r^lij im\ IfeSi .;.>:; i'^' vflBM yy L^-u^.i 41- •fBo-t- PUBLICATIONS ON PLANT PATHOLOGY FROM WISCONSIN EXPERBIENT STATION RESEARCH HJLLETINS A? a.cV\ SC>v> , Vs/ 1 'b C o v^'^l Vi ri^.nr- ii2>"^ INDEX (Alphabetically by Authors) Dudley, John E., Jr., k C. L. Fluke, Jr. Res.Eul. 82-- Spraying Versus Dusting - To Control the Potato Leafhopper in Conimercial Potato Fields of Wisconsin Johnson, James Res. Bui. 32 - Black Rot, Shed Burn, and Stem Rot of Tobacco Johnson, James Res. Bui. 31 The Control of Damping-Off Disease in Plant Beds Johnson, James Res. Bui. 63- Transmission of Viruses From Apparently Healthy Potatoes Johnson, James Res. Bui • 76 '' The Classification of Plant Viruses Johnson, James Res. Bui. 87 '^ The Classification of Certain Virus Diseases of the Potato Johnson, James, & Herbert F. Vurwin Res. Bui. 62^ Experiments on the Control of Wild- fire of Tobacco Johnson, James, &: V;'. B. Ogden Res. Bui. 95- The Overwintering of the Tobacco Mosaic Virus Jones, F. R., 8c K, B. Linford Res. Bui. 64^' Pea Disease Survey in Wisconsin (?) Jones, Leon K. Res. Bui. 59 Anthracnose of Cane Fruits and Its Control on Black Raspberries in Wisconsin Jones, L. R., & J. C. Gilman Res. Bui. 38 The Control of Cabbage Yellows Through disease Resistance Jones, L. R., ^^ Iviiller, k E. Bailey Res. Bui. 46 Frost Necrosis of Potato Tubers Jones, L. R., J. C Walker, & W. B. Tisdale Res. Bui. 48- Fusarium Resistant Cabbage Jones, L. R., H. H. McKinney, & H. Fellows Res. Bui. 53 The Influence of Soil Temperature on Potato Scab Jones, L. R. , Sc Regina S. Riker Res. Bui. Ill Wisconsin Studies on Aster Diseases and Their Control Keitt, G. W., & Leon K. Jones Res. Bui. 73 Studies of the Epidemiology and Control of Apple Scab Linford, l.!aurice E. Res.Eul. 85 A Fusarium V»ilt of Peas in Wisconsin llelhus, I. E. Res.Eul. 15 Experiments on Spore Germination and Infection in Certain Species of Oomycetes Melhus, I. E. Res.Eul. 37 Germination and Infection with the Fungus of the Late Blight of Potato (Phytophthora infestans) (3) Rands, H. D, Res. Bui. 42^ Early Blight of Potato and Related Plants Shands, R. G. , E. D. Leith, J. G. Dick- son, H, L. Shands Res. Bui. 116 ' Stripe Resistance and Yield of Smooth-Av?ned Barley Hybrids Stout, A. B. Res. Bui. 18 - A Sclerotium Disease of Blue Joint and Other Grasses Wade, B. L. Res. Bui. 97 Inheritance of Fusarium Wilt Re- sistance in Canning Peas Walker, J. C. Res .Bui. 107 - Resistance to Fusarium Wilt in Garden, Canning and Field Peas \^ Experiments on Spore Germination and Infection in Certain Species of Oomycetes' I. E. MELHUS INTRODUCTION Some two years ago special work was undertaken upon Cysto- piis and other Oomycetes aiming to learn more as to the methods of spore germination, zoospore fonnation, methods of infection, and as to the possible existence of so-called "physiological spe- cies" in the genus Cystopus. The parasite Cj/stopus candidus was common on the garden crop of radish, Raphanus sativus, and attempts were made at the outset to produce infection by transferring spores from this to radishes growing in the green- house. Although repeated trials were made in connection with these early studies, only meager and irregular infections resulted. This suggested that some variable factors of unknown nature were present in the greenhouse trials. Difficulty was also en- countered at the outset in securing uniform results in spore germination by the methods described by earlier workers. Thus it soon became evident that some more specialized methods were necessary in order to secure the germination of the conidia and host infection in abundance and with a satisfactory degree of certainty. We were thus led to attempt to determine the relations of various conditions to spore germination and infection with tliis fungus, not only host relations, but also relations of the age 1 The author wishes to express to Dr. R. A. Harper and Dr. L. R. .Tones his sincere appreciation for the kind criticism and keen interest shown during the progress of this work an(J preparation of the manu- script, 26 WiHCONSiN Experiment Station. and maturity of the spores, moisture, food, chemical stimuli, liglit and temperature. Other species were suhsequently tested. The results secured are of such definiteness and breadth of applica- bility as to justify their publication, although much remains to be done upon the problems as originally defined. Before discussing my own work and conclusions, a brief review will be made of the results of previous studies upon these and closely related mat- ters. Review of Eartjer "Work The germination of the asexual spores of Cystopus was first observed over a century ago by Prevost (1807:29). He studied the species commonly parasitic on crucifiers and purslane respectively, then known as Uredo Candida and Uredo portidacae. His description is clear and interesting. He states that the spores germinated one or two hours after immersion, some- times within 40 or 45 minutes, owing in all probability to differ- ences in temperature, wdiich, during the observation, fluctuated between 12° and 16° R. (equivalent to 15° and 20° C). In the process of germination the spores absorb water and become bottle shaped; soon a globule (zoospore) is seen on the outside and this is immediately followed by several others, -sometimes as many as six more. He states that these globules instantly reunite into a mass which moves as a unit by rolling about in the water. The globules, as a rule, separate from one another in a very short time ; sometimes, however, two or even three globules may remain attached together, either immediately touching or as if joined by a string. Those globules which separate from one another, and they are by far the greater number, are sometimes a little angular and possibly a little hollowed or pushed in at one side. They smm about in the same way as when united in mass. Soon the movement of the globules ceases and they become fixed at the surface of the water or at the edge of the drop. He observed that the swarinspores developed germ tubes, regarding which however he gives little detail. Prevost likcA^ise studied the Cystopus on salsify and Amarantlms and louiul tlio last two forma much more difficult to germinate. Tulasne (1854:77) states' that he germinated the spores of Uredo porhdacae and Uredo Candida but was unable to get them to form swarmspores in tlie manner deserilx^d by Provost, find- ing only the germination by a tube. KXI'KKI.MKNTS ON Sl'OKK ( IkKM I XATIO.V. 27 Iloflfinan (1859: 210) was also iuuil)lc to confinn Prevost as to the formation of swarm spores in Cystopus. lie describes the germination of the spores l)y tul)cs in tlie same manner as deserihed by Tulasne. DeBaiy (I860: 23G) studied the method of germination in Cystopus cnhicKs and Cystopus candichis and found the germina- tion was by zoospores, as described by Prevost, but that germina- tion miglit take place at any tcmper-ature ])etweon 5° and 25° C. He emphasized for the first time that tlie spores are really sporangia, producing from five to eight zoospores in Cysto- pus candidus and from eight to twelve in Cystopus cubicus. DeBary describes the changes in the sporangium very clearly. The spores abso"rb water and swell when sown in a drop of water. On one side an obtuse papilla is developed and vacuoles forrii in the granular protoplasm which disappear later. At this stage of development, fine lines of demarcation divide the protoplasm into five to eight polyhedral portions leaving at the center a small f(>ebly colored vacuole. AVhen the division of the content of the sporangium is complete, the papilla swells, opens and the zoo- spores are pushed to the outside one by one, showing no sign of movement. Once outside the sporanginm they become lenticular in form and group themselves l)efore the oi)ening of the sporan- gium in a spherical mass. Very soon the swarm spores begin to move, vibratile cilia appear and the globular mass is set oscillat- ing. The zoospores ultimately become free and swim away singly. The motile spores are plano-convex or slightly concavo-convex having a small disk-like vacuole on one side. Attached near the vacuole are two cilia, a short one in front and a long one behind, both on the same side. In from one and a half to three hours after being placed in water the escape of the zoospores begins. They will develop either from sporangia freshly formed or from those which have been kept as long as six weeks, DeBary (1863:14) found the conidia of Cystopus germ- inating on the leaves of the host plants. Zoospores were found in the drops of water on the leaves. Infection experiments with Cystopus candidus were made on varions hosts. In the case of Lcpidium sativum the zoospores readily entered the stomata of both leaves and cotyledons but produced infection only in the ktter. Various species of Brassica showed the same tendency, though not to so marked an extent. 28 Wisconsin Experiment Station. Farlow (1875:319) studied the germination of the conidia of Phytophthora infestans and observed* that sometimes the con- tents of the spore discharged in one mass, and from this mass zoospores are produced as before. He believes that the produc- tion of zoospores is favored by darkness, whereas germination by a germ tube takes place more frequently in the light. He states, however, that he has repeatedly sown spores in watch glasses and both methods of germination resulted. The germinating power of the spores was retained for several weeks, but tliey did not germinate after a winter's exposure. Farlow (187G :410) also describes the germination of the conidia of Plasmopara riticola. During the month of October when the disease is most prevalent, he found that in one and one-fourth hours zoospores were formed and began to make their way to the outside of the sporangium. The conidia might also germinate directly, i. e., by germ tubes. Darkness was more favorable for the germination of the conidia, whether directly or indirectly. He found the zoospores to swim about for fifteen to twenty min- utes, after which the motion gradually became slower until they finally came to rest. In another quarter of an hour an outgro^\^:h appeared on one side which rapidly developed. Scribner (1886 :10) states that temperature exercises a consid- erable influence over the germination of Plasmopara viticola from the grape. The most favorable temperature lay between 25° and 35° C. At a lower temperature, germination took place more slowly, but the temperature could be reduced to zero without d(\stroying the vitality of the conidia. Under exceptional cir- cumstances, Scribner states, another form of gei-minatinn might occur in which a conidium may push out a tube. This method, he reports as undoubtedly rare. The question of spore germination and physiological species in the genus Cysto^^us has quite recently been studied in consid- erable detail and with very interesting results by Eberhardt (1904:622). Pie began his work in June 1902 and continued it until the fall of 1903. Cystopus candidus on various crucifer- ous plants were used except in one experiment Avhere he used Cystopus ciibicus on Tragopogon pratensis. Eberhardt (1903:655) found, as we have, considerable diffi- CL;lty in germinating the conidia of Cystopus. He tried the dif- ferent methods used by DeBary and Zalewski, but obtained only a low germination. To solve the question of spore germination, liXPEUIMENTS UN Si'UUK GeUMINATION. li'J Eberliardt turued his attention to methods oi' properly muturiug iiiu cuiiKiia. jjurjiig all ilie inontli oi Mny aiid llie begnming oL (iiims ,CiJ6lupUi> oandidas was available ou Lapsella, twelve expeniiienis were iiiaue at as many uillerent limes ui wiueh llie iollowiug may be lakdn as typical. Conidial pustules not <)j)e(iea were eolieeled irom time to liiue. 'Uie contents ot tiie pustules were placed in a vial eontaining a small amount ol rain water. 