ms iy Ay <5 ef a res ae sears aa: Seen Lee Se Caen eames es parent epee aeees 3 rank nae ae a x SEE 3 way cee Se ean aS Ee rae aa pat — Pees TE ores Syaeirs ess ws paar & oy LIBRARY Michigan State University MSU RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from yw your record. FINES will be charged if book is returned after the date stamped below. - Tomato leaflet badly infected with Cladosporium fulvum, Note that the dark colored growth covers nearly the entire lower surface of the leaflet. THE LEAF MOLD OF TOMATOES, CAUSED BY SLADOSPORTUM FULVUM CRE. THESIS FOR DEGREE OF M, 8. WALTER KENNETH MAKEMSON 1917 THE For helpful advice and assistance given, the writer wishes to extend his thanks to Dr. E. A. Bessey and to Dr. G. H. Coons, under whose direction the accompanying investigation was conduoted. Acknowledgment is also made to Miss Eugenia MoDaniels, of the Department of Entomology, and to Mr. J. E. Kotila, to whom the writer is indebted for assistance in making the accompanying photomicrographs and photographs. I. II. IIf. IV. Ve VI. VII. CONTENTS . The Leaf Mold of Tomatoes, caused by Cladosporium fulvum Cke. Introduction. (a) History and distribution of the disease. The Disease. (a) Economic importance. (b) Description. 1. Appearance on the leaves. 2. Appearance on the stems. o- Appearance on the fruit. 4. Appearance on blossoms. Etiology of the disease. (a) Previous work. (b) Formal proof of causation. lL. Pure culture. 2- Infection from pure culture. &- Reisolation and reinoculation. Taxonomy of Cladosporium fulvum. Morphology of the parasite. (a) Mycelium. (b) Conidiophores. (c) Conidia. Relation of parasite to host. (a) Infection and conidiophore production. 1. Type of infection. 2. Conidiophores. (b) Morbid anatomy of the host. Physiological studies. (a) Cultural studies. 1. Growth upon various media. 2. Effect of humidity. 5. Effect of temperature. 4. Effect of light upon spore germination. S- Relation of light to fungous growth. 6. Effect of reaction of culture medium. 7. Effect of air exclusion. (b) Translocation of starch. VIII.Dissemination of fungus. IX. Xe XI. Pathogenicity. (a) Artificial ftnoculation of blossoms. (b) Field inoculation. (c) Period of incubation. Life history of the organism. Immunity phenomena. XII. Control measures. (a) Spraying. 1. Resistance of the organiem to fungicides. (1) Germination and growth tests in moist ohambers, (a) Infeotion on sprayed and dusted plants. (b) Fumigation. 1. Formaldehyde gas. 6. Sulfur fumes. (o) Additional prophylactic measures. XIII. Summary. XIV, Literature oited,. THE LEAF MOLD OF TOMATOES, CAUSED BY CLADOSPORIUM FULVUM ORE. INTRODUCTION. Tomato Leaf Mold (caused by Cladosporium fulvun), probably the most serious parasitic disease of greenhouse tomatoes in Michigan, was first desoribed from specimens sent from South Carolina to Prof, M. ©. Cooke in England, (1883)*. The article containing this desoription gave the diagnoses of the new speoies contained in Ravenel's distributed sets otf specimens entitied "North American Fungu", ** Since 1883 the fungus oausing this leaf mold has been mentioned in various experiment station reports, agricultural publications and texts on Plant Pathology, but these reports, other than giving a description of the appearance of the fungus on the tomato plant, contain very little additional intormation to that first given in "Grevillea® in 1883. Nor is any efficient method of control given, most authors assuming that Bordeaux spray will effeotively check its spread. * fhe date in parenthesis refers to the literature oited at the close of this paper. ** The species in question was desoribed as follows: "Cladosporium fulvum. Effusum, lanosum, fulvun,. Hyphis ereotis, flexuosis, septatis, nodulosie, paroe ramosis, fulvis. Sporis elliptiois, unisepta- tis, vix constrictis, pallide fulvis, hyalinis (,O1 - 02 x 0045 mm), * -3- In the United States the leaf mold seems to be almost uniformly distributed over the country, especially where tomatoes are grown under glass. Halsted (1885) reported the disease as ooourring on tomatoes grown in the open air in the vicinity of Ames, Iowa. The fungus was desoribed in the Report of the U. S. Department of Agticulture for the year 1888, from material sent to the Department by a Mr. Wilson of Vineland, N. J., where the mold oocourred in the greenhouse, Bailey (1893) reported the fungus in greenhouses in New York State where it was not regarded as being a serious pest; Selby (1896) reported its ocourrence in Ohio; Rolfs (1898) in Florida; Orton (1904) stated that the disease was serious in Maryland and Ohio; Jones and Morse mentioned the ocourrence of the mold in Vermont in a disease survey for the same year; and Barre (1910) reported it as being prevalent in South Carolina. 