where the roots of the chestnut pene- trated to bedrock. Actual proof or disproof of the truth of this idea was peculiarly difficult, since chestnut is but rarely found growing naturally on calcareous soils. During the summer of 1914, however, a careful study of the chestnut on certain portions of limestone areas in west- ern Maryland and western Connecticut was made. These localities were chosen because they were convenient in connection with other work, the blight had been present for several years in both States, and thorough State geological surveys made the location of the lime- stone areas very eas}^. The two States also are sufficiently far apart to eliminate sources of error that might arise from local weather conditions. In western Connecticut chestnut was abundant on glacial till over the Stockbridge limestone of this region. Chestnut was also grow- ing directly over limestone at various points near Danbury, Twin Lakes, Chapinville, and Lakeville. Several localities near the latter place were kindly pointed out by Dr. George E. Nichols. Near Dan- bury every tree examined showed the blight in a more or less ad- vanced stage, while near the other towns, all in the northwest corner of the State, nearly 50 per cent of the trees were blighted. About 30 inoculations were made on sprouts in this region, and all except two developed cankers quite as rapidly as did check inocula- tions made on the trap ridge west of Hartford. Chestnut is very rare on the Shenandoah limestone in the Hagers- town and Frederick valleys of western Maryland. A number of chestnut trees were, however, located growing on limestone soil near Frederick Junction and Adamstown in the Frederick valley. The disease was already established west of Adamstown, where 20 per cent of the chestnuts were either diseased or dead. Twenty-two in- oculations were made on nine chestnut sprouts in these two regions, and all developed typical cankers quite as rapidly as the checks made in similar sprouts growing over Baltimore gneiss 50 miles east. RECESSION OF THE CHESTNUT IN THE SOUTHERN STATES. While it has been definitely proved that Endothhi 'parasitica is pathogenic on healthy chestnut trees, one of the points brought for- 72 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. ward by the advocates of the " weakened host " theory seems to be fully established; that is, that the chestnut trees have suffered se- verely in the southern Appalachian regions previous to the present epidemic, in some cases being practically exterminated, so that the range is now considerably less than formerly. The evidence on this point has been summarized by Clinton (18, pp. 408-413). Various writers quoted by him cite fire injury and borers and other insects as causes for this recession. Long (48, p. 8) considers a root rot due to ArmAllaria mellea as " very probably an important factor in the gradual recession of the chestnut" in North Carolina. It seems probable that all of the above-mentioned factors, and perhaps others, have played a part in the destruction of the chestnut in this region. RELATIVE SUSCEPTIBILITY OF SPECIES OF CASTANEA. The importance of Castanea dentata as a timber and nut tree and its abundance in eastern North America, where the blight is preva- lent, has made the chestnut blight an object of much investigation. Descriptions of the nature and importance of the disease, the rate of its spread, methods of distribution, and attempted methods of control have been given in detail by Anderson (1-5) , Clinton (12-15) , Heald (39-41), Metcalf (51 and 52), Metcalf and Collins (53), Kan- kin (62), and others. It may be sufficient here to state that the fungus enters the host through a wound in the bark, probably never or very rarely through lenticels or natural cracks, grows chiefly in the cambium, penetrating for only short distances into the wood, and kills the tree or