'I'his was ealletl viai ^\o. i. In vial ^N'o. 2, eontaining ram water was placed, the spores shaken irom open pustules, in still another test a shoot was taken which bore many unopened pustules, it was wrapped in a moist cloth to be kept for lurtlier observations. In vial No. 1 the eonidia remained in chains, the protoplasm became grantilar and later bacteria developed decomposing the couidia. Vial No. 1 may show a small number of eonidia ger- minated, but the larger part disorganized. In vial No. 2, containing spores shaken irom open pustules one third of the eonidia formed zoospores wliile the remainaer decomposed. ^Vhen taking spores from pustules just opened, more than one half germinated, it is permissible to think that the former were relatively old and that they had lost their capacity of germination whereas the later eonidia were taken from pustules just opened and gave a germination of more than one half of the spores. The third test, in wiiich the shoot containing pustules was wrapped in a moist cloth, gives further evidence in support of the above interpretation. After the pustules had been kept moist for one day a microscopic examination of three of the pustules showed the following conditions on being placed in a drop of water. The young pustule was found to contain the zoosporangia in chains which gave no germination. The second showed a small fraction of the spores empty or bottle- necked, ready to germinate. The other eonidia were decom- posing. The third presented all the eonidia disassociated and about one third produced zoospores. The shoot was kept fur- ther until almost all of the pustules had ojjened liberating the spores that germinated in fairly large numbers. This same shoot was kept two weeks longer when it had wilted and dried up in places. The eonidia that fell from it gave no germination. These experiments show us that the zoosporangia are the organs of immediate infection, requiring for germination to be collected at the time when the pustules open. Infection occurs 30 Wisconsin Experiment Station. when these spores fall upon the young host plants humid from rain or dew. In the numerous cultures made for infection purposes none were kept after the third day if germination had not already occured. It is well here to notice the variation in time required for germination of conidia. A few notes from Eberhardt's experiments will suffice. Ebcriiardt used "room temperature" in every case except in one test made on the twenty-nintii of ]\Iarch. The exact temperatures are given in only two cases, however. In the first it varied from 11° to 17° C In the experiment referred to, on March 29, the temperatui'e varied from 2° to 8° C. The latter condition was obtained by placing the culture on a win- dow sill wliere the influence of tiie out-door temperature had some elfect. His germination experiments that are described were carried on from ]\Iarch 28 to August 9, 1902, on five different dates, all resulting in germination in from three to forty hours. There was also another difficulty greater than that of the selection of the conidia, it was the choice of the proper time to inoculate the host plant. The task of growing and caring for the young plants, the delicacy of the material of infection and the considerable space required for growing the crucifers has made it iinpossil)l(' lo oflcn repeat the same infections. In view of the fact that tlie following statements of the jnithoi- in'c not clear we quote dii-ectly. "Un facteur important etait la coincidence qui doit toujours exister entre la recolte du parasite et I'etat de germination propice de la plante a infecter. C 'est pour ces raisons que nous ne pouvons poser en ce moment aucuuc regie basee sui- 1 'experience, relative a I'optimum de receptivite du pai-asite par 1(> vegetal nourricier. ]Mais ce (pie nous pouvous cei-tainement avancer, c'est (|u'il ne suffit pas dans toutes les Crucifei-s d 'avoir des cotyledons bicn etales. La ques- tion de roj)timum ch' rece])tivite demande ])lusieurs annees de recherches. Comme nos infections tendaient plutot a prouver ] 'unite d'espeee, ce n'est que tres taid que nous nous sonunes aperu, apres des insucces nombreux, combien I'etat du jeune plant influe sur la reussite de 1 'experience. Ainsi nous avons vu (juc certains cotyledons, sortant de la gaine refusent I'entree du parasite, tandis (|ue lorsqu 'ils sont bien etales, ils sont sus- ceptil)les d'infe(!tions. 11 nous a semble que plusieurs de nos especes qui avaient etc infectees au moment oil les eotyludons Experiments on Spore Germination. 31 ('taiciit r.-iiit's jx'uvciit i't'(H-v()ir rciidopliyte par Ic jeuue bourgeon foliairi' drroiilaiil scs tViiilles. Mais nous no pouvons encore rien aHii'iner k I'e pro])os. Au reste, ])c Bary lui-inOine indiquc que Il< h'ophila crilliDu folia est apte <\ etre ijifectee par les jeuues f.'uillrs." Conii)aratively s[)eaking, but very little work has been done on the question of physiological species in the Oomycetes. Eber- hardt (1904:714) has investigated this problem in Cystopus can- didtis occurring on various crueil'ers. Plants were inoculated by germinating the conidia and placing the liquid containing the zoospores on the lower side of the cotyledons of the plants to be infected or simply dipping the cotyledons to be infected in the water cont^iining the zoospores. The seedlings Avere grown in large flower pots and small bunches were removed as needed for infection. His infection experiments were carried on both out of doors and in an ordinary laboratory. The conidia of five different species were used as material for infecting other cruciferous plants. It is further interesting to note that EberliardL mailc inocula- tion experiments with the oospores of Cystopus caiKlidus from Lepidiutn sativum. In order to infect the plants of Lepidium and Capsella with the oospores from Lepidium satiium, the parts of the host containing tlie oospores were placed in small bags and hung out of doors in the open air during the winter months. In March and April the oospore material was distributed over the surface of pots where it might decay and liberate the oospores. Two of these pots were then seeded with Lepidium satiinim and Capsella Bursa-pastoris and the young seedlings became infected. It should be noted in this connection that Eborbardt used no con- trols in this series of experiments. His results may be sumnuirized as follows: "With the conidia of Cystopus candid us fi'om Capsella Bursa-pastoris he infected Capsella Bursa-pastoris. Lepidium sativum, Iheris amara, and Arahis alpina. Conidia from Capsella Hccgcri infected Capsella Heegcri, Capsella Bursa-pastoris and Jjepidum sativum. Conidia from Lepidium sativum infected Lepidium sativum and Capsella Bursa-pastoris. Spores from Brassica rapa infected Brassica rapa, Brassica- oleracea (varieties: hotrytis, capitata, and cuugylodcs), Brassicoi nigra, Sinapis arvensis, and Diplotaxis tenuifolia. The widest range of infec- 32 "Wisconsin Experiment Station. tion was obtained with the conidia from Arahis alpina whlcl; infected Arahis alpina, Arahis hirsuta, Arahis Halleri, Arahis iurritis, Lepiditmi sativum, Iheris amara, CarcCamine pra- tcnsis, Cardamine amara, Capsella Bursa-pastorts, and Sene- bicra coronopus. "With the oospores from Lepidium sativum lie infected Lepidium saiivum and Capsella Bursa-pastoris. It was iinjxjssible to infect any of the Cruciferae with conidia from Tragopogon praiensis, but quite easy to infect ticorzonera It-.s panica. From these darta Eberhardt believes that there are no biological species in the species, Cystopus candidus. It shoiiid be noted, however, that the above conclusions are based upon only one trial in some cases with each species or variety of plant. Eberhardt states in this connection that the laborious task of growing, inoculating and recording results, together with the fact that the ground used for growing the plants could not be had during the next year, made it impossible to duplicate any of the series of experiments performed. The fact that Eberhardt 's work was incomplete and not fully convincing, and that it is especially important that each group of parasitic fungi be fully understood as to the existence of physiological species, led me to study this problem. EXPERIMENTAL STUDIES IN SPOKE GERMINATION ]\Ietpiod As has been previously stated, the major portion of the experi- ments recorded in this paper were carried on with the common white rust, Cystopus candidus, as it occurs on various garden plants. The culture M^ork was all carried on either in the green- house or in an ordinary laboratory. Both distilled and tap water were used to. germinate spores. The tap water in this case is un- filtered water dra^^^l directly from Lake Mendota. The conidia were gathered from infected plants growing out of floors until frost, when they were taken from infected plants in the green house and were sown the same day as gathered. The spores were sown in a drop of water placed near the center of a clean slide. It was always difficult to make the conidia sink, but by stirring the drop with a scalpel, a large per cent could be finally made to settle I'] M'l; KM mi; NTS ON Si'OKK ( J I :iC.M I NATION'. ,^3 to the bottom. The aim was always to add only a moderate ii um- ber of spores to each drop, since too many spores make the water opacjue and difTicult to examine. Care was taken to obtain fresh spores in every ease, wludi were stirred into the drop of tap water. Sometimes small i)ieces of leaves bearing pustules were dropped into water on a slide. The first experiments were made at room temperature, ])ut the irroii'ularity of the results and the small per cent of the total which germinated even in the most favorable eases led to experi- ments with low temperatures. With this idea in mind, the slides sown with spores were placed in an ice box of the usual construc- tion. A Richard's self-regislering thermometer was also placed in the ice box so that a complete record could be had of the temperatures to which the spores were exposed. In the case of the controls at room temperature, the stands holding the slides soAMi Avith spores were placed on a wet earthen plate and a small liell glass placed over them to prevent rapid evaporation. The effect of darkness was also tested by placing similar cultures in the dark room. The temperature of the dark room was noted when the experiment was started and stopped and the average taken. The cultures at room temperature, not in the dark roojn, were kept in a greenhouse where another self-registering ther- mometer recorded the temperature. Here also the temperature was noted Avhen the experiment was started and stopped. For convenience in tal)ulating, the average of the two extremes was taken as tlie prevailing temperature. C'nltui'es in the ordinary Van Tieghem cell were also used. Tie cell was partially tilled with tap-water and a luinging drop made containing the spores. Vaseline was used to prever"" evaporation and to hold the cover glass in place. The cells were laid on a stand as described above for the slides. Watch crystals were used when it was desired to secure large quantities of spores in differ- ent stages of germination. Ref.ultp. of Germination Experiments 'My earliest experiments in germination of the spores had the double aim of providing material for the cytological study of the processes of nuclear and zoospore formation in the germination of the conidia and of obtaining a reliable method for germination of the conidia for infection experiments in determining whether 34 "Wisconsin I'].\i'j;ki.mi;ni' Siaiiun. physiological species are to be found in the group. As noted earlier, it "was found difficult to obtain germination and a long series of one hundred (experiments was made, testing the general questions as to the effect of age and maturity of the spores ; the weather conditions under which they were collected ; possible in- fluence of rain water, dew, tap water and distilled water; and the effect of gathering the spores at different times of the day. Attention was turned to the possible effect of artificial media and a number of attempts were made to germinate tlie spores of Cystopus in artificial media, these experiments lieing made in a greenhouse the temperature of which varied from about 33° C. in the day time to 22° C. at night. First ordinary nutrient agar was tried (3 gr. meat extract, 10 gr. agar, 3 gr. salt, 1000 cc. water). The conidia were sown on this agar in petri dishes and kept under observation for twentyfour hours, but no germination resulted in any of the ten experiments tried. Beef bouillon (3 gr. meat extract, 3 gr. salt, 1000 cc. Avater) was tried in two experi- ments, but gave no germination. Following this, some special media were tested. Ten experiments were tried with lima bean agar, as prepared by Clinton (1908:904) for growing Phytoph- thcu'a, and eight experiments with his pumpkin agar ; none of the cultures showed any signs of germination at the end of twenty- four hours. Various other forms of artificial media were tested, including mustard leaf decoction, four trials; corn meal agar, five trials; a two per cent sugar solution, six trials. The cultures were kept under observation in cacli case for twenty-four hours. In no case did germination result either by germ tul)es or by zoospores. The above experiments were made during July and August, 1909. The conidia used were from Cijsiopus candidus, G. hliti, C. cuhicus and C. porlulacae. They were collected at various times of the day ranging from seven o'clock in the morn- ing until seven in the evening, the great majority being gathered about eight o'clock in the morning. The infected leaves were cut off from the host plants and carried to the la])oratory. where the material was used immediately. A num1)or of times the infected leaves were immediately placed in a damp chamber after they were removed from the host plant. A large number of tests were made with both young and old conidia., before and after the epi- dermis of the pustule had i-uptured. These conidia were, of course, also tested in ualei- on a slide or in a hanging drop in a so-called Van Tieghem cell. The slides I'lXI'IOIflMKNTS ON Si'OKI'; TJ KKM | N \'||( (N". 