1 have the verbal report of Mr. H. CG. Young of the Department of Botany at the Michigan Agricultural College that the leaf mold was a serious pest in greenhouses in the vicinity of Raleigh, N.C. in 1915. The earliest specimens from Michigan in the Agricultural College herbarium date back to 1898, since whioh time it has been reported from practically every part of the State where tomatoes are grown under glass, ~3— Abroad the fungus seems to have gained as wide a distribution as in Amerioa. Plowright (1887) reported the organism in England for the first time; Prillieux and Delacroix (1891) reported it in the Department of the North in France; Traverso (1890), in Italy; Briosi and Cavara (1898) at Pavia, Italy; Froggatt (1906) in New South Wales, Australia; Bos (1901) in Holland; Lind (1907-'09) In Denmark; Sorauer (1908) in Germany; M. T. Cooke and Horne (1908) in Cuba; Sohecohner (1910) in Austria, Just where the fungus originated is hard to determine but it evidently came from some part of the New World, probably the western side of South America to whioh the tomato plant is indigenous and where it has been in cultivation for more than three hundred years, At any rate, it is significant that the disease was first observed in a southern state of North Amerioa and when we consider its ease of dissemination and the rapidity with which it has epread to all parte of the world, it does not require a great deal of imagination to see how it could have developed along with the tomato plant in its unoultivated state and finally become a serious parasite of cultivated tomatoes, especially on the foliage of tender greenhouse plants. THE DISEASE. Eoonomic Importance. An acourate estimate of the loss caused by this tomato disease ie diffioult to make because the infection, under different oonditions, is so variable. In one greenhouse the disease may get an early start before the tomato vines have yet begun to bear and in such a oase, the loss may be the entire tomato orop, due to the exosedingly rapid epread of the fungus. Again, the fungus may not get a good start until the vines have set most of the fruit and in this condition, the plants may be able to withstand the fungus and mature the majority of the fruits. In either case, the loss is brought about by two faotors: the failure of the new fruit to set because the blossoms are blasted by the fungous attack and the failure of green fruit to grow to marketable sise. Prillieux and Delacroix (1891) stated that to tell which plants are diseased it is “only necessary to count the fruits, whose number is greatly diminished". One of the largest growers of greenhouse tomatoes in Michigan states that "if he could hold the vines below, there would be some sise to the fruit above", meaning that if the lower leaves, which are the first out and oonsequently tirst subject to the attaok of the fungus, could be kept green, there would not be much third olass or cull fruit. 5 = These oull fruits, borne on the upper fruit branches, were about the size of emall walnuts when the author visited the infected greenhouses and it is unlikely that they became larger as the foliage of the plants bearing them wae nearly dead. In another part of the greenhouses, where younger plants were growing, the other factor which causes even a greater loss of fruit could be observed. In this case, the last fruit branch had in almost every instance failed to set any fruit, due to blasting of the blossoms; the calyces, corollas and even young ovaries being infected. (See Fig. 1) This was on the 10th day of August, before the field grown tomatoes were ripe and while there was yet a good demand for the hothouse fruit. Some idea of the seriousness of the disease may be had when one considers the oapital invested by the grower mentioned above, These greenhouses were covered by 180,000 sq. ft. of overhead glass, under which were set 35,000 tomato plants. Not one plant of the 35,000 was found to be free from the disease and most of them were very badly infeoted, the older plants having at least 75 percent of the foliage killed at the date of visit, The younger plants still retained their badly intected foliage but all of the later blossoms were blasted. The plants were headed back at the third wire, or about six feet from the ground which gave on an average about five fruit bearing branches per plant. Fig. 1 - Blasted tomato blossoms, Only one : fruit has set on this fruit branch. Te Leaving out of consideration the stunting effect upon all developing fruits and considering only the loss from blasted blossoms and undersized fruit on the last fruit-bearing branch, a