branch by girdling. Once a tree is attacked, it is only a question of time till it succumbs. The chinquapin (Castanea pumila) was found by Murrill (58) in 1908 to be attacked b}^ Endothia parasitica. Rogers and Gravatt (65) in 1915 made inoculations of E. parasitica on C. pumila and found that the parasite grew as rapidly on this host as on C. dentata. They attribute the apparent resistance of the chinquapin to its com- parative freedom from bark injury, a view also held by other writers. Pantanelli (60) and Metcalf (52) have proved that the European chestnut is readily susceptible to the disease. The only chestnuts thus far observed which show any resistance to Endothia parasitica are those of oriental origin. Metcalf (51) first pointed out the resistance of the Japanese chestnut. This observation has since been confirmed hy Clinton (18, p. 375), who " failed to produce the disease in a Japanese variety in the [Conn.] station 3^ard, although the bark was inoculated in 16 different places." Van Fleet (84), in describing the spread of the chestnut blight in his breeding plats at Washington, D. C, says (p. 21) : " The Asiatic chestnuts and the chinquapin-Asiatic hybrids are plainly highly resistant." ENDOTHIA PARASITICA AND RELATED SPECIES. 73 Morris (56) sums up eight years' observation of the effect of the chestnut blight on 2G species and varieties of chestnuts at Stamford, Conn., as follows: Every one of the 5,000 American chestnut trees became blighted * * * None of [the grafted varieties or seedlings of European and Asiatic varieties appear] to be as vulnerable as the American chestnut, but most of mine are now dead. Korean chestnuts and chestnuts from the Aomori regions in Japan resisted the blight until six years of age. Since that time they have shown a marked tendency to blight, but resist it better than does the American chestnut * * * None of the American species of chinquapin * * * has blighted with the exception of two limbs * * * None of the specimens of Castanca alnifolia [or] * * * of Castanca mollissima has blighted, but these latter include only five trees. These observations as to the resistance of the oriental varieties of chestnut when grown in America are of particular interest in con- nection with the observations of Meyer in the region where he dis- covered Endothia parasitica native. In his letter to Fairchild, writ- ten from Santunying, China, June 1, 1913, Meyer makes the following notes with reference to the effect of the blight in that region : This blight does not by far do as much damage to Chinese chestnut trees as to the American ones * * * Not a single tree could be found which had been killed entirely by this disease, although there might have been such trees which had been removed by the ever-active and economic Chinese farmers * * * Dead limbs, however, were often seen and many a saw wound showed whore limbs had been removed * * * The wounds on the majority of the trees were in the process of healing over * * * Old wounds are to be observed here and there on ancient trees. Meyer's photographs taken near Santunying substantiate his state- ments. Certainly no specimens of C. dentata in a blight-infested region in this country could survive to the age of the Chinese chest- nuts shown in his photographs. That the Chinese chestnuts are by no means uniformly resistant, however, is clearly shown by Meyer's later notes. On the label of a package of Endothia parasitica collected on chestnut at Tachingko, Shantung, China, March 21, 1911, he writes, "Trees very severely attacked, many dying off," and in a letter written from the same place he says, " A serious canker ; many of the trees here were killed by it." Further evidence that the virulence of Endothia parasitica on Chi- nese chestnut differs in different parts of China is found in subse- quent communications from Meyer. From a point near