35 worn ]>l.i(M'(l on ;i siiinll sljiiid on ;i wot earthen plafo nM'oung . 'S'oung . ^'oung . ^'()ung . ^'oullg . ^'oung . ^'oung . Young . >'<)ung . ^'oung . ^'()ung . Young . Young , ^'oung , ^'onng , ^'oung , ■^'oung , ^'oung ^■oullg ^'oung N'oung Young Old.... \'oung Old.... Old.. . Old.... >'oung ^'oung ^'ol^ng ^'oung ^■oung ^■oung ^'(>ung ^'ouiig Period of re- fbigebation. Hrs. Temp. 1.5 2 1 1 1.5 2.60 3.5 1.5 11.75 2 24 2.5 4.5 24 7 24 36.5 28 4.5 4.5 5.75 4.5 3 3 4 3,5 16 15 2,66 7.5 6 6,25 3 3 47 10 22 12 5.5 6 3 4.25 18.5 2.12 10.75 2. 5 2 5 5 25 1 a 25 6 0 4 25 " 2.33 2 75 3 4 87 ct 5 it 25 Out -door Re- icnipera- sult. lure. Minimum. 8 9 10 10 13 12 12 11 10 10 10 11 10 11 11 18 14 15 11 11 10 10 10 10 8 12 * E-\periments carried on in watch cr.vstals. •I' l*ieces of leave.s, on vvhi;'li tlicrc vere pjstules, were laid on tlie slide. Experiments on Spore Germination. 37 Tai!i,i, I. Continuod.- Kkkkct ok F>()\vkiun(; Te.mi'kuatcke on thk (ilOUMINATION OK TIIK (JONIDIA Ob' CKUTAIN Sl'KCIES OK CySTOI'U.S. Date. Sept. 21. SeiJt.21. t;ept.21. Sept. 21. Sept. 22. Hv)pt.24. Sept. 25. Sept. 25. No. of cultures. Species tested. C. candidus C. candidus C. cubicus. C. bliti C. cubicus. . C. cul)icus. . C. cul)icus. . C. cubicus. . Pehioi OF RE- KUKJEKATION. Asre of spores. Re- sult. His. Temp. 14 Young . . 23 • + Younsr . . 23 14 + Vount^ .. 4.87 12 + N'ount? .. 29 14 + ^ oung .. G 11 + "iountr .. 3.75 8 + Old 23 7 -Voung .. 3.25 ( + Out-door tempera- ture. Minimum. *E.\perirneiits carried on in watch crystals. a considerable number of spores remained in the pustules; and old after the pustules were nearly empty. The conidia were either taken out of the pustule and placed in a drop of water or small pieces of leaves with pustules were laid in a drop of water. If the pustules on the pieces of leaves were not already open when they were laid in the drop of water the epidermis was broken with a needle. An ordinary ice box was used and in it was kept a self regis- tering thermometer. By referring to the table it can be seen that tlie temperatures from August 10 to August 25 ranged from 15° to 21° C. The temperature was usually above 18° C. The ice box was kept in a rather warm room adjoining the green house, and it was also used for other purposes so that the doors were opened and closed quite often. The temperature curve was very irregu- lar. There Avere fluctuations of 10° C. in five hours in some cases although usually it was less. Since the fluctuations were too numerous to explain in connection with each test, the average of me maximum and minimum temperature iias been tai^en as the prevailing temperature and is that recorded in the table. This, in some experiments and especially in those before August 13, does not give the correct temperature conditions. In the tests after August 13, there was much less variation and the average of the two extremes is much nearer the prevailing temperature condition. The exact temperatures during two tests are given in detail to show more clearly the existing conditions. For example, the temperature varied as follows during the experiment on August 25 : The test was started at 9 :15 a. m. with a temperature of 20° C. The temperature remained constant until 10 o'clock. Thirty minutes later the temperature was 21° C. At 11 o'clock 38 Wisconsin Experiment Station. it was 19° C. where it remained until 12 o'clock noon. At 1 o'clock it was down to 14° C; at 2 o'clock it was 13^^ C. and at 2 :45 p. m. it again rose to 14° C. During 5 1-2 hours, the tem- perature varied 8° C. After August 25, the temperature of the ice box was much more constant and often fluctuated only one degree during the time of an experiment. The period of refrig- eration on September 2 started at 9 :30 a. m. -with a temperature of 10%° C. and stopped at 2 :35 p. m. with a temperature of 11° C. During the five hours of refrigeration, the temperature only varied 3/, degree. It is to be noted also tliat the temperature grew gradually lower until October 9. This was due to the fact that before this time, no artificial heat of any consequence was used in the green house ; while after this date it was heated. For comparison with the temi)erature in the ice box, tlie minimum out door temperature is given for each day on which an experiment was made, as pub- lished by the local weather bureau of Madison in their monthly meteorological summary. It will be seen that the temperatures in the ice box varied two degrees or less from the minima out doors in 72 per cent of the tests until about October 14. This suggests that germination can readily take place at temperatures equal to or varying two degrees ox less from the minima for the outdoors. The length of time required for germination varied from one to thirty six hours. The one hour period required for germination was on August 12 and the thirty-six hour period was on October 10. All of the experiments that required an unusual length of time for germination were examined from one to six times before the final observation was made. Water was added to replace the amount that evaporated. It should be said, however, that the usual period required for germination in the majority of the cases was less than six hours. From August 10 to 31 the average length of the period of refrigeration necessary to produce germi- nation was about 3% hours. The longest period was ISy- and th(3 shortest period, one hour. During the month of September experiments were made on twentyone different days, one more than in August, and the average length of the period of refrigeration was about 7^2 hours; here the longest period was twentynine hours and the shortest period two hours and five minutes. In October, experi- ments were made on nine different days. The average of the periods of refrigeration used where germination resulted was nine Experiments on Sporf, iiKK.MixATiON. 39 hours, the longest period being 36^^ hours, and the shortest two hours and forty minutes. From these facts it is strongly sug- gested that the period of refrigeration is longer in the fall than in the summer as has already lieen pointed out by Zalewski (1883:215). No germination experiments Avere carried on from October 28 to January 24, 1910. However, on January 24, 25, 20, 1910, fourteen triabs were made in which six germinations occurred. The time required was ten hours in the trials on Janu- ary 24 and twelve hours in the three successive tests on the follow- ing days. Although only a small number of tests were made dur- ing the 2nonth of January, it was quite evident that the conidia responded differently at this time than in the summer. In the tests made in January the zoospores lost their motility in less than one hour and developed long germ tubes. In none of the tests made before that time had germ tubes been seen. The different behavior of the conidia in. the late fall and winter as compared with spring and summer, are attributed to the loss of vitality of the host and fungus or to the improper maturing of the spores. Table II. — Effect ok Loweuixg Te.mi'ekature in Germixatiox op Conidia of Certain SrEciES of Cystopis, With Controls at Room Temperature. Species tested. Experiment at Low Tem-^ pekature. ] Controls at Room Tem- perature. Date. No. cul- tures. Period of re- frif^'-eration. Re- sults. No. cul- tures. 4 8 3 4 9 5 2 2 2* 2 4 Time con- tinued. Re- sults. Hrs. Temp. Urs. Temp. Augr. 11.. Aug. 18.. Cystopus bliti.. C. liliti 4 8 3 4 9 5 2 2* 0 4 1.5 2 2.5 10 7.3 1.6 5.5 24 5.5 28 2.6 21 18 19 18 17 18 10 8 11 10 20 + : + + + + + + + 1.5 o 2.5 10 7.3 2 5.5 24 23.75 28.5 14 25 27 27 25.5 27 28.5 22 22 22 21 27 - Autr. 20.. C. bliti Aug. 21.. C. bliti Aug. 23.. C. bliti Au^'. 16.. Sep. l.'>.. C. I'aiididiis C. bliti — Sep. 25. . Oct. 9.. Oct. 9.. Aug. 13.. C. caiulidus C. caiulidus C. candidus C. bliti — *Conidia placed on watch crystals, instead of slides. Summary of Table II As noted, no control experiments were kept in connection with the trials reported in Table I. The conidia were germinated as material for cytological study which will be reported on later. A second set of similar experiments (Table II) with controls, was 40 Wisconsin Experiment Station. carried on to demonstrate beyond question that chilling is neces sary for abundant germination. In this series of forty live cul- tures subjected to low temperatures, six failed to germinate. That is, about 85 per cent of the tests gave germination. In the controls, no germination was observed. The temperature during the period of refrigeration was very high for ice-box temperature in most of the experiments, due to the conditions explained in the summary of Table I. In all of these experiments the conidia were placed in tap water on the slides. In order to prevent too rapid evaporation in the trials at room temperature, the slides were laid on a metal stand placed on a wet plate under a small bell jar. The final observa- tions were made in both sets of experiments at the same time and the results recorded. The average room temperature at which the experiments were made varied from 21 to 28.° C, as can be seen in Table 11 ; while the temperature during the period of refrig- eration varied from 8'^ to 21° C. Otherwise the conditions were the same in both sets of experiments. These results show that a slight lowering of the temperature stimulates the conidia of Cystopus to germinate with the production of zoospores. Tab[,i<; III.- -Effect of Ligiit on tiik Gkrmination of tiii': Conidia ok Certain Species of Cystopus. CONTBOI.S ChILLBB. Cur. TUBE AT High Temi'er- ATURES. Date. No. of cul- tures Species tested. Period of refrig'ra'n. 9 P3 Period in flai'ic room. Period in liyht. 3 o d S o W s 3 o W a S V EH 3 01 Aug 11 4 4 3* 4* 2 2 2-1 Cystopus canciidus C. bliti 2. 2.60 2.5 9. 5.5 24. 5.5 21 20 18 17 10 8 11 + + -t- + + 8.T5 14. 1.5 4. 5.5 24. 6. 26 27 28 27 27 28 27.5 - Aug. 13.. 14. 2.5 7.33 5.5 24. 23.75 27 27 27 22 22 22 Autr. 20.. C. bliti Auyr. 23.. C. bliti Seul. 15.. C. bliti Sei)l. 25.. Oct. 9.... C. candidus C. candidus - * Pieces of leaves on which there were pustules, were laid on the slide, ■I E.xpLMinient.s carried on in wtltcli crystals. Summary of Table III The possible effect of light on the germination of the conidia of Cystopus was also tested. Controls were kept to ascertain the viability of the conidia and were chilled as described above. Twenty-one cultures were exposed to light and seventeen were Experiments on Spore Germination. 41 kept in darkness, in each case without chilling. All failed to germinate. The controls all germinated except two. These results are shown in. Table J II. All the experiments were carried on simultaneously and all the conditions were the same except that no bell jars were used to cover the controls while bell jars were used in the experiments in the light and in the dark. The conidia were taken from freshly matured pustules and placed in a drop of water on slides that were well cleaned. One series of cultures was kept in the diffused light of an ordinary laboratory where no direct sunlight fell upon them; the other series was kept in a dark room. The high temperatures, from August 11 to 25, have been explained in connection with Table I. The question might naturally be raised as to whether germination of the conidia in the ice box were not due to the dark moisture saturated atmosphere of the ice box rather than to the low tem- perature. These experiments answer this question. The series of seventeen cultures kept in the dark room were in a saturated atmosphere the same as the controls in tlie ice box. The only dif- ference was in the temperature which varied from 26° to 28° C. in the dark room and from 8° to 21° C. in the ice box. The ice box cultures all germinated except two ; while none of the seven- teen cultures in the dark room germinated. From these results it is clear that light is not a determining factor in germin;ition. It is also clear that a saturated atmosphere at high temperatures will not cause the germination of the conidia of Cystopus. Taui.e IV. — Effect of Ustncj Still Lower Te.mpeuatures in Gekmixa- Tiox OF Conidia of Certain Species of Cy«toihs. Date. No. Of ful- Agre of spores. Period of refriger- ation. ■ Re- sults. e)ut-floor tempera- 1 tures Hours. Temp. ture, minimum. Sept. !