conservative estimate of the loss involved would be about 30 peroent of the total crop. This is a considerable amount as the owner of these greenhouses estimated that the crop, altho cut short, would yield a ten thousand dollar return, Growers over the state report infeotionse of varying seriousness but, as before stated, it is diffioult to estimate the percentage of loss which indirectly affeots the fruit by consuming the plant's food supply, destroying the leaf tissue, and blasting the blossoms. Members of the Office of Cotton and Truck Investigations report in the Plant Disease Survey Bulletin No. 1, Auguet 15, 1917, "Cladosporium fulvun observed in one large field in Southern Florida, causing about 30 percent injury, leavés turning yellow and dropping, Yield from this field as compared with adjoining fields was reduced about one third®. Description. Common termes which have been applied to this disease are "Tomato Leaf Mold", "Tomato Leaf Rust", “Leaf Spot", "Common Blight" and inaptly, "Soft Rot", or "Tomato Rot", Tomato Leaf Mold seems to be the most fitting term used to designate the disease sinoe the fungue presents a B= characteristic moldy appearance on the tomato leaves. This name is also less apt to be confused with the other diseases of the tomato. The term "Blignt*® (preferably Southern Blight) is usually associated with a disease of the tomato caused by Bacillus solan- acearum Erw, Smith, or with a disease of the tomato caused by Fusarium lycopereici and often called blight, especially in the South. "Leaf Spot*® is usually applied to the disease caused by Septoria lycopersici and the fungus can hardly be said to produce a rusty appearance on the host, hence the unsuitability of the term *"Rust*, Appearance on the Leaves. This is essentially a disease of the tomato foliage. On the under side of the tomato leaf newly infeoted areas show at firet a whitish downy growth which, compared to Ridgeway's oolor charts, is a pale olive-buff oolor. Figure 3 shows a leaflet greatly magnified, with the younger downy growth of the fungus at the margins of the leaflet. The growth below is usually accompanied by a yellowing of the oelis on the upper surface opposite the infeotion, This at first appeare as a rather indefinite spot whose pale yellow color gradually merges into the chlorophyll green of the leaf but it later turns a deep ochre-yellow color, When the leaf tissue is finally killed, it becomes a red-brown color Qe Fig. 3 - Leaflet magnified eight times to show the younger, downy, fungous rowth at the margins of the nfected area. -10—= above. As the infeotion develops, the fungous growth on the lower surface becomes a characteristic tawny- Olive color and the infected area increases rapidly in size. Intected leavee usually show a great many inteoted areas (See Fig. 3) so that ina very short time the entire lower surface of the leaf may be covered with the tawny-olive colored growth. (See Fig. 4). The leaf may, however, show only a few large infected spots or areas (See Fig. 5). Under certain conditions, the upper surface of the leaf may show @ scattering growth of the fungue but thie never becomes as heavy as on the lower side. (See Fig. 6, whioh shows infection on the lower side of the two leaflets shown above; the two lower leaflets were photographed from their upper surface, which is yellowed in the infected areas and bears numerous tufte of fungous growth ). The infeoted leaves soon commence to die, due to the growth of the fungus. Areas which were first inteocted are killed first and the fungous growth in these dead areas scmetimes becomes a beautiful purple color on the lower surface. Thus the lower surface of @® partly killed leaf may show a great many violet-purple colored areas, surrounded by the tawny-olive color of the later infeotions. Fig. 3 = Infeoted leaflet with many infeoted areas somewhat evenly spaced over the surface. Fig. 4 - The entire lower surtace of this tomato leaflet is covered with fungous growth composed of spores and conidiophores. a i Fig. 5 = Tomato leaf inteoted in - large separated sreas with Cladosporium fulvum Cke. -14—- Fig. 6 = Conidiophores on the upper surface of the two leaflets shown below. Note that the growth is scattered. Stomata are fewer on the upper epidermis of the tomato leaflet. -15- Plowright (1687) desoribed this color production, which Cooke (1883) did not mention in the original desoription, and suggested that the fungus be known as "Cladosporium fulvum, Cooke, var. violaceum; Hyphis conidileque violaceis®, Since physiological studies,(see section on Physiology), have shown that the oolor is not due to a difterent variety of Cladosporium but to the action of certain factors