Chingtsai, Chekiang, China, on July 15, 1915, he writes: "All around Hang- chow and west of it one finds the chestnut trees seriously attacked by this destructive bark fungus." On July 11, 1915, near Changhua, Chekiang, China, he com- ments, "With the exception of near Taianfu, Shantung, chestnuts 74 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. are much more severely attacked in the Chekiang Province than either in Chihli, Shansi, or Shensi. Could the greater humidity of central China be of assistance to a more vigorous development of this destructive fungus?" COMPARISON OF HOST RELATIONS. It will be seen from the above description of the host relations of the various species that while some other members of the genus (E. gyrosa, e. g.) may have slight parasitic tendencies, Endothia parasitica alone is an active parasite. The contrast is still more striking in the section of the genus to which E. parasitica belongs, for E. fluens and E. fluens mississippiensis, which resemble E. para- sitica so closely in their morphological characters, and to a less degree on culture media, and are common on Castanea, are almost purely saprophytic. This fact is established by the work of Ander- son (2), Clinton (18), and others, and by two years' field observa- tions and several thousand inoculations made by the writers and their colleagues. The host relations of the parasite are equally striking. Although Endothia parasitica is so pathogenic on Castanea dentata that this tree has been practically exterminated over several hundred square miles of its natural range and its extinction is threatened, the fungus has been only occasionally found as even a weak parasite on the closely related genus Quercus, and never, to the writers' knowledge, on Fagus. During the course of this work the writers have been continually impressed with the possibilities of a physiological study of E. para- sitica and one or more closely related species which might throw some light on the fundamental question of the nature and cause of parasitism. No other case is known to the writers of a virulently parasitic fungus and a closely related purely saprophytic one which will grow readily and fruit on a large variety of artificial media, which are readily distinguishable on those media, and remain con- stant for hundreds of generations. SUMMARY. The pathological and economic importance of this group of fungi was first recognized when the chestnut-blight fungus was discov- ered in New York in 1904. This organism was first referred to the genus Diaporthe, but was later shown to belong to the genus Endothia. The specific identity, relationships, and native home of this para- site were at first uncertain. Some pathologists considered it a native organism which was attracting attention and causing injury chiefly ENDOTHIA PARASITICA AND RELATED SPECIES. 75 by reason of the weakened condition of the chestnut trees. Others believed it to be of foreign origin. Its recent discovery in China and Japan has settled this question. To determine positively the identity of the organism, a thorough study was made of the types or authentic specimens of all the species of Endothia obtainable. As a result of this work a revision of avail- able species of the genus is presented. This is based upon the field and laboratory study of over 600 collections. Over 4,000 cultures have also been studied. Endothia gyrosa (Schw.) Fr. is the type of the genus, which is naturally divided into two sections, chiefly by the character of the ascospores. In section 1 they are short, cylindric to allantoid, and continuous or only pseudoseptate. This section contains two species, E. gyrosa and E. singularis. Section 2 has oblong- fusiform to oblong-ellipsoid uniseptate as- cospores. This contains four species and one variety, Endothia fluens, E. fiuens mississippiensis, E. longirostris, E. tropicalis, and