>.. 1 Sept. it.. 2 Sept. 9..I 1 Sepl.K?.. 3* Sept. 8.. 2 Sept. 4.. 1 + Sept. 27.. 1 1-t- Sept. 27..' 1+ C.vstopus bliti C cuhiciis C. candidus C. cuhicus C. cul)icus Young:.... Young.... 'S'oung Young,. . . ^'oung.. . . Young.... Young.... Young.... Old Young.... Young.,.. Young.... Old O'd l.lt) 8 S 2.5 4.5 lit. 5 27 27 10.75 10.75 3 3 3 3 —1 § S S s 10 10 12 12 + -t- + + + + + + + 15 15 15 18 12 C. bliti 14 C. cuhicus 6 ti Oct. 4.. 1 Oct. 4.. 1 C. cubieus C. bliti 6 6 Oct. 12.. 3 3 4 C. blitif —4 Oct. 12.. Oct. 16.. Oct. 16.. C. cubieus* C. cubieus? C. bliti* —4 0 0 *This experiment was made with a Van Tieghem cell and hanging drop. +Piec^^s of leaves with pustules in water on watch crystal. $Conidia taken from frozen leaves. iSThe slides in these experiments were laid on a block of ice in the ice box. 42 Wisconsin Experiment Station. Summary of Table IV The effects of lower teiiipcratures than those ordinarily obtained in the ice box were also tried. The results of these experiments are shown in Table IV. The slides were laid on blocks of ice, except in the case of the experiments on September 13 and October 12 and 16, wliich are further described below. Twelve cultures were made. In nine, the spores germinated while in three, there was no germination. These results are somewhat at variance with DeBary, who found the minimum to be 5° C. There can be no doubt from the above results that the minimum for germination is very near 0° C. To test the effect of still lower temperatures, three Van salt giving a temperature of 1° C. At the end of thirty minutes no germination had oceuiTcd. These slides were allowed to remain in the laboratory at 28° C. for ten hours after being re- moved from the freezing mixture, but no germination resulted. This indicates that a change from high to low and then back to high does not lead to germination. The effect of frost on the conidia of Cystopus outdoors was also tested. The conidia were collected October 12 and 16 from frozen leaves which had been allowed to thaw out in the laboratory. Twelve cultures were kept at 10° to 12° C. for three hours. Nine of the twelve cultures germinated and three did not, indicating that the conidia were not killed by a frost. Summary of Table V In view of the fact that germination was obtained in the ice box at temperatures above 20° C. in some cases, it was thought advisable to make a further study of the relation of temperature to spore germination at room temperatures. ])uriug the latter l)art of IMarch and first part of April, 11)10, further experiments were made to determine whether the conidia of Cystopus would germinate at room temperatures. As previously described, the experiments with cultures at green house temperatures in the summer of 1909 had given only a low percentage of germination. In fact gennination was observed in only one or two cases. In the new series of experiments, seventythree cultures w^ere made from March 27 to April 8, at temperatures varying from 17%° to 25° C. Controls were kept at ice box temperatures in cases Experiments ox Si'oin; dKimix a-iiov. 43 ^3 +++++++ M ffi w m in ifs M en c<5 .-o iM C'l M e-t in 1/5 irt •^ ci « ■■ +++ I I I I 1 +++++++++++ !+++*! +*ri*l++x*| 5^ ;^ i<: z', x: -° >^ ^ X Z ;', iJ X i'. z .t:.- 5 S £ os- nJ^-ii^xicZi-:^ X X X X <5 Z X X ?; X X ?: X Z ,< X Z X X X X X Z «^ X X X Z X X X Xj 05 •ad .2.3 CO cc fc CO I— i^ i^ i^ in in u'^ lO in in »n cOin-*-s-. cricio: 3"-2S"-502 (P*J CO bus t** g= 050>o»o;oMe'»oeoc<5-*'*-«»<'*in>ninininininincnci cj CI CO c<3 oi CO en ■■ ■«• -i ;D t- 00 SSi^^'s?^':^^?^ ?;<<<<<<<<<<<<<<<<<<<<<< 44 Wisconsin Experiment Station. where the temperature of the room was about 20° C. The room where this series of experiments was carried on was lighted by a skylight. It was steam heated and had only one door. The temperature during any one experiment never varied more than four degrees. The temperature was noted when the experiment was started and stopped and the average of the two readings are given in Table V. The conidia were obtained from stock cultures kept growing in the green house during the wdnter. The age of the conidia was determined as far as practicable by noting the time of appearance of the pustules and each time using conidia from the marked pustule. The plants serving as the source of conidia were placed on a separate bench where no wind could strike them, so that the conidia could not be blo'SATi away. Care was exercised in watering the plants not to jar the infected leaves. In this way it was possible to obtain conidia from pustules of known age. The spores were placed in tap water except on March 27 and April 4 when ice water was used. Twenty tests were made without covering the cultures with bell jars; while the remaining fiftythree were kept in a saturated atmosphere obtained by plac- ing the cultures under bell jars placed on a wet earthen plate. The approximate amount of evaporation in the cultures Avas ob- served and recorded in each case. In this case of seventythree tests at room temperature varying from 11%° to 25° C, twenty- six failed to germinate. Controls were kept at lower temperatures all of which germinated readily, indicating that the conidia used were normal. The time required for germination varied from forty-five minutes to four hours. The former being ;l:e shortest period at which germination was observed. Twenty tests were made in a non-saturated atmosphere and fifty- thre.3 in a saturated atmosphere. Nine of the cultures that were sub- jected to ordinary room conditions failed to germinate, and eleven of the cultures that were in a saturated atmosphere failed. This suggests that a saturated atmosphere is not necessary so long as the conidia are in water. The germination apparently took place as readily in cultures nearly dry r.s when no evapora- tion occurred. It should be noted still further that twenty-five cultures were made at temperatures above 20° C. in which 58 per cent germinated. The remaining forty-eight euitixro in the series were at temperatures between 17%° and 20° C. or below and 60 per cent germinated. From these results it is again clear that temperature is an important factor Experiments on Spoiuo Cermination. 45 ill the germination of the conidia of Cystopus. These results would have heen still more strikinf^ had the temperatures in th(; last series been nearer tlie optimum, 10'' C. Tabi.k— VI. OnsEuvATioNs ON Outdoor Gekwination ok Cvstoihs ON Radish and Salsiky. ,^ . Time, Date, ofdavr Sept. 9.. Sept. 11.. Sept. la.. So|)t. 2(".. Sept. 2(1.. Sept. 28.. Oct. 7.. Oct. 21.. Spocios of plant 9:15 8 StiKify.. ."^aNily... Siilsit\-... U;i(lisli.. SaWlv... Sal-.ir.v... Salsif.V... Sal.sify... Results observed. Zoospores. Present. abundant, abundant . almndutit . aliundant. al)undant . aljundant . Absent. al)undant. absent. Outdoor weather (•onditit)n.s. Ma.Y. Min. Preci- pita- tion. Charaiter Temp. Temp. of da.v. ! 1. s 0 C'loai'. 'M 10 .(4 Cloufly. 20 • 11 0 Clear. 21 10 0 Clear. 21 10 0 (.:]ear. 19 8 0 Clear. 19 (J 0 Clear. 1 10 0 22 Cloudy. Summary op Table VI Since it was found that the spores of Cystopus germinated readily in the laboratory when chilled, my attention was directed to the pos.sibility that chilling also favored germination on the host plant out doors when a dew was present. Observations were made on seven different days from September 9 to October 21. The observations were made on the leaves of salsify (Tragoporjon pcn^rifoliiis) and radish (Raphanus sativus) which were badly infected with Cystopus. The leaves were gatliered from 5 to 0 :ir) ;i. m. and were carried to the laboratory and examined for motile zoospores. The zoospores were found in every case except on October 7. On this date the observation Avas made at 9 :15 a. m. rather than at 8 :00 a. m., which was the usual time. It should be noted that no observations were made after 9 :15 a. m. wlieii the leaves were M'et. This should undoubtedly he done to fully prove that germination takes place only in the morning or when the temperature is low. The minimum for the days on which ob- servations were made varied from 5 to 11 2-3^ C. as seen from the table. The lowest temperature for the days in question came about sunrise as is usually the case.. It is quite clear from these observations that germination outdoors does take place at the time of lowest temperature for the day. The question as to whether germination takes place at any other time of the day will be further investigated this coming summer. 4/) "Wisconsin Exi-khimiont Station. J-iKst'i/rs WITH Other Si'ixmks of Pekonosi'okaceae It has been found that the favorable effect of chilling is not restricted to the conidia of Cystopus, but is manifest as well in the case of the conidia of other Oomycetes. I have incidentally tested the germination of the conidia of Plasmopara viticola at two different times and each time found that the conidia germ- inated readily at a temperature of about 10° C. Four tests were made with Peronospora effusa and germination resulted in two to four hours when kept at 10° C. Two trials were made with the conidia of Ffirono^pora parasitica and germination resulted in two and one half hours at 12° C. The conidia of Phytophthora infestans also germinated when chilled for 21^ hours at 12° C. The controls for each of the above experiments failed to germinate at room temperature. These results are based on only one to four trials Avitli each species, but show that chilling favors the germ- ination of conidia of other Oomycetes as well as those of Cysto- pus. Further experiments are in progress to detennine more fully the effect of chilling on the germination of the conidia of Plasmopara, Peronospora and Phytophthora. GROWING CYSTOPUS IN STOCK CULTURES UNDER GREEN HOUSE CONDITIONS It might naturally be supposed that so vigorous a parasite as Cystopus candidus would maintain itself quite easily under green house conditions, but it was soon evident that this is not the case. New infections were very reluctant to appear even when the host })lants were well infected in the beginning and the old infections after a time would die out. In order to maintain through the mnter vigorous stock cultures several methods were tried. I first attempted to learn the affect of vaiying the light in the case of well infected plants. In this experiment ten radish, three Amaranthus and two white nmstard plants, all of which were well infected with Cystopus were placed on an isolated bench in the green house Avhere only diffuse light was accessible. Over these ]ilants a large bell jnv was ]daced, l)ut free ventilation was |)i'()\i(lc(l by allowing llic licll jai- to rest on two bricks, one on each side of tlic ])ots containing the plants. On September 24-, 1909, fourteen other plants for controls were placed where they Fi XI 'Km MKN'I'S ON Sl'OKK Tl i;i:M I N ATM INT. 47 could get cliroct Kiiiiliulil, also iiiidcr siiiiilai-ly nciiI ilatcil Im'II jars. All these ])lants were; <^ro\vii in tlii-(!o inch pots, and wci-e four to six inches high and had four to eight leaves when the experiment was begun. Daily observations were made when the plants were watered. The experiments were continued from September 24 to October 29. It was quite evident that internal growth of the fungus took place both in the diffuse and in the direct light because hyper- trophy of the tissues about the infected areas occurred on the leaves of the radish plants as well as the production of the oospores in the leaves of the specimens of Amaranthus rctro- flextis. At no time did new pustules appear on any of the plants used. It was thus evident that Cystopus would grow under bell jars in the green house both in direct and diffuse light but at no time did new conidial pustules form. Some condition necessary for the production of conidia was lacking. Further experiments were made to learn whether the fungufj could be made to spread from an infected to a non-infected plant wlien in close contact and under the same conditions of moisture, light, and temperature described in connection with the above experiment. On Septem])er 27 a three inch pot that had three infected plants in it was buried in sand on a bench in the sunlight. Around the infected plants were seven pots containing twentyone plants not infected. A large bell jar was placed over the whole in such a way that plenty of air was accessible. On October 13, new infections were found on seven new leaves which had become infected during a period of sixteen days. The plants w^ere much healthier and more vigorous than the plants that were kept in the diffuse light in the above experiment. The spores were possibly carried from plant to plant by aphids or by slight currents of air that might enter imder the bell jar. The culture was maintained for tw^enty days more but the fungus gradually disappeared. In the following experiment attempts were made to produce new infections on a large number of plants and thus to retain and propagate the conidial stage. Six inch pots, in which there were about two hundred plants, Avere used. A more crowded condition was thus produced whicli showed itself more favorable for the fungus. When the plants were eight or ten days old, with oidy two cotyledons, they were inoculated with spores from C'ljstopu.'