upon the fungus itself, the writer does not believe the erection of a separate variety tor the violet colored fungus justifiable. Leaves are infected as they develop on the plant, consequently the lower leaves are the firet to be killed. The fungus attacks the new growth as it is produced so that a diseased plant oharaoteristioally has killed and infeoted leaves on the lower portion of the stem with the newly produced foliage above as the only part remaining green and uninjured. Growers of greenhouse plants “top* them or “head them baok® when they become five or six feet tall; therefore, we may find that no part of the plant is free from disease after a short time, since good greenhouse practice favors the removal of all new growth subsequent to the "heading back® process. Where White Fly is present in a greenhouse, a saprophytic mold known as Cladosporium herbarum may often be found. This mold is not a parasite on the -16— tomato plant but obtains ite subsistance from the honey- like exudate deposited upon the tomato leaves, stems and fruit by the White Fly. Cladosporium herbarum may be easily differentiated from the parasitic Tomato Leaf Mold in four ways: (1) Cladosporium herbarum usually ocours upon the upper surface of the tomato leaf, Cladosporium fulvum is found in greater abundance on the lower side. (3) Cladosporium herbarum is altogether superficial, i.e. the entire growth may be scraped off of the foliage or fruit and the latter found to be un- injured while Cladosporium fulvum myoelium is found to be ineide the leaves or foliage and cannot be removed. (3) Cladosporium herbarum does not cause a discoloration of the leaf tissue; Cladosporium fulvum does. (4) Clad- osporium herbarum presente a dirty, greyish-black appearance on the plant; Cladosporium fulvum is a tawny- Olive color (see Figs. 7 and 8.. The under side of the leaf showing C. herbarum is intected in three small areas by C, fulvum but shows none of the C, herbarum growth). Appearance on The Stems. Plowright (1887) described the fungus attack on the stems in the following way: "Not only are the leaves covered by the Cladosporium but the stems have also succumbed, long brown lines and elongated patches being present upon then". a3 7= Fig. 7 = Cladosporium fulvum on leaflet at left, Cladosporium herbarum on leaflet to the right. The Cladosporium herbarum growth is seen to be greater on the upper surface of the leaflet than the growth of Cladosporium fulvum at the left, also upon the upper surface. -18- ~N Q.guivern under side Fig. 8 = Cladosporium fulvum shown on under side of leaflet at left, also in two small patches on leaflet at right which is infected with C, herbarum on the upper surface, Note that Cladosporium herbarum does not show in under surface of leaf where honey dew exudate of the white fly is not deposited, -19- Massee (1910) described the fungus as attacking the stems,and Leaflet #263 of the Board of Agrioulture and Fisheries (1913) desoribed it upon the tomato stem as follows: "The fungue forme long, rust colored afterwards blackish streaks on the stem and more or less oiroular, scattered patohes on the fruit". The writer has found the fungus present abundantly on the young stems of the fruit branches, also on young side branches where it apparently had spread from an inteotion starting on the foliage at the end of the branch but no case has been observed where main stems of the plants were intected, Certainly main stem infection, if it does ocour, is rare in Miohigan as no infection of this kind was observed on over 50,000 intected plants in greenhouses visited. Furthermore, many attempts to intreoct artitiocially large tomato stems, both by sowing spores on cut and on uninjured areas, reeulted negatively. Appearanoe on Fruit. Haleted (1885) reported a destructive fruit rot of the tomato at Ames, Iowa, ascribed by him to Cladosporium fulvum. He: performed numerous inoculation experiments upon the tomato fruit with spores taken from the leaves, These bnooulations were reported as being successful in part but from the description, it seems doubtful that the rot was caused by Cladosporium fulvuna. -30- Halsted himself was somewhat doubtful as to this point. Orton (1903) called the disease "Soab" and Ferraris (1915) stated that "on fruits appear very similar little soabs but only on green fruit is there much change", According to Plowright (1887), "the fruit itself once set, seems generally to escape®*, Numerous attempts by the writer to produce an infection on tomato fruits, ranging from the size of a pea upward, have failed wherever the skin was uninjured, Various methods ot inoculating the fruit were tried. Thirty inoculations were made by sowing fungus