E. parasitica. E. tropicalis is a hitherto unrecognized species. Radiating layers of yellowish or buff mycelium situated in the bark and cambium of the host are found to be constant and dis- tinctive characteristics of Endothia parasitica. None of the other species studied shows this character. All species of the genus possess a stroma having a distinctive yellow to reddish color. There is no division of stroma into distinct layers, as described by some authors. Pycnidia or perithecia may arise in any portion of the stroma. Most commonly where pycnidia and perithecia are both present the pycnidia are above the perithecia, though the reverse arrangement is sometimes observed and all intermediate conditions frequently occur. The stromata of the species of section 1 are larger, more erumpent, and contain more numerous pycnidia than those of section 2. En- dothia singularis is especially striking in this respect. The stromata of section 2 are smaller and very similar in all the species. The pycnidia consist of more or less irregular chambers or locules in the stroma. The pycnospores are small in most species and furnish no very distinctive specific characters. The pycnospores of Endothia trop- icalis are, however, constantly larger and more variable in size than those of the other species. Paraphyses have been described by some authors, but have never been observed by the writers. The ascospores in the species of section 1 are very similar in size and shape. Those in section 2, though similar, have been found by thorough study and careful measurement to show constant though 76 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. slight differences, as indicated in the tables of measurements and ratios. Numerous cultures of all the species on a variety of media show that each species has constant and distinctive characters of growth and color. All the species grew equally well in light or darkness, and no de- cided differences in temperature relations have been demonstrated. The species appear to have well-defined geographic limits of distribution, which have been approximately determined for the American species. The distribution of the species does not coincide with that of the hosts, but seems to be determined in part by soil and climatic conditions. Endothia fiuens has the widest distribution, being frequent and widely distributed in Europe and the eastern United States, and also occurring in Asia. Endothia parasitica is evidently of oriental origin. Specimens have been received from five rather widely separated localities in China and from two localities in Japan. In the eastern United States it is now abundant from Maine to North Carolina and is rapidly spreading south and west. It has already destroyed most of the chestnut trees within a radius of 100 miles of New York City. The species have rather definite host relations. Endothia gyrosa has been found on five genera of plants, viz, Cas- tanea, Fagus, Liquidambar, Quercus, and Vitis. Endothia singularis occurs, so far as known, only on Quercus species. Endothia fiuens has been found in America only on Castanea and Quercus, but in Europe it occurs on Alnus, Carpinus, Castanea, Corylus, Quercus, and Ulmus, and has been reported on Aesculus, Fagus, and Juglans. Endothia fiuens mississippiensis has been found only on Castanea and Quercus. Endothia tropicalis is known only on Elaeocarpus. Endothia parasitica has been found on Acer, Carya, Castanea, Quercus, and Ehus, but at present is only known as a serious para- site on Castanea. Upon the American species of Castanea it is actively parasitic under all the conditions of soil and climate observed. Oriental species of chestnut are more or less resistant to the disease both in America and their native homes. None of the species except Endothia parasitica has thus far been found to be actively parasitic. LITERATURE CITED. (1) Anderson, P. J. 