^ candidits. The conidia were placed in water and sprayed on with 48 Wisconsin Experiment Station. an atomizer. It was in this way easily possible to inoculate every cotyledon. The method of chilling for abundant spore germina- tion was also tried on these stock cultures. The pots containing the plants were placed in the ice box on a wet earthen plate over which a bell jar was inverted. The plants were usually left in the ice box from two to twelve hours, so that ample time might be provided for the conidia to germinate. December 6, two pots, A. and B., were treated as outlined above and became heavily in- fected by December 17. On this date the pots were placed on an isolated bench where plenty of sunlight was available for normal development 'of the plants, to learn whether or not the fungus Avould maintain itself under green house conditions when the plants were closely crowded together. On December 28 about lialf of the plants in the pots had died. These were carefully pulled out and more seed so"wti in these same pots among the living plants. In a short time a new crop of plants was at hand in the place where the dead plants had been. A supply of young host ])lants for the fungus was thus provided. On this same date, a third pot, C., thirteen days old and not infected, but otherwise exactly like the two described, was ph-^.ed as close to the other two as possible. Observations were made on January 8 and it was foiuid that five small pustules had appeared on the plants in pot C. On January 11, no changes Avcre observed in tlie extent of infection. Three pots, D., E., and F., of mixed radishes eight days old and iii»i infected were pnt with 1he above three, making six in all. The next day two six inch pots of mixed radishes, G. and II., nine days old, were placed with the rest, all in a group around the; infected plants. Observations on January 21 sliowed that pot C. had about 10 per cent of its leaves infected. Four more pots, I., J., K., and L., containing young seedlings eighteen days old, thickly sown in large pots, were added to the group. January 24 two of the pots, eight days old, D., and E., showed two pustules each. The following day the third, F., showed one pustule. On the same day the pots nine days old, G. and H., showed infection, each pot having four pustules. On February 5, two pustules had appeared on each of the two pots, I. and J., placed on the bench on J.a.nuary 21. On the fol- lowing day the third one, K., showed five pustules, and the fourth. L., nine pustules. Daily observations were made of the pots of plants as they were watered, The disease spread gradually, th^ EXPERIMKNTS OM SrOUE (JERMINATIONT. 40 number of infected areas iiicreasng as well as the numljcr of pustules in each area. As the cotyledons died the young leaves became infected. On February 2G, it was evident that the dis- ease had reached its maximum. It was estimated that 15 per cent of the leaves were infected. The spread of the fungus, however; had hardly kept pace with the growth of the plants. The leaves increased in number faster than the fungus spread. The disease gradually decreased from this time on. On .Marcii 20, it was found that about 5 per cent of the leaves were infected. The plants now were six inches high with six to ten leaves each and the number of pustules had decreased considerably. At least 40 per cent of the plants in the six inch pots were pulled out and seed sown among the remaining plants. Observations were made from time to time on the twelve pots of plants and the fungus was found to be gradually decreasing in amount. On April 8, a damping off fungus attacked the young plants just at the surface of the soil, causing them to fall over and dry up. The young crop had become infected in every pot, but because of the damping off fungus, the infection was not as ex- tensive as had been the case before. Three pots were so badly injured that they were discontinued. The large leaves of the plants in the pots were shading the surface of the soil too much for healthy gro"\\"th of the young plants. To relieve this condi- tion, the tops of all the plants in the pots were cut down even with those of the young plants. It is cjuite evident that this is a method that can be used to maintain the white rust under green house conditions. Two infected pots of crowded seedlings in- fected ten more standing near them. At no time was more than 15 per cent of the leaves infected and as the plants grew older, the amount of white rust decreased. The inoculation period in these experiments varied from eleven to sixteen days, due pos- sibl}' to lack of optimum conditions for germination. This Avas longer than was required when the stock cultures were chilled. An attempt was made to show whether or not the white rust could be maintained directly en the green house benches without chilling the seedlings in the beginning as was done in the above experiment. The radish seed was sown directly on the bench and not in pots. It was thought that possibly the disease might do better under these conditions. On November 1, four varieties of radish were sown in a box built on the bench, which was 8 inches wide and 3 feet long. When the plants were nine days old and 50 WiscoNsiK Experiment Station. showed two cotyledons they were sprayed with water and inocu- lated with Cystopus candidus spores from "Early scarlet globe" radish. On November 18, one leaf w^as found having one pus- tule. November 24, eleven cotyledons showed infection. From that time on the disease increased gradually. On December 16, it was estimated that about 300 of the 2000 plants showed white pustules. December 29, the pots of plants that had been inoculated but not chilled and the plants in the box on the bench showed about .the same degree of infection. A record of the temperature was kept by placing a self regis- tering thermometer as near the box of plants as possible. The range of temi)erature was about 12° C. for the first thirteen days. During the day the temperature was about 27° C. and at night about 15° C, showing that temperature conditions are some- times favorable in the greenhouse. The plants in this experiment were given the most favorable conditions possible and they showed normal growth. This was evident from the dark green color on the cotyledons and leaves. When the radish plants are sickly and not doing well in the green house the infected cotyledons and the leaves tend to become varie- gated and fall off. None of these symptoms were noted in the plants studied in this experiment. The experiment indicates that it is impossible to get any such development of the disease as was secured by chilling the ])lants at the time of inoculation. Still the disease appeared without chilling and increased in amount until alxmt 300 of the plants became infected. There- fore, it is possible to maintain the w^hite rust under green house conditions hy the method suggested in this experiment. Experiments on ?>pore rrERMiN.\TroNr. 51 T.\i5LK \^II.— Rkcoki) Oh- M.\iNT.\iNiN(i Tk\ [xfectku Stock Clii-tukem. cd a Variety of radish. Affe Cult. Time in ice box. Date Inoc. Date inf. Date re- planted. Results. .1. Nf I'his Fltra cla.vs 5 18 42 46 77 ^3 118 148 1G4 192 Hrs. Hi 0 6 19 12 Sept. 29 Oct. 12 Nov. 5 Dec. 17 Feb. 23 Oct. 6 Nov. 9 ,lan. 24 Apr. 8 Oct. J 2 Dec. 11 Jan. 13 Mch. 11 Heavy inf. Vounsr leaves well infected. Well infected. Second crop heavil.v infected. Well infected. Heavy inf. Old plants with little inf.: yountrer plants, plenty of inf. Old 1)1. about toljloss- oni. Tendiny to shade young pi. Not well inf. Two old pi. in blos- som. Pot filled with soil. T^' Crimson Giant 9 43 53 80 121 105 6 19 12 Dec. 25 Dec. 17 Feb. 23 Nov. 9 Dec. 31 Mch. 24 Apr. 8 Dec. 7 Jan. 13 Mch, 11 Well infected. Well infected. Young heavily inf. Well infected. Low in inf. Young crop well inf. 6 big plants about to blossom, other pi. shaded. Poor inf. One pi. in tlower. Others in bud. Poor inf. L. New White Cliint'se No Plus Ultra 6 27 41 57 6 20 12 Dec. 17 Jan. 13 Feb. 23 Dec. 29 .Ian. 13 Jan, 13 Feb. 23 Mch. 11 Well infected. Disease increased in amount. Well infected. Young crop well- inf. M. Scarlet turnip, white tip. Ne Plus Ultra 5 86 6 12 Dec. 17 I'eb. 23 Jan. 2 Nov. 1 Jan. 13 Mch. 11 Well infected. Coty dead. Young pi. Ne Plus Ultra l)ecoming inf. Young pi. well inf. Infection decreased. Young plant inf.. old also tho not as much as young. N Old's Snowball 9 83 8.5 12 Dec. 21 Feb. 23 ,lan. 2 .Jan. 13 Well infected. Well infected. Well inf. Pot brolcen. O. Cincinnati Market NePlus Ultra 7 95 23 Dec. 13 Dec. 26 Mch. 11 Well infected. Pi, 5 in. high. b% of Pi. inf. 1' Mixed radish 8 67 7 Jan. 11 22 Feb. 23 22 Mch. Jan. 22 Well infected. Ne Plus Ultra Feb. 23 Well infected. Infection low. 52 Wisconsin Experiment Station. Tablk Vil Continued. — MAiNTAiJfiNG Tkn Infected Stock Cux^tures as Variel.v of li^idish. Age Cult. Time in ice lio.\. Date inoc. Date inf. Date re-r Rpsnlts planted. Ke;,ults. Q. our.s Twenty I)a.v 7 95 24 9 22 Dec. 13 Feb. 25 Mcli. 11 Dec, 23 Jan. 22 ' Well infected. Jan. 13 ; Less than ' of pl.inf. You ntr crop well inf. •R Winter 1 Jan. 3 Jan. 12 l'\'l). 23 Well. 11 ; 1 Well infected. 11 78 24 22 Feb. 25 s 9 67 24 24 7 Jan. 12 Feb. 23 Mch.ll Jan. 20 Feb. 25 Well infected. 75% of leaxes inf. 75''b of leaves inf. Decreased in inf. Summary of Table VII On the basis of the experiment with stock cultures just de- scribed the following method for maintaining stock cultures was marked out. Good infections were secured by spraying the spores on the young seedlings in a crowded six inch pot with an atomizer and then setting the pot on a wet earthen plate over which a bell jar was inverted. Under these conditions the pots of plants were placed in the ice box where the temperature was about 10° C, long enough to allow the conidia to germinate. After this they were placed on a bench in the green house. As the fungus continued to kill the young seedlings, others were supplied by sowing more seed among the remaining plants. When this new crop came up, the pots were again placed in the ice box as described above so that the young plants might become infected. It will be convenient in the following work to speak of a six inch pot thickly sown with radish plants infected with Cystopus candidus, as a stock culture. Nine of these stock cul- tures w^ere maintained in this experiment, each designated by a capital letter from J to S. In all the stock cultures from J to S good infections were maintained by chilling and reseeding when- ever the disease showed signs of disappearing. These stock cul- tures became the source for the conidia used in the subsequent experiments up to the time when the jilants were about to bloom. In stock cultures J and K good infections were kept in the same pot for 192 and 165 days respectively. On March 8, as a result of the pots becoming so filled with soil from continued reseeding, it was impossible to grow more new plants among the old ones. Experiments on Spore Germination. 