spores on the skin of the fruit in a emall drop of water, afterwards covering the inooulation with a fleck of wet cotton. This method never failed to produce inreotion upon leaves but failed in every case with the fruit, superficial mycelium only being produced which finally died, Fruit was also bagged after being heavily inoculated with Cladosporium spores but no infection occurred, On fruit injured by cutting with a soalpel and then inoculating with spores, but one case of infection oocurred and this was accompanied by an Alternaria growth so the result was considered negative. Green fruit when injured and inoculated with spores, showed great power of resistance; the injury wae generally healed by the formation of callus or soar tissue and -21l- the fungus seemed to be unable to obtain a foothold. In infeoted greenhouses, only one case was found of a fruit of any size being infected (see Fig. 9). This fruit was about one-fourth grown and a microscopical examination showed the presence of Cladosporium fulvum conidiophores and conidia intermingled with those of @ especies of Botrytis. Fruits from the size of a pea downward and ovaries were unquestionably intected with Cladosporium fulvum as spores and conidiophores were found in abundanoe upon them. Although Cladosporium tulvum may inteot fruit through an injury or in connection with injury from other fungi, it cannot be said to be a disease of the sound fruit under ordinary Michigan conditions and is of minor importance in this regard, Appearance on the Blossoms. Blossom infection is equally as serious and important ae the inteotion on the leaves. Here the fungus obtains e hold on the calyces, the petals, stamens and pistil, and the developing ovary itselr becomes intected from- the other parts so that blasting of the blossoms is of common and widespread occurrence, Figure 9 shows young intected ovaries, the intected areas of which have become browned and dark colored. Figure 10 shows four young infected ovaries removed from the infected branch shown in Figure 1. Figure 11 shows one of these ovaries with the attached oalyx considerably enlarged. -2324=— Fig. 9 = Young tomato fruits intected with Tomato Leaf Mold. Healthy fruit shown at left, diseased fruit at right e -25- Fig. 10 - Infected calyces and ovaries of tomato. Ovaries are browned and calyces covered with masses of spores. ~Z4< Fig. 11 = Young tomato fruit (shown natural eize in Fig.10), greatly enlarged to show fungous growth on calyx. Growth on ovary itself are not conidiophores but glandular Plant hairs. -25- The fungus growth can be easily seen upon the calyx lobes and the darkened area shows the inteoted part of the ovary. Such fruits never develop but soon fall from the vine. Microscopic examination of these blasted blossoms showed conidiophores and conidia upon the Calyx and in one case, upon the style and stigma of the flower. No conidia could be deteoted upon the ovary itself but it was abundantly infeoted in the interior with the fungous myceliun. That the fungus will attack the blossome and young ovaries and not the more mature fruit may be explained by the fact that the calyx has stomata and is similar in structure to a leaf, 80 that the fungus can find ready entrance, The young ovary of the flower also has stomata and lenticels whioh may afford an entrance tor the parasite. Blasting of the ovary was obtained on two difterent occasions by placing conidia upon the stigma. On the other hand, the stomata occurring on the wery young fruit are rapidly transformed into lenticels and by the time fruit has reached the sise ot @ pea no more stomatal openings are to be found but numerous lenticels have taken their place. ithe ekin of the tomato is naturally very tough, consequently atter the stomata disappear no openings are left by which the fungus oan enter the fruit except by going direotly through the epidermis. Intrection studies have -36- shown the method of infection of Cladosporium fulvum to be stomatal hence its inability to inreot the tomato fruit after the fruit has attained a certain sise. ETIOLOGY OF THE DISEASE. Previous Work. Inrection experiments reported for Cladosporium fulvum are soarce; most authors have been content to describe the fungus which is so plainly assooiated with the host that intection experiments seemed need- less, Halsted (1883) did the first work of this kind but, as stated above, it is doubtful whether the organism observed was Cladosporium fulvun, Prillieux and Delacroix (1890) reported that infection experiments succeeded perfectly; at the end of three weeks fruoti- fication or the fungus took place and the leaves began to die, That pure oultures of the fungus were used is not