1914. Morphology and life history of the chestnut blight fungus. Com. Invest, and Control Chestnut Tree Blight Disease in Penn. Bui. 7, 44 p., 17 pi. 1913. (2) and Anderson, H. W. 1912. The chestnut blight fungus and a related saprophyte. In Phyto- pathology, v. 2, no. 5, p. 204-210. (3) 1912. Endothia virginiana. In Phytopathology, v. 2, no. 6, p. 261-262. (4) 1913. The chestnut blight fungus and a related saprophyte. Penn. Chestnut Tree Blight Com. Bui. 4, 26 p., 6 fig. (5) and Babcock, D. C. 1913. Field studies on the .dissemination and growth of the chestnut blight fungus. Penn. Chestnut Tree Blight Com. Bui. 3, 45 p., 14 pi. (6) and Rankin, W. H. 1914. Endothia canker of chestnut. N. Y. Cornell Agr. Exp. Sta. Bui. 347, p. 531-618, fig. 77-101, pi. 36-40 (1 col.). Bibliography, p. 611-618. (7) Balls, W. L. 1908. Temperature and growth. In Ann. Bot., v. 22, no. 88, p. 557-591, 11 fig. (8) Berkeley, M. J. 1836. British Fungi ... 386 p. London. (Smith, J. E. English Flora, v. 5, pt. 2.) (9) and Broome, C. E. 1875. Enumeration of the fungi of Ceylon. In Jour. Linn. Soc. [London] Bot., v. 14, p. 29-140, pi. 2-10. (10) 1877. Supplement to the enumeration of fungi of Ceylon. In Jour. Linn. Soc. [London] Bot., v. 15, p. 82-86, pi. 2. (11) Cesati, Vincenzo, and Notaris, G. de. 1S63. Schema di classificazione degli sferiacei Italici aschigeri ... In Comm. Soc. Crittog. ItaL, v. 1, p. 207, 240. (12) Clinton, G. P. Chestnut bark disease [1907]. In Conn. Agr. Exp. Sta. 31st/32d Ann. Rpt. [1906] /OS, p. 345-346. (13) 1908. Chestnut bark disease, Diaporthe parasitica Murr. [1908]. In Conn. Agr. Exp. Sta. 31st/32d Ann Rpt. [1906] /08, p. S79-S90. (14) 1911. Chestnut bark disease. In Conn. Agr. Exp. Sta. Bien. Rpt. 1909/10, p. 716-717, 725. (15) 1912. Some facts and theories concerning chestnut blight. In Penn. Chestnut Blight Conf. Rpt. 1912, p. 75-83, 9 pi. (16) 1912. Chestnut blight fungus and its allies. In Phytopathology, v. 2, no. 6, p. 265-269. (17) 1912. The relationships of the chestnut blight fungus. In Science, n. s., v. 36, no. 939, p. 907-914. 77 78 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. (18) 1913. Chestnut hark disease, Ehdothia gyrosa var. parasitica (Murr.) Clint. In Conn. Agr. Exp. Sta. Ann. Rpt. 1911/12, p. 359-453, pi. 21-28. (19) Cook, M. T., and Taubenhaus, J. J. 1911. Relation of parasitic fungi to the contents of the cells of the host plants. I. The toxicity of tannin. Del. Agr. Exp. Sta. Bui. 91, 77 p., 43 fig. (20) Cooke, M. C. 1878. The fungi of Texas. In Ann. N. Y. Acad. Sci., v. 1, no. 5/6, p. 177-187. (21) Cukrey. Frederick. 1858. Synopsis of the fructification of the compound Sphaeriae of the Hookerian herbarium. In Trans. Linn. Soc. London, v. 22, pt. 3, p. 257-287, pi. 45-49. (22) 1865. Supplementary observations on the Sphaeriae of the Hookerian herbarium. In Trans. Linn. Soc. London, v. 25, pt. 2, p. 239-262. (23) Curtis, M. A. 1867. ... A catalogue of the Indigenous and Naturalized Plants of the State. 158 p. Raleigh, N. C. (Geological and Natural History Survey of North Carolina, pt. 3. Botany.) (24) Detwilek, S. B. 1914. Observations on sanitation cutting in controlling the chestnut blight in Pennsylvania. Final Rpt. Penn. Chestnut Tree Blight Com. 1913, p. 63-73, 1 pi., 3 maps. (25) Earle, F. S. 1901. Some fungi from Porto Rico. In Muhlenbergia, v. 1, no. 2, p. 10-23. (26) Ellis, J. B.. and Everhart, B. M. 1892. The North American Pyrenomycetes. . . . 793 p., 41 pi. Newfield, N. J. (27) Fairchild, D. G. 1913. The discovery of the chestnut bark disease in China. In Science n. s., v. 38, no. 974, p. 297-299. (2S) Farlow, W. G. 1912. The fungus of the chestnut-tree blight. In Science, n. s., v. 35, no. 906, p. 717-722. (29) Faull, J. H., and Graham, G. H. 1914. Bark disease of the chestnut in British Columbia. Forestry Quart., v. 12, no. 2, p. 201-203. (30) Fries, E. M. 1882. Systema Mycologicum. ... v. 2. Lundae. (31) 1828. Elenchus Fungorum. ... v. 2. Gryphiswaldiae. (32) 1830. Eclogae fungorum. . . . In Linnaea, Bd. 5, Heft. 4, p. 497-553. (33) 1849. Summa Vegetabilium Scandinaviae. . . . sect. 2. Upsaliae. (34) Fuckel, Leopold. 