53 At this time the old plants in the stock cultures were in blossom and the leaves were large and broad. This condition was not con- ducive to the growth of young plants in the same pot. Even in this stage of development, the plants were not free from disease although the pustules Avere much less abundant than when the plants were younger. The stock cultures from L to S were all younger than J and K, and on j\farch 11, all the plants were about to blossom. The white rust was evident in every stock culture but was not as plentiful as when the plants were smaller. Had it not been that it was desired to make a study of the pro- duction of oospores, the larger plants might have been cut off, and the culture inoculated again whenever the disease showed signs of marked decrease in amount. Stock culture J was chilled in the ice box five times, reseeded four times and maintained a good infection for 192 days or until the plants blossomed. Stock culture K was chilled in the ice box three times and re- seeded three times during the 165 days. Infected plants were always abundant until the old plants blossomed. The remaining stock cultures L to S were not as old, but in every case tliey were chilled at least twice and reseeded from one to three times. Here also, well infected stock cultures were main- tained until March 11, when the plants grew tall and began to blossom. This is undoubtedly a method by which white rust can be maintained abundantly under green house conditions with the use of only a few pots. The full data as to these cul- tures are given in Table V. To fully test the value of chilling in maintaining stock cultures of Cystopus, parallel cultures, chilled and not chilled, Avere made during a period of 116 days. Chilling at the start a crowded stock culture of Ixaphamis sativus such as is described above and inocu- lated with Cystopus canclidus from the radish undoubtedly causes a larger per cent of infection than when the stock cultures are not chilled. Six inch pots were thickly seeded with radish so that each pot contained about two hundred seedlings when it was about ten dajn old. The plates except YII and YIII are photographs of such cultures. At this time the stock cultures consisting of two hundred seedlings, were inoculated with con- idia of Cysiopiis candichis from Tiaplianus sativvs. The conidia were sprayed on the young seedlings with an atomizer. Fol- lowing this treatment, the stock cultures were placed on a wet 54 AViscoNsiN Experiment Station. ^ y. o o K H ^ o a H <) U D o O Q S^ W >— 1 ►J ^ K r-) 05 H 'SO •0 0 ^ CO §■ 0 '-< 'A 0 0 u w ai H M « j; U> H ►q U 0 H 0) O O *~ • QO 00 iC S^ 5; ^ ^^ ift I— »f5 ^ t '■ j5 i; S; ^1 «■' M — cj c! CM « e>' c 5^ '-:'-. 1-5 -ii^^cc-CHjC;:; o aj c-i ^ 00 CO 22 oi CI 'i) 't- CQ ro •_* • -cici-^d : — . ci -^ -^^ « — ' M 0^ IN — ( -^ •"• 00 00 in ^J g, ^ ;z; t_ =' d ^ !i « '-j' d =^ c *-' 1^ i; 53 ^ ^ c c d d d ■ — ^■lM■*!i:lU<^"^Mm- '-- rv^ ,_ ,_, j,i '^ ^^ — ■ — ^ ^» ^1 - > o d d c ^ ^ - '^ d u d «3 o !i j; !:' j; d d d d d O^ rt 3 rt =5 3;^ CS^ ri aj 0; *JS^ rt rt rt d rt rt o — o om if^ o o tc Oj 1(5 o <: noooo — — O ^ _■ ,, „ 50 to 00 b- 00 GC 1-- -^ -* to O re CQ t- to '-0 iftlft ift iftl- 1- CI t^ t^ d '^ 00— -f— C-IOOO — TICOTIOO — oc~occ.o i^ Q 8 IB ^O — — OCIO-*M — C>10'HM5>10 5'|-r>l^-HC>t~-«l Ed a ,-^^,-l^.-,CI ^^-,^^H^— *^CI^CI^^-^" ^^ -^ o.r;u^.f.>«u2-o=.ooo = oM222t:;g;S?".?'.?3^SgSS;'-g'"S'""S'-S'" oi rt >■ > t-' >■ >■ >■ >■ >■ >■ >■ >■ > >■ >■ >■ >■ >■ >■ i-' >■ > > > > >■ >■ > > > -J > -J > -J -J :>■ -J >' JO OOOOOOOCOOOOOOOCOOOOOOOOOOOOOl/OoOiC-SOOai ^Z>^?;^zzz;^.?;/'/^>^z<'.^J5>5?;^.zz;^2:/^;<>^5';<;;:;/5=;^£Cx;iZi. 2, 05 to -O -a — ' -- — • C> CI CI « CI CJ CI C» M I- ^ 5 o a! s ffl-3 00 S to in — -" — ■ ci ci CI ci « ci CI CI cj t- 00 00 Oi' to to in to to to' -^ — — — — ---------- — ''.'Z'^'^'^ZX'^ n ^^ ti^ ^^^^^^ «»m ■■r tx yi a- '2 lO ifl lO .ft .n in .re in un in in in in in in o in un in in in in \a_ D O cj 31^ .: !>; od -^' -^ « — -- — — — — — in .f^ in .n irt in in in in ■* in in in t- C'j M « ci M oi ci c-i ci 2 " ?^ -^ 1-1 w ci cs^w*j*j®^*j*ii-w_^.-_ — — — *--. o ^ .^ „^ ^ __ .^ _• _M ^ .^ .iM —4 — « 1 -^ ~^ "^ •"— * ^ — < ^ —M -^ — * ^^ . >•« ^ >-• — ^ — • — ^ — ■ — ^ — • "— < —• — ^ —1 c3 S^ -J -J O „ -J - ^ 03 •/! 73 r, /; •/! /^ -y) ry- y; -yi y; y; yj -/■ 1— >^ -H S y; -yj X T. — *-* W 7) X y: yi X X X X 01 x x CB 3, S§ S33 = 3333 = S35 = = S, ©3333,,, S3S333S33SS33? ««^.~^ — .^.M^-4.— < — _-.^ — —■ — ^ w — ^ /,__ — — -^ w — ^ — — 1— . — ,_—.—, — ___._--. ^ =£" ChZ.Q< — ^ — ^S- — — — — — — — .C-- .z: =:_i.i. — .t;.t:c- — — — — — — — — — — — — 3 a)a>SJiii.3,®ajiJ:^a,aj:,:/i ^'~i-c;-. ^:^w::.-ivi:i.--i.*a)i. ?r; /^ j^ z', ?^ 2^ ?; Z >'. -^ >'. ^, ^ ;/'. i', ;? ^ j; '^ 5C ^ /' ' ' ' 1 ! — ?::::: '■ '• ■ a ■ 2, :d r^ ^ • • * • • ■ * * 3 * 3 : : I b I ^ • a "^ . I . I 1 • ■ ■ 3 • 3 ^ Q : ^ ^ ' ^ ! I • . : -.T* --'rr* 5 2 2 5 a Z c 0 c: a ■J a S ' 3 J a, it - tc 'x C x'S 0 _ c ; "3 a |l — ■I p — > r. - H : : : : . ]7 X - jl ^ a ^ • >. =3 : 3 Sc S ri = s = •J X ^M -^^..^--.^ ,-,_^.i_ , , . —4, ',3**C''C— ■— '„-'^"„-*-i'', -^--, ''.«*-'*„-*->.t^ llllilllllllllllllilllllisllllllsllllll '*^^ft^r5 0saiO^'H^^^^^^C:a55sOiWC^I^^lwc•^c•»^-^-^*«M«SJ5«Mcoc^«« d ?aCMwc*i wojco ,— 1,— ,^-,^N^^.-«.— 1.— ^s^— CJc^MOidWricjwci *^-;*j j«.j^ > s-' >■ •> > > >■ >■ > > > &•' >■ > > > > > > > >■ > ■> > > > > > > > > s u ut3^ 0 0 0 cooooooocooooooooooocooooo 2 00 0 = 5 COCSOSC^Z^:?'.>^Z?^^;^i^i^ZZ5^i^2;>^;<;^/^Z?:S5Z;s?:Z^2:V!:Z-. .■J'O «°l| w to e») «>« M eo ■* N e>> «"« et w •»« m m N to «> M 'T M ^ ■«<•*■«<••♦ »^ N •»» T-t -< N M «-. M ♦.1 ' 60 Wisconsin Experiment Station. SUSCEPTIBILITY OF OTHER CRUCIFERS TO CYSTO- PUS CANDIDUS FROM THE RADISH Tests were made to leam if the spores of Cystopus candidus from the radish can infect turnips (Brassica rapa). The tur- nip plants used in this series of experiments were grown in the same way as the radish plants, as explained in connection with Table VII. Three healthy plants were allowed to groAv in each pot and these were inoculated Ivy spraying them with the spores of Cystopus candidus from the radish. The spores were sprayed on with an atomizer and the plants chilled in the usual manner. As shown in Table X, ten varieties of turnips (Brassica rapa) were inoculated with conidia of Cystopus candidus from the radish, variety Ne Plus Ultra. In each variety, at least nine plants were inoculated, while in the case of Snowball, eighteen plants were tested. This would mean that eighteen or more cotyledons were inoculated. These experiments extended from November 13 to January 28, and at no time did any infection result, although in every trial the controls were infected. It was impossible to infect any of the ten varieties of turnips with Cystopus candidus from Raphanus sativus, variety Ne Plus Ultra. These results suggest that there may be a physiological species of Cystopus candidus occurring on radish and turnip respectively. Tai;i,r X. — Ri<;r-ATrvK Stjscioi'TrniriTTY ok Diffeurnt Vahieties Braxxicn rapa 'I'o Gj^xtopus candidus from Raplianus Hatirus, Variety Ne Plus Ultka. '6 0) 0) ^ ■^ Dat e. o o o '/T, ^ Nov. .30. 3 Nov 30. 3 Deo. 4.. 3 Dec. fi.. 3 Dec. 2S. 3 Dec- 28. 3 Dec. 28. 3 Dec. 28. 3 Dec. 28. 3 .Ian 10. 3 .Ian. '■><> 2 •Ian. 28. 1 .fan. 2S. ITosl I'huit. Yellow al)erclcon Olfls" heavy couper , (ioldon ball White flat dutch Eai-l.v white niilan (^ovv horn Extra early purple top milan rniple top white erlobo White esrsr Snowt»all, Snowball Early white milan Golden ball Source of conidia. Ne Pln.s Ne Plus Ne Plus No Plus Ne IMus Ne Plus Ne IMus Ne T^lus Ne Plus Ne Plus \f> IMus \e IMus Ne Plus TTltra. intra, ritra. Ultra. Ultra. Ultra. Ultra. Ultra. Ultra. Ultra, intra. Ultra. Ultra. 6 o Per od Results. c refi-ig- era- o o tion. ..; .^ ..-» d a o y d 3 O d s Date. o 55 55 - H *- 9 5 4 16 Dec. 17.. 0 9 5 4 16 Dec. 17.. 0 9 .5 ;> 17 Dec. 22.. 0 9 5 S 11 Dec. 17.. 0 9 5 17 9 •Ian. 13.. 0 9 5 17 9 .Ian. 13.. 0 9 5 17 0 .Ian. 13.. 0 9 5 17 (1 .Ian. 13.. 0 9 a 17 9 .Ian. 13.. 0 9 5 23 10 .Tan. 22.. 0 9 f) 2fi,5 12 Feb. 15.. 0 6 5 26.5 12 Feb. 15.. 0 3 5 26.5 12 Feb. 15.. 0 Tnf. Inf. Tnf. Tnf. Tnf. Inf. Inf. Tnf. Inf. Tnf. Tnf. Tnf. Tnf. Experiments on Spore Germination. 61 Summary op Table XI An attempt was made to infect three different varieties of rutal)au;a {Brassica campcstris) , namely: Olds' lmi)!'OV(Hl Purple Top, New Necklace, Olds' Ijai'ge White, with the conidia from the radish, variety Ne Plus Ultra. Three seedlings were grown in two-inch pots in the same manner as described in connection with Table IX. Controls were kept in every experiment jind always became infected. Final observations were made in the ease of the different varieties of rutabaga several days after the controls had become infected. This was done so as to exclude any possibility of overlooking the disease, which possibly might take longer to develop on the rutabaga. The age of the seedlings inoculated varied from eleven to twentynine days. The plants had one to five leaves. Both coty- ledons and leaves were inoculated. In all, sixteen separate trials were made. In each trial, three plants were inoculated, making a total of fortyeight plants inoculated, and in no case did infec- tion result. These experiments extended from September 25, 1909, to Januaiy 28, 1910. Plants of different ages were tested. It is quite evident that Cystopus candidus occunnng on Bapha- vus sativus, variety Ne Plus Ultra, will not grow on Brassica campestris, variety Olds' Improved Purple Top, New Necklace and Olds' Long "White. Table XI.— Relative Suscrptibility op Different Varieties Brassica cam- pestris TO Cystopus candidus from Raphanus satims VariktyNe Plus Ultra. Period Results. -d s o o c OF Re- FRI- r/l Date. Host plant. Source of conidia. S 71 GEBA- TION. C •>> oi 3 OS- 'S p^ o 0 0 '£ . o ri- oo ^ i. 0 ?: ^; Z K H ^ Z 3 Dec. 4 Olds' improved purple top. Ne Plus T'ltra. 0 5 5 17 T)ec. 22. 0 Inf. 3 .Ian. 10 New neckless Ne Plus TTltra. 9 a U ;o Jan. 0 Inf. 3 .Ian. .Tan. 10. Olds- Olds' lai'tre white Ne Ne PlusTTltra. Plus TTltra. 9 9 26..-) 10 12 Jan. Feb 5"! 0 0 Inf 3 larsre white Inf 3 Jan. 00 New neckle.ss .Ne T'lus T'ltra. 9 5 24. ^ 12 Feb. 5. . 0 Int 1 Jan. ?S Olds- lartre white Ne Plus T'ltra. 3 a 20 12 Feb. lo. 0 Inf 0 Sept ?3 Olds' larsre white Ne Plus intra. 6 5 9.3 12 Oct. 8... (I Inf 1... Seut 25. Olds' larg-e white .Ne Plus Ultra. 3 a B 9 Oct. y... 0 Inf. 62 "Wisconsin Experiment Station. Summary of Table XII Experiments were undertaken to determine whether Cystopits candidus from the radish could be made to grow on different varieties of cabbage (Brassica oleracea). About twenty plants were grown in each three-inch pot. Fifteen different varieties of cabbage were tested from January 10 to March 13, 1910, in- oculations were made in the usual way by spraying the spores on with an atomizer and placing the pots containing the plants in the ice box under a bell jar on a wet plate. Three tests were made on each variety, so that at least sixty plants of each variety were inoculated. Four of the varieties tested became infected. All Head Early showed two cotyledons infected out of a total of sixty plants tested. The Volga was the most susceptible to Cyslopus candidus. Nine cotyledons became infected out of a total of sixty. One cotyledon became infected out of sixty plants of Olds' Selected Early. Of the sixty plants of Olds' Ballhead inoculated, two cotyledons were infected. In all the fifteen varieties of cabbage tested, nine hundred plants and of course nearly twice that number of cotyledons were inoculated and only fourteen cotyledons became infected. The results show that it is possible to infect at least four of the fifteen varieties of cabbage tested with Cystopus candidus from the radish, yet not to any marked extent. The fungus had a more marked effect on the cotyledons of the ca])])age than on those of the normal liost, the radish. The in- fections found were always on sickly looking cotyledons and in two or three days after the pustides appeared the cotyledons v.ould dry up and drop off. In the radish plants used as con- trols the cotyledons lived for two or three weeks after becoming infected. This would mean that the fungus was more virulent on the cabbage, and immediately killed the host when it was able to establish itself, or that it attacked only the sickly, weak cotyledons. I am inclined to believe that the fungus attacked all of* the cotyledons alike, but the further development of the fungus was overcome by the host in all the cases except in the few infections described above. This is of course a point that needs further investigation ])cfore definite information can be had. To still further determine the susceptibility of different varie- ties of Brassica oleracea to Cystopus candidus of the radish, an- Experiments on Spore Germination. 63 Op o«-: O ed O S o a. cir. o — ciocooocooooc — OOC.OOOOOO — coo-/ ^ — — tc t^ t- o <- 1- 1^ o o -<• o o o o o -r -* -* -T ■* cn« in in >n ■?: o in >n '■': '_2 'j2 2 2 2 t~-i-t^t-«~i-t--i -st^-atc-n -a -o ■£: -^ -o -a -^ ^^^^'^•^^^ • '■^'^•^'^'^''O'O'S iCii)a)a>s»aaia;o OOOOOOOOOOOOOOOOOOOO00>ntDOOOOOOOOOOOO cScacS£S:3eSo3caci3:Se3eSoSsSeSsScSe3eScS^cjcaoSeS •r. X r. X x x X X ;il; X X X X X X X X X X J23 ^ n s s =; K :i cS i-a 01 cj X :J-=a^SS . ;- — x. - Vj >; . - . _ _ C !»cj±i! r;-r 5,of "j: $•? :SjS 4) X x r; X p:^ — — ■/. ;; / X <:tHX~3K3EfciK5-5xo -^ r> ■^ i. o a t£ ^ — - ^ ta. £ S X •? 3 ■J. - X r^ >i * 0. :5 n j; Z^ K ^ ^^ =■3 ?C3 w X -r c = — — -' ? x-r' 3 -H -Ifx ti i." i." t," u t; ti fc. t.' - « 1. 1, u « t^ t- - - - _• ^* _• - I. >> >i >, ti t.) i^ >. >. >> >> cftccM — ?!