mentioned, however, so the presumption is that the inoculations were made from spores on intected leaves directly to other plants. Formal Proof of Causation. three separate single spore isolations of the organism were made, The first isolation was made from suspected tomato leaves sent in by Mr. C. W. Waid of the Department of Horticulture, from the greenhouse or Melvin Lennon ot Ann Arbor, Michigan, on June aoth, 1916. The intected leaves compared favorably with the descrip- tions of the dieease given in Saccardo and standard texts on Pathology and a single spore isolation was made by means of the "loop dilution" and "poured plate*® method, using a oorn meal agar medium, Pure Culture. Cultures obtained from this isolation made on July 1, 1916 were used for observation of the fungus in pure culture on dirterent media until November 5, 1916 when artificial inoculations were tried upon tomato plants but these resulted negatively. Accord- ingly on November 33, 1916 a new single spore inocula- tion was made from infected leaves sent in from the greenhouse of a Mr. Dudley of Redford, Michigan. Cultures were made from the spores produced by the growth from this single spore inooulation and on December 31, 1916 eighteen inoculations were made upon eix young tomato plants. The spores were sown in small drope or water placed on the leaves, then covered by emall flecks of wet ootton whion served to maintain a moist condition and also retain the spores in a definite place, a method used by Levin (1916) in studying the tomato leat spot disease. -28— Produotion of the Disease on Healthy Plants by Inooulation rrom Pure Culture. The above inoculations were uniformly successful and infection on all plants was found to be present on January 10, 1917. In three weeks spores were being produced freely and the characteristic yellowing of the leaves was noted. Many of these intected leaves were wrapped in paper, placed in flower pote and covered with soil. Some of the pots were placed out doors, others left in the greenhouse and others put in difrerent parte of the Botanical Building where conditions of temperature and moisture varied, This was done to see it possible what the eftect of "wintering" or a "rest period® would have upon the spores. Reisolation and Reinooculation. On April 34, 1917, nearly four months after placing the intected leaves in flower pots, they were examined and a single spore isolation made from spores on the leaves. The isolation was successful and after growth in culture on corn meal agar plants were again inooulated with spores from the isolation, Intrection was positive and characteristic of the organien, the rest period did not seem to have lessened the viability of the spores in the least and the plants were succesefully inoculated with spores direotly from these -2Z9~@ leaves on June 321, 1917, which shows that the spores may cause infeotion after a considerable time, TAXONOMY OF CLADOSPORTUM FULVUM. Cladosporium fulvum Cke., was first described by Cooke in 1883 as an imperfeot form belonging to the family Dematiaceae of the group of "Imperfect Fungi*" and it still remains in this group ae no perfect stage has been reported for it. The fungus may some time be found to have a perfect stage, presumably in the Myoosphaerellaceae, since Janozewski (1893), established the connection of what he called Cladosporium herbarun, causing an injurious disease of cereals, with the Ascomyoetous germ Mycosphaerella. Infected tomato leaves wintered over in the soil of flower pots, gave no indication of having a perfeot atage formed upon them but inoculations of spores from pure oulture upon autoclaved oorn grains in test tubes on June 19, 1917 showed on August 10, 1917 rather thiok mats of fungus hyphae distributed through which were numerous spherical bodies ranging from 50 to 1320 microns in diameter, These bodies presented a structure typlioally peritheoial in appearance with a thin, pseudoparenchymatoue wall but although partially hollow contained neither asci nor spores, and could not be made to form spores by ordinary variations in the cultural teohnique. ~30— MORPHOLOGY OF THE PARASITE. Myoéeliun. The mycelium of the organism waries considerably according to the culture media used, age of culture, eto. The young hyphae from germinated spores are about 32 microns in diameter, delicate, hyaline and septate, at first unbranched but soon sending out numerous lateral branches at right angles to the main etrand. These hyphae grow very well in Van Tieghem moist chambers or upon the tomato leaf surface (see Figure 14), where they may attain a length of several hundred microns, Some have been observed in moist chambers which were 400 microns long. On