1861. Enumeratio Fungorum Nassoviae. 126 p., 1 col. pi. Wiesbaden. (35) 1869-70. Symbolae Mycologicae. Beitrage zur Kenntniss der Rheinis- chen Pilze. 459 p., 6 col. pi. Wiesbaden. (Jahrb. Nassau. Ver. Naturk., Jahrg. 23/24.) ENDOTHIA PAEASITICA AND BELATED SPECIES. 79 (36) FUISTING, WlLHELM. 1867. Zur Entwickelungsgeschichte der Pyrenomyceten. In Bot. Ztg., Jahrg. 25, No. 23, p. 177-181; No. 24, p. 185-189; No. 25, p. 193-198; No. 39, p. 305-311. (37) Fulton, H. R. 1912. Recent notes on the chestnut bark disease. In Penn. Chestnut Blight Conf. Rpt. 1912, p. 48-56. (38) Gulliver, F. P. 1914. [Report of] geographic work. In Final Rpt. Penn. Chestnut Tree Blight Com. 1913, p. 52-53. (39) Heald, F. De F. 1913. The symptoms of chestnut tree blight and a brief description of the blight fungus. Penn. Chestnut Tree Blight Com. Bui. 5, 15 p., 16 pi. (40) Gardner, M. W., and Studhalter, R. A. 1915. Air and wind dissemination of ascospores of the chestnut blight fungus. In Phytopathology, v. 3, no. 6, p. 493-526. (41) and Studhalter, R. A. 1913. Preliminary note on birds as carriers of the chestnut blight fungus. In Science, n. s., v. 38, no. 973, p. 278-280. (42) Hitchcock, Edward. 1829. A Catalogue of Plants Growing without Cultivation in the Vicinity of Amherst College. 64 p. Amherst, Mass. (43) Hohnel, Franz von 1909. Fragmente zur Mykologie. IX. Mitteilung, Nr. 407 bis 467. In Sitzber. K. Akad. Wiss. [Vienna], Math. Naturw. Kl., Abt. 1, Bd. US, Heft 9, p. 1461-1552, 1 illus. (44) 1913. Fragmente zur Mykologie. XV. Mitteilung, Nr. 793 bis 812. In Sitzber. K. Akad. Wiss. [Vienna], Math. Naturw. Kl., Abt. 1, Bd. 122, Heft 2, p. 255-309, 7 fig. (45) Keefer, W. E. 1914. Pathological histology of the Endothia canker of chestnut. In Phytopathology, v. 4, no. 3, p. 191-200, 3 fig. (46) Koppen, W. P. 1901. Versuch einer Klassification der Klimate vorzugsweise nach ihren Beziehungen zur Pflanzenwelt. 45 p., 2 illus., tab., 2 fold. maps. Leipzig. (47) Livingston, B. E., and Livingston, Grace J. 1913. Temperature coefficients in plant geography and climatology. In Bot. Gaz., v. 56, no. 5, p. 349-375, 3 fig. (48) Long, W. H. 1914. The death of chestnuts and oaks due to Armillaria mellea. U. S. Dept. Agr. Bui. 89, 9 p., 2 pi. (49) Merkel, H. W. 1906. A deadly fungus on the American chestnut. In 10th Ann. Rpt. N. Y. Zool. Soc. 1905, p. 97-103, illus. 80 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. (50) Mekriam, C. H. 1895. Laws of temperature control of the geographic distribution of terres- trial animals and plants. In Nat. Geog. Mag., v. 6, p. 229-238, pi. 12-14 (col. maps). (51) Metcalf, Haven. 1908. The immunity of the Japanese chestnut to the bark disease. In U. S. Dept. Agr., Bur. Plant Indus. Bui. 121, p. 55-56. (52) 1912. Diseases of the chestnut and other trees. In Trans. Mass. Hort. Soc. 1912, pt. 1, p. 69-95. (53) and Collins, J. F. 1910. The chestnut bark disease. In Science, n. s., v. 31, no. 802, p. 748. (54) Montagne, J. F. C. 1834. Notice sur les plantes cryptogaines, recemment decouvertes en France, contenant aussi l'indication precise des localitgs de quelques esp£ces les plus rares de la flore francaise. In Ann. Sci. Nat. Bot, s. 2, t. 1, p. 295-307, 337-349, pi. 11-13 (2 col.). (55) 1855. Cryptogamia Guyanensis . . . In Ann. Sci. Nat. Bot. s. 4, t. 3, p. 91-144, pi. 5. (56) Morris, R. T. 1914. Chestnut blight resistance. In Jour. Heredity, v. 5, no. 1, p. 26-29, fig. 14-15. (57) Mureill, W. A. 1906. A new chestnut disease. In Torreya, v. 6, no. 9, p. 186-189, fig. 2. (58) 190S. The chestnut canker. In Torreya, v. 8, no. 5, p. 111-112. (59) Nowell, William. 