-"?> cccj»HN«rt«-«Mm—<^^:mi.- CI ri ooooooooooooooooc^ioooooooooooo _••-< '/] ^ »--*oc a; c V. t^ >. >i ^ t; I.' i: u u ti -■ t: -' -j: c -^ c c c c C'^.>> >, sj bi t; c c;B!tsescia2.=.=.a=.s:iazaa^::aaacdceoss32,a sssss«t:<i >> t>> >S >. C fci fc. P]XI'i;ivlMKNTS ON SfOIU': (iKKMINATlONr. 07 were eleven days old they were inoculated with conidia from the radish. All of the experiments showed infection on May 18. From 35 to 50 per cent of the total iuiml)er of cotyledons were infected as shown in Tabic XIIT. Only a small number of leaves ])ccame infected due to the fact that not many of the plants had leaves when they were inoculated. The results in these experiments show clearly that the white mustard is sus- ceptible to the cystopus that occurs on the connnon radish. Summary of Table XIV During the spring and summer of 1910, various other hosts of Cystopus were grown from seed in the green house and inocul- ated as already described. The plants were grown in three inch pets and were vigorous and healthy when inoculated. In every species, at least two separate tests were made. The number of plants in each pot varied from one to twenty. If the plants to b*- inoculated were large before inoculated, enough were removed to allow the remainder to develop normally. Twenty-six plants cf Capsella Bursa-pastoris, nineteen plants of Sisynibrium offici- nale, twenty plants of S. altissiynum, forty plants of Lcpidiurn sativum, forty plants of Nasturtium officinale, sixty plants of Brassica nigra, and thirty plants of Ih(ri-< coronata were inocu- Ic.ted and no infections were observed. Inoculations on radish were used as controls in every case and infections were always obtained. Further tests must be made with these plants before positive statements can be made as to their susceptibility to the spores from Cystopus from the radish. It should be noted that a larger percentage of infection was obtained in the spring than in the fall. Thirteen tests were made in the fall of 1909 from October 20 to December 28 in which thirtynine plants were inoculated and only three cotyle- dons and six leaves became infected. "While in the tests in the spring from 10 to 50 per cent of the cotyledons became infected. The difference in extent of infection was possibly due to a differ- ence in the host plants. The white mustard seedlings grown in the fall were not as vigorous as those obtained in the spring. 68 Wisconsin Experiment Station. Table XIV. — SuscEiTiBiLrrv op Other Crucifers to Cystopus From Radish. candUiu-i Date. Host Plant. Source of Coiiiclia. 3 Period 3 'J of refri- Results. o ger- "-" ation. '£, S 2l Si2 ^ =* <)-i a o -J t 0) OJ a -^ Date. « o o 3 HH ^• i5 ^ HI ^=^ o Con- trols Inf. June 9. June 9. May a,), Apr. 29. Apr. 2^. Apr. 2». Mav (i. An- 24. Apr. 31). Capsella Bursa-pastorls Ne Plus Ultra. Sisymljriuin officinale...! Ne Plus Ultra. Sisymi)rium officinale. ..j Ne Pius Ultra. Lepidium sativum Ne Plus Ulti'a. Brasslca niirra Ne Plus Ultra. Capiella Bursa-pastorls. Ne Plus TMtra. Nasturiurn officinale | Ne Plus Ultra. [Ijaris umbellatta | Ne Plus Ulti'a. Sisymbrium altissimum. Ne Plus lUti-a. 21 20 8 14 June 24. 0 9 20 8 14 June 24. 0 10 200 22.5 i7 June 8. 0 ! 40 12 27.5 9 May 25. 0 tf) 12 27.5 9 May 25. 0 5 12 27.5 9 May 25. 0 40 10 5 10 .May 21. 0 SO 200 14 12 Sep. 9. 0 41) 20 U 13 May 13. 0 Inf. Inf. Inf. Tnf. Inf. Inf. Inf. Inf. Inf. DISCUSSION AND CONCLUSION Germination of the Conidia These studies with various species of Cystopus and other Oomycetes have shown that germination of the conidia is con- trolled by certain factors of which temperature is the most im- portant. Prevost (1807:33) in his studies made over a century ago states that at a temperature of 12° to 16° R. Cijs- topus candidus spores sometimes germinated in 40 to 45 minutes, whereas they ordinarily required from one to two hours. DeBary (1863:14) found the conidia of Cystopus to germinate at tem- peratures ranging from 5° to 25° C. My own results have led me to conclude that chilling the spores to a relatively low tem- perature is necessary for the most vigorous germination. In our preliminary series of over one hundred cultures of Cystopus spores at green-house temperature, 22 to° 33° C. during the sum- mer, only a very low percentage of germination occurred. In none of the cultures were zoospores observed and the only indi- cation of germination was the presence of an occasional empty sporange which may possibly have been empty before the spores were brought into the laboratory. In a later series (Table II) during the summer, including fcrtyfive cultures chilled with controls at room temperature, it Experiments on Spore Germination. 69 was found tliat 85 per cent of the chilled cultures germinated, whereas none of the cultures kept at higher temperatures germ- inated. In a third series (Table V) of seventy three cultures during the following spring where about one third were held at joom temperature (i. e. above 20° C.) scanty germination occurred in 48% of the cultures,^ whereas 69% of the cultures which were kept at a temperature below 20° C. showed abundant germination. This last series shows that germination may occur at room temperatures and above, as has already been pointed out by DeBary (1863:14), Zalewski (1883:215), Biisgen (1882:22), and Eberhardt (1904:614). But it was also clear that the percentage of germination was much increased by using lower temperatures as was furtheo* shown in the behavior of the stock cultures described in connection with Table VIII. The chilled stock cultures became heavily infected while the controls not chilled showed only a low percentage of infection as is conclusively shown by referring to plates I and II. Two things are clearly indicated from our results. First, and most important; temperature exercises a marked influence upon germination ; second, this influence was more marked with spores obtained in the late summer and autumn than with those de- veloped in the spring. "We interpret this latter difference as probably due to the greater vigor of host and fungus during the spring months. Be this as it may, it was at all seasons evident that comparatively low temperatures were necessary to induce strong normal germination. Various media were used for germination trials Avith spores of Cystopus; such as rain water, distilled water, tap water, ex- tracts of the host, sugar solutions and certain nutrient media. No germination was obtained in any medium except water and no marked difference was noted in the percentage of germina- ticii obtained whether it was rain water, distilled water or tap water from Lake INIendota. In all of the subsequent tests this tap water was used. Where spores were immersed in a drop of water, the condition of the surrounding atmosphere, whether saturated or dry, and the amount of evaporation had no effect on the percentage of germination. Other physical changes in the culture containing the spores, such as diffusion of the drop and changes in surface tension, were of no consequence. No attempt has been made to determine vnth exactness the optimum temperature for germination. Indeed, the variations 70 Wisconsin Experiment Station, a&sociated with seasonal and host conditions as just noted and witli maturity of the spores may preclude perfectly definite con- cjusions upon this j)oiut. It was at least strongly suggested by the large number of experiments made in the laboratory and from the observations made out doors that the optimum for normal spores produced under the best conditions was about 10° C. Results as to the maximum temperature of gennination tend to substantiate DeBary, who found, as previously noted, that the maximum temperature was 25° C. In a series of over one hundred cultures carried on in the greenhouse during the months of July and August of 1909 at temperatures varying from 22° to 33° no germination was obtained. Again in a later series (Table V) in the spring of 1910 scanty germination was secured at 25° C. Although these experiments wore not planned espe- cially to test this point, yet they show that the maximum temp spores of Cystopus on the various species of Raphanus are quite limited Imt it may be that Bras.'^ica alha serves as the bridging species. These arc questions lliat can lie fnlly determined only by a large Experiments on Spore Germination. 77 number of cross inoculations with the spores from various hosts of Cystopus. As has been pointed out, the infections that were secured on Bvassica alba and B. oleravea with the conidia from radisli dif- fered in appearance from those usually occurring on the radish. Tlie infection on the radish is vigorous, causing marked hyper- trophy and developing large, white, plump pustules. On the white mustard and cabbage this was not the case. No hyper- trophy occurred and the i)ustules were small, showing none of tiie signs of vigor evident on the radish. Not only was there a marked difference in the appearance of the fungus pustules on the hovsts in question but also in the effect upon them. The fun- gus killed the host tissues very much faster on the white mustard and cabbage than on the radish. A possible explanation of these results would be that the infection of the white mustard and cabbage occurs only in the most vigorous cotyledons; that in tJiese the fungus is able to overcome the host cells and persist in only a few cases and that in such, the host cells when overcome die immediately. In my observations, plants infected with aphids or thrips seem to be quite immune to Cystopus. At no time w'as I able to get infection on a plant that was badly infected with insects. Reed (1907 :381) also found that it was quite impossible to infect grain seedlings with mildew that were already infected with thrips. This was more evident in the ease of wild plants such as Capsella, Lepidium and Sisymbrium than in the case of SL'ch cultivated plants as radishes and mustards. The lack of irfection can not be attributed to the aphids eating the spores, since some of the plants w^ere fumigated, killing the insects and then inoculated, with similar results. These facts lend support to Cook's (1911:624) view that plants injured by plant or ani- mal parasites develop an excess of tanin which causes more or less immunity. Not only was it quite impossible to infect plants attacked by insects, but likewise, plants that showed signs of not being vigorous and healthy from other causes. It was also im- possible to infect wild and cultivated seedlings that showed yel- lowing of the cotyledons and first true leaves. This was also true of the more mature plants. If for any reason a stock cul- ture of radishes showed signs of not being healthy and vigorous the extent of infection was at once reduced. As has been 78 Wisconsin Experiment StatioisT. stated earlier, Eberhardt believed that the various hosts of Cystopus do not at all stages of development show the same susceptibility for the fungus. Nowhere does Eberhardt have any data to substantiate this conclusion, nor has he taken into consideration host abnormalities as a factor influencing the question of susceptibility. From my results it is at least very evident that Cystopus reacts differently to healthy and sickly j)lants respectively. It is impossible to infect Capsella Bursa-pastoris, Lepidium virginicum, or Sisyinhrium officinale when the plants are not vigorous and healthy. Many attempts were made during the fall of 1909 to infect Sisymbrium officinale with Cystopus can- didus from the same host but the infections were very scanty. Out of the fourteen experiments on fiftysix plants, only eight plants became infected. I attribute this to the weakness of the plants that were grown at that time. In every case, it was the largest and healthiest looking plants in the lot which took the disease. Although I have not succeeded in proving entirely to my own satisfaction that the extent of infection is dependent upon the vitality of the liost; yet jt seems highly probable that this is the case. Reed (1907:381) has fully described a similar relation between the host and fungus in the grain mildews. Since there is this evidence in both the mildews and white rusts that sickly hosts do not readily become infected, in testing a species for socalled physiological species, all possible care should be exercised in cases where plants are used as hosts that are at all difficult to grow. Failure to infect may be due to weakness of the host plant. Experiments on Spore Germination. 