nutrient media, such as corn méal agar, the mycelium averages about 3.5 miorone in diameter, is much Longer, more numerously branched than in the moist chamber and presents a granular rather than a hyaline appearance, Septa are here more numerous and constrict- ions often ooour between the cells which sometimes become greatly swollen, Older myoelium grown in a sterilised extraot expressed from tomato foliage is a dark brown in color, frequently septate, granular and averages 5 microns in diameter. On autoclaved corn kernels the mycelium produces dark colored, stromatoid masses, composed of -3l- hyphae possessing greatly swollen and globular shaped cells which are easily broken apart; cells of this kind often measure 10 microns in diameter. Conidiophores. Conidiophores are four to eight septate, nodulose, sparingly branched and dark brown in color. Their length and width varies from 75 to 140 x 3 to & microns with an average of 80 to 90 x 4 to 5 microns. The conidiophores project from the leaf surface in clusters which emerge from the stomata. Consequently it usually occurs that the conidiophores are of greater diameter as the top than at the base where they are densely crowded and form a stroma-like mass. (See Figs. 12 and 13). Branches usually occur below the septa, often at right angles to the conidiophore and may be long or mere wart- like processes from which the conidia are produced. A Single branch usually occurs below a septum but sometimes a whorl of them is found. The conidiophore generally ends in a chain of conidia but often tapers to a narrow tip or forms a "boot-like" process. (See Fig. 13). Conidia. Conidia are produced acropetally either at the tip of the conidiophore or laterally from wart-like projections which arise immediately below the septa or sometimes laterally from other conidia. They begin to be formed. ~33- Fig. 18 = Cross section of tomato leaflet near midrib (on left), which shows a@ cluster of conidiophores emerging from a stoma. Note a few darkly stained hyphal branches extending from the cluster into the leaf tissue and also inter-cellular mycelium, Fig. 13 = Cluster of conidiophores photographed from section cut from a badly infected leaflet. Note the peculiar "Boot" shape of the conidiophore tip in several cases. Spores are mostly broken from the conidiphores but many are shown in the field. -34— when the conidiophore is comparatively short (40 - 50 microns) and continue to be produced in chains as the conidiophores elongate. The youngest conidium of each chain is the terminal one. As many as six conidia are often found composing one chain. Conidia are ovoid to elliptical in outline, apiculate at one end, usually uniseptate, pale brown in color and measure 10 = 35 x 5 = 7 microns, The average length and width is 6 x 18 miorons. RELATION OF PARASITE TO HOST. Infeotion and Conidiophore Production. Type of Infeotion. To determine the type of infection two methods were employed: (1) Microscopes were set up with moisture supplied slides (see Fig. 14). Spores from pure oulture were sown in a droplet of water on the leaflet of one plant and the leaflet then placed between the cover glass and slide of the miorosoope shown on the right in Fig. 14 A long narrow strip of filter paper, having & emall square opening cut from one end, was placed under the cover glass and over the leaf in suoh a way that the square opening lay opposite the inooulated area on the leaf. This permitted observation of the inoculated surface through the microscope, held the cover slip away from the inoculated area and supplied moisture from a tall vial containing water. The other microscope was used in the same way, with the exception Fig. 14 = Apparatus used in preliminary work to determine method of fungous infeotion. -36- that spores taken direotly from an infected leaf were sown on the leaflet. (3) Inooulatione of spores were made on a number of tomato leaflets and the epidermis stripped from the inooulated areas after 13, 3, 36, 60 and 72 hours. By means of the microscope observation method germ tubes of various lengths were observed after six hours. At the end of twelve hours some of these were quite long, producing a superficial mycelium on the leaf surface but no infeotion was found until the next day, when the germ tubes had become very long and much branohed. (See Fig. 15). Several oases of stomatal entrance by the hyphae were noted after thirty-six hours but penetration direotly through the epidermis was not observed. That the type of infection is stomatal was proven beyond a doubt by the second method of stripping the epidermis from inoculated areas.