1915. Diseases of lime trees in forest districts. Imp. Dept. Agr. West Indies, Pamphlet Series 79, 41 p., 2 fig., 5 pi. (60) Pantanelli, Enrico. 1912. Su la supporta origine europea del cancro americano del castagno. In Atti R. Accad. Lincei, Rend. CI. Sci. Fis., Mat. e Nat., s. 5, v. 21, sem. 2, fasc. 12, p. 869-875. (61) Petri, Ltjigi. 1913. Sopro una nuova specie di Endothia, E. pseudoradicalis. In Atti R. Accad. Lincei, Rend. CI. Sci. Fis., Mat. e Nat., s. 5, v. 22, sem. 1, fasc. 9, p. 653-658, 2 fig. (62) Rankin, W. H. 1914. Field studies on the Endothia canker of chestnut in New York State. In Phytopathology, v. 4, no. 4, p. 233-260, 2 fig. pi. 11. (63) Rehm, Heinrich. 1907. Ascomycetes exs. fasc. 39. In Ann. Mycol, v. 5, no. 3, p. 207-213. (64) Ridgway, Robert. 1912. Color Standards and Color Nomenclature. 43 p., 53 pi. Washing- ton. (65) Rogers, J. T., and Gravatt, G. F. 1915. Notes on the chestnut bark disease. In Phytopathology, v. 5, no. 1, p. 45-47. ENDOTHIA PARASITICA AND RELATED SPECIES. 81 (66) Rudolphi, Fr. 1829. Plantarum vel novarum vel minus cognitarum descriptiones. Decas secunda. In Linnaea, Bd. 4, Heft 3, p. 387-395. (67) Ruhland, Willy. 1900. Untersuchungen zu einer Morphologie der stromabildenden Sphae- riales auf entwickelungsgeschichtlicher Grundlage. In Hed- wigia, Bd. 39, Heft 1, p. 1-79, pi. 1-3. (68) Saccakdo, P. A. 1873. Mycologiae venetae specimen. In Atti Soc. Veneto-Trentina Sci. Nat. Padova, v. 2, fasc. 1, p. 53-96, pi. 4-5 (col.) ; fasc. 2, p. 97-264, pi. 6-17 (col.). (69) 1882-1905. Sylloge Fungorum, v. 1, 8, 17. Patavii, 1882, 1889, 1905. (70) 1883. Genera Pyrenomycetum Schematica Delineata. 8 p., 14 pi. Patavii. (71) 1906. Notae mycologicae. Series VII. In Ann. My col., v. 4, no. 3, p. 273-278. (72) Schweinitz, L. D. VON. [1822]. Synopsis Fungorum Carolinae Superioris . . . 105 p., 2 col. pi. n. p. (73) 1825. Description of a number of new American species of Sphaeriae. In Jour. Acad. Nat. Sci. [Phila.], v . 5, pt. 1, p. 3-16, 2 pi. (74) 1S32. Synopsis fungorum in America boreali media degentium ... In Trans. Amer. Phil. Soc, n. s., v. 4, p. 141-316, pi. 19. (75) Shear, C. L. 1912. The chestnut-blight fungus. In Phytopathology, v. 2, no. 5, p. 211- 212. (76) and Stevens, Neil E. 1913. The chestnut-blight parasite (Endothia parasitica) from China. In Science, n. s., v. 38, no. 974, p. 295-297. (77) 1913. Cultural characters of the chestnut-blight fungus and its near rela- tives. In U. S. Dept. Agr., Bur. Plant Indus. Cir. 131, p. 3-18. (78) 1916. The discovery of the chestnut-blight parasite (Endothia parasitica) and other chestnut fungi in Japan. In Science, n. s., v. 43, no. 1101, p. 173-176. (79) Sowerby, James. 1814. Colored Figures of English Fungi or Mushrooms. Sup. London. PI. 438 is pt. of pi. 420. (80) Sprengel, Kurt. 1827. Systema Vegetabilium. v. 4, pars 1. Gottingae. (81) Sydow. Hans, and Sydow, Paul. 1912. Novae fungorum species — VII. In Ann. Mycol., v. 10, no. 1, p. 77-85. (82) Traverso, G. B. 1906. Pyrenomycetae . . . In Societa Botanica Italiana, Flora Italica Cryptogama. pars 1, Fungi, v. 2, fasc. 1, p. 180-182. (83) Tulasne, L. R., and Tulasne, Charles. 1863. Selecta Fungorum Carpologia ... t. 2. Parisiis. 43737°— Bull. 3S0— 17 6 82 BULLETIN 380, U. S. DEPARTMENT OF AGRICULTURE. (84) Van Fleet, Walter. 1914. Chestnut breeding experience. In Jour. Heredity, v. 5, no. 1, p. 19- 25, fig. 9-13. (85) Winter, Georg. 1887. Die Pilze . . . 928 p. Leipzig. (Rabenhorst, Ludwig. Krypto- gamen-Flora . . . Aufl. 2, Bd. 1, Abt. 2.) (86) Zon, Raphael 1914. Meteorological observations in connection with botanical geography, agriculture, and forestry. In Mo. Weather Rev., v. 42, no. 4, p. 217-223, 1 fig. ADDITIONAL copies OF THIS PUBLICATION MAY BE PROCURED FROM THE SUPERINTENDENT OF DOCUMENTS GOVERNMENT PRINTING OFFICE WASHINGTON, D. C. AT 35 CENTS PER COPY : Gaylord Bros. Makers Syracuse, N. Y. PAT. JAN. 21,1908