79 SUMMARY The studies outlined in the preceding pages were carried on chiefly witli Cystopus candidus as it occurs on the common radish, l.'aphitnus salivns. The leading problems considered are: Con ditioiis influencing gennination of the conidia; conditions in- fluencing iut'ection ; and, the occurrence of so-('alIed physiological speci( s of Cijslopus candidus on the various crncifers. Gekaiination op Conidia When the conidia are placed in water they germinate better a strikingly low than at high temperatures. The optimum was not definitely determined, but the results tend to show that it was 10° C. The minimum temperature of germination was very near zero, while the maximum was, as DeBary has shown, about 25° C. ; It was found that water is the most favorable medium for germination. No germination was obtained on various nutritive culture media. The time required from the immersion of the conidia to the escape of the zoospores usually varied from two to ten hours. The shortest period in which such germination was observed was 45 minutes. Environmental factors, season and host vitality, seemed to in- fluence the time required for the spores to germinate. It was strongly suggested that the time required in spring and suimner is sliorter than in the late fall and winter. No difference was observed in the time or percentage of germ- ination which occurred in light as compared with darkness. Spores obtained from leaves after a killing frost germinated. Such factoi's as evaporation, surface tension, and diffusion of the drop containing the conidia did not influence the percent- age of germination. The conidia germinated as readil}^ in a non- saturated as in a saturated atmosphere. CoxDiTioNS OP Infection Chilling was also found to have a very marked effect on the degree of infection secured, as can be seen by referring to plates I to X. Ninetyfive per cent of the seedlings chilled became in- 80 Wisconsin Experiment Station. fected while the controls not chilled usually showed less thati 5 per cent of infection and never more than 15 per cent. This difference in extent of infection, I helieve was due to the in- creased percentage of spore germination. It should be noted, however, that the chilling process may have had some effect on the host, possibly making it more susceptible. This is a point tl'.at needs further investigation. The favorable effect of chilling on the conidia of Cystopus is plainly an adaptation to the environment of the fungus. The spread of a^ fungus by zoospore infection is directly dependent upon the presence of water on the foliage of the host. DeBary found the motile zoospores of Cystopus in the dew drops in the morning on the host plant and I have often made the same ob- servation. The fall in temperature which leads to the deposition of dew and thus provides a medium in which the zoospores may develop serves at the same time as the necessary stimulus to the gerniiiiation of the conidia. The results obtained suggest that a close relation exists be- tween host vigor and susceptibility in that healthy plants are more susceptible than sickly or abnormal ones. No marked difference in the susceptibility of leaves and cotyle- dons of the radish, shepherd's purse, white mustard and gar- den cress Avas observed. So-called Physiological Species Repeated infection experiments were made ' using conidia of Cystopus candidus from the common radish, Raphanus sativus. upon this same and other cruciferous hosts to learn whether there is any difference in susceptibility. A large number of experiments were made testing the suscep- tibility of twentytwo different varieties of radish, and it was fcund that no marked diff'erenee in their susceptibility existed. It was also readily possible to infect Raphanus candatus with the conidia from Raplianus sativus which shows that species of the s;ime genus are susceptible to the form of Cystopus that occurs on the common radish. Species of crucifers from other genera known to be hosts of the white rust were investigated as to their susceptibility to Cystopus candidus from the common radish. Infection was se- cured on the wiiite mustard, Brassica alba, and cabbage, Bras- Experiments on Spore Germination. 81 sica oleracea. At no time was it possible to infect more than 50 per cent of the cotyledons or leaves of white mustard which were inoculated. With the cabbage, it was even more difficult to secure infection, iilthough fifteen varieties were tested. Less than 1 per cent of the plants inoculated became infected. No infection could be secured on any of the other crucifers tested. These included turnip, Brassica rapri, ten varieties; black mustard, B. nigin, rutabaga, B. ccimpcstris, three varie- ties; shepherd's purse, Capsella BnrKa-pa.^toris; garden cress, Lcpidium sfiiivum; wild pepper grass, Lrpidium virgimcum; hedge mustard, two species Sisymhriiim officinale and S. altis- simum; candy tuft, Ihcria umhellata; water cress. Nasturtium officinale, and wall flowe7\ Chciranflnts cJiciri. LITERATURE CITED 1807. Prevost, B. : Memoire sur la Cause immediate de la Carie ou Charbon des Bles, etc., pp. 33-35. 1854. Tulasne, L. R. : 2nd memoire les L^redinees et les Ustil- aginees. Ann. Sci. Nat. Bot. Series IV 1&2:77. 1859. Hoffmann, II.: Ueber Pikkeimungen. Bot. Ztg. 17:210 1860. DeBary, A. : La formation de zoospores. Ann. Sci. Nat. Bet. series IV 13&14:236. 1863. DeBary, A.: Recherches sur le developpement de quel- que^ champignons parasites. \\nn. Sci. Nat. I5ot., series IV. 20:14. 1873. Wiesner, J.: Sitzber Akad. AViss. (Vienna) :\Iath. Phys. Kl. 68, ]. (Abstract from DeBary Morphology of Fungi, p. 349.) 1875. Farlow, W. G. : The Potato Rot. Bui. Bussey Inst. Part IV p. 319. 1876. Farlow, W. G. : The American Grape Vine JMiklew. Bui. Bussey Inst. Art. I. p. 419. 1882. Busgen, IM. : Die Entwicklung der Phycomycetensporan- gien. Jahrb. "Wiss. Bot. [Pringsheim] 13 :22. 1883. Zalewski, A.: Ueber Sporenabschnurung und Sporenab- fallen bei den Pilzen. Flora 66 :251. J883. Zalewski, A.: Zur Kenntniss der Gattung Cystopus. Bot. Centbl. 15:215. 82 Wisconsin Experiment Station. 1886. Scribner, F. L. : The Dowiij^ Mildew. Bot. Div. U. S. Dept. Agri. Bui. 2, p. 10. 1893. Via.la, P. : Les maladies de la Vigne, p. 93. 1895. Eriksson, J. : T^eber die Forderiing dor Pilzensporen- Ivcimung durcli Kalte. Ceutbl. I'.akt. (etc.). 2 Abt. 1 :557. 1896. Eriksson and Ilenning. : Die Getreideroste. Stoekhohn. p. 73. 1901. Liidi, E. : Beitriige zur Kenntniss der Chytridiaeeen. Hedwigia, 40:1-44. 1901. Duggar, B. ]\I. : Phj'siological Studies with Reference to the Germination of Certain Fungous Spores. Bot. Gaz. 31:38-66. 1902. Ward, IT. M. : On Relation between Host and Parasite in the Broines and their Brown Rusts. Ann. Bot. 16:265.' 1902. Ferguson, IMargaret C. : Germination of the Spores of Agarieus campestris and Other Basidiomycetous Fungi. IT. S. Dept. Agri. Bur. Plant Indus. Bui. 16:16. 1903-4. Eberhardt, Albert: Zur Biologie von Cy.stopus candi- dus. Centbl. Bakt., (etc.). Abt. 2. 10 :655-656. 1904. Eberhardt, Albert: Contribution a I'etude de Cystopus eandidns. Centbl. Bkt. (etc.). Abt. 2. 12:614-631 and 714-727. 1904. Clinton. G. P.: Downy I\Iildew, or Blight of Musk Melons and Cucumbers. Conn. (New Haven) 7\gri. Exp. Sta. Rpt. Part 4. Botanist Rpt. p. 338. 1905. Jahn, E. : IMy.xomycetenstudien. Ber. Deutsch. Bot. Gesell. Abt. 2. 23:489. 1906. Reed, Geo. M. : Infection Experiments with Erysiphe graminis. Wis. Acad, of Sci., Arts and Letters. Part 1, 15:135. 1907. Reed, Geo. M. : Infection Experiments with Erysiphe Cichoracearum DC, Bui. of Wis. Univ. No. 250, Sc. series 3, No. 2. 3 :381 . 1904. Smith, R. E. : The Water Relation of Puccinia Asparagi. Bot. Gaz. 38:19. Experiments on Spore Germination. 83 1!)0S. Clinton, (I. P.: Artificial Cultures of Phytophthbra with Especial Kcference to Oospores. Conn. (New II a veil) Agri. Exp. Sta. Rpt. Part 4. Botanist Rpt. p. 904. 1!)0S. Salislniry, H. 1).: Pliysiojrrapliy. p. ^56. 11)10. Jaczcwski, A.: Studicu uhci- das Verhaltcn des Schwarz- rostes des Getrcidcs iti liussland. Zt>ehr. IManzon- krank. 20:21. J!)ll. Cook, ^1.: Protective Enzymes. Science, n. s. 33:G25. 84 Wisconsin Experiment Station. DESCRIPTION OF PLATES The following plates arc all from pliotogra])lis of radish plants grown in six inch i)ots, taken from above. In some cases the entire culture is shown with a slight reduction. In the rest only a portion of the culture is shown, but so selected as to be fairly representative. The plates illustrate the advan- tage of chilling in securing optimum spore germination and favorable conditions for infection with Cijsiopiis ccnulidus. The seedlings were incculated, covered with bell jars and either placed in an ice box or kept at room temperature. I. Radish. Var. — Ne Plus ITltra. Sowed ]\Iay 16. Inoculated May 26. Infected June 2. Photographed June 6. This culture was chilled. II. Control. Radish. Var.— Ne Plus Ultra. Sowed :\Iay 16. Inoculated May 26. Infected June 2. Photographed June 6. This culture was not chilled. III. Radish. Var.— Ne Plus Ultra. Sowed Nov. 1. Inoculat- ed Nov. 8. Infected Nov. 15. Photographed Nov. 17. This culture was chilled. IV. Control. Radish. Var.^Ne Plus Ultra. Sowed Nov. 1. Inoculated Nov. 8. Infected Nov. 15. Photographed Nov. 17. This culture was not chilled. V. Same culture as shown in III. Photogiaphed twelve days later, Nov. 29. A'^I. Same cultui-e as shown in IV. Photogi-aplicd twelve days later, Nov. 29. VTI. Radish. Var.— Ne Plus I^ltra. Sowed Nov. 6. Inoculat- ed Dec. 3. Infected Dec. 12. Photograi)h('d Jan. U 1910. Cul- tures at right of page chilled ; at left of page controls, not chilled. VIII. Radish. Var.— Triumph. Sowed Dec. 12, 1909. Inoculated Jan. 3, '10. Infected Jan. 12, 1910. Photographed Jan. 18, 1910. IT])])('i- ciiltui'c chilled. Lower culture control, not chilled. IX. Radish. Var.— Ne Plus ITltra. Sowed I\Iay 26, 1910. Inoculated June 6. Inf. June 12. Photographed June 15. This culture was chilled. X. Control-Radish. Var.— No Plus I^tra. Sowed May 26, 1910. Inoculated June 6. Inf. June 14- Photographed June 15.. This culture was not chilled, EXPLCRIMKNTS ON Sl'CUK (JKKM I NATION. 85 I.— Radish seedlings inoculated with Cystopus, chiKed. (Compare with II.) EXPERIMEXTS OX SpOHE f J KRMIXATIOX. 86 n.— Eadish seedlings inoculated witli Cystopus, not chilled. (Compare with I.) Experiments on Spore Gkhmixation. 87 ^"^^^^B^ Jf n IP- ^^ ^By^tM /^ JT M^V^ *^[^^j^Bj ^ > T*-^ ^^^^^P^^^VSjr^l^HJKi^i T ^ KX w ^^^ l^V^^^/^'^^^K^^'^'''-^ ' M^'i m r ■Li3lr ^ K w^ l^tofcrJ^ ^^i MK^ l^y &"*^ ^S Mr P^^^^^ttlBA'^ ^HF^'^'^SIIfl i^^ ^m^ III.— Radish seedlings iiioenlnted witli Cystopiis; cliillcd. IV.— Radish see(niiigs inoculated with Cystopus: control not chilled. ExrERIMEXTS OX Sl'ORf: (JkKMIX ATION. 88 v.— Radish seedlings inoculated with Cystopus; chilled. VI. — Radish seedlings inoculated with Cystopus; control not chilled. EXPEKIMKNTS OX Sl'OUK CjiKKMINATION. 89 VII.— Four small cultures of radish st'O.lliiigs; two at right of page, chilled; two, controls, at left of page, not chilled. Experiments on Spore Germination. 90 Larger rmiisli plants inoculated with Cystopus; chilled. VIII.— Large radish plants inoculated with Cystopus; control not chilled. Experiments ox Spohk ( Jkk'mixa'iid.v. 91 1 r^^^^m *f .'lijiy ^Tw^il ■'■■• * » • 1 •. ^» — -■■ -'ti " ■ ■ •^BBk mm iiiMii^M^^^H V ' ^ • A^^^K^SiBB^^^^^^ ^? ^^•^^^^' ^. ^^^ iil"-' IX.— Radish seedlings inoculated with Cystopus; chilled. wK^mi mmm^^^m ^ m. .. * *vfl X.— Radish seedlings inoculated with Cystopus; control not chilled. 'Rj... V.UJU. IV '. H A Sclerotium Disease of Blue Joint and Other Grasses^ A. B. STOUT Tntroductiox During the siiinincr of 1!)<)7 the ^vriter oIksoi-vpcI a fuiifrns vvliit'li was appearing in the iiiarsii meadows about ^Fadison. Wis., as a parasite on the leaves of the grass eonnuoiily known as ])lue joint {Calamagrostis canaih nsis) . The principal symp- toms of this disease a,re as follows. The portions of the leaves att.aeked soon lose their green color and become dry and rigid. Often an entire culm is killed. Wlun a large number of culms within a small area are infected, the general appearance of iho dead and whitened leaves is somewhat similar to that oft-^n produced on young grain ]>l;ints by a frost. A cursory exam- ination, however, showed that there was present on the dying leaves a delicate gray felt of mycelium from which sclerotia often developed. When iiuitui-e these sclerotia project into the air as small but conspicuous bead-like bodies. These symptoms clearly indicated that this fungus was tho cause of the death of the leaves and culms of the grass in ques- tion and the severity of the attack made it a matter of consider- able economic importance inasmuch as blue joint is the most valuable of the wild hay grasses of Wisconsin. Davis (1893, p. 183) has briefly described a fungus which produces sclerotia on Calamagrosfi.t canadensis and which he found at various points in