* Epidermis from twenty-four hour inoculations showed no entrance of the germ tubes but the ends of many were turned toward stomata, in some cases reversing their original direotion of growth, This seemed to show the existance of a chemotactio influence exerted through the stomata upon the germ tube. Epidermis. stripped from leaves after * Pieces of epidermis stripped from the leaf were fixed in hot 85% alcohol, then run down to 50% alcohol and stained with an Erlich's Haematoxylon- clove-oil- eosin stain. This stained the epidermal celis a deep pink and the hyphal threads on their surface blue. Fig. 15 = Photomicrograph taken from upper surface of epidermis stripped from inoculated tomato leaf. Note.fungous hyphae overrunning the surface, three of which penetrate stomata (in oircles). -38< thirty-six hours showed numbers of germ tubes entering stomata. The germ tube may immediately enter a stoma but more oases were observed where it had reached considerable length before doing so and lateral branches were obeerved to enter as often as main branches, (See Fig. 15). Some- times two branches, given off at approximately the same piace on the germ tube, enter the stoma. A pronounced enlargement in the diameter of the germ tube occurs immediately upon its entrance through the stoma and the protoplasmic contents become more granular in appearance, Inoculations made upon both lower and upper surfaces of leaves have given positive reeults, showing that infeotion does oocur upon either side of the leaf which is what would be expected, as stomata are present upon both leaf surfaces, in greater numbers, however, on the lower side. Conidiophores. Numerous hyphal branches pass to the stomata (see Fig. 1), where a stomatoid structure is formed. Conidio- phores arise in clumps through the stomata from these etromatic masses of cells, whioh often measure twenty to thirty microns in diameter and muet rupture the stomatal opening during their growth, since the latter does not measure over thirty microns in length and is only about four microns in width, The conidiophores spread outward as they grow, the top portion of the cluster often reaching a diameter of one hundred microns (see Fig. 13). -39— As the stomata are spaced from thirty to eighty microns apart on the under side of the tomato leaf, it is no surprising thing that badly infected leaves seem to be a solid mass of spores on the under side when viewed by the naked eye, Viewed by means of the low power of the compound microscope, the conidiophores appear like little clumps of shrubs thickly and evenly planted over a level field, with their branches interlaced at the top. Morbid Anatomy of the Host.* Sections out from newly infected leaves show mostly intercellular mycelium but with growth of the fungus, the hyphae are found to be both inter- and intra- cellular, Haustoria are not produced, the hyphae ramifying through all parts of the leaf tissue, particularly around the traoheary tubes, Tracheary tubes out in oross section show a ring of fungous hyphae around them (see Figs. 16 and 17), and in longitudinal section the hyphae are * Histologioal Teohnique. The fixing agente used were Flemming'’s fixative medium formula, alcohol-formaldehyde and Gilson's Fixative, Flemming's triple stain and Durand's stain were used in staining sections but no success was had with the latter. @ alcohol-formaldehyde fixative seemed the better where sections were stained but specimens fixed in Flemming's fixative, sectioned by hand and mounted without staining gave better sections and more detail than thase cut by the microtome and stained. By this method the fungaus hyphae showed up very black with the septa clear and the host cells became greenish brown, giving plenty of differentiation for histological study. Flemming's triple stain gave better results than the Durand stain but the resulte were far from satisfactory. -40- Fig. 16 — Photomicrograph of cross section of tomato leaflet, showing ring of fungous hyphae around the badly disorganized tracheary tissue. Note also the badly disorganized appearance of the infected host's leaf tissue, hte Fig. 17 = Photomicrograph showing ring of hyphae in Fig. 16 under greater magnification, The mycelial septa may be seen in the lower portion of the cirole,. ~-43— found parallel to the sides of the tubes and closely appressed to them, (see Fig. 18) which indicates that the parasite is a strong contender for the water supply of the host. The mycelium at first does not greatly affect the host cells but its growth is very rapid and the host tissue and cells soon become disorganised and broken apart. The spongy parenchyma and traoheary tissue seem to be more susceptible tothe fungous attack than the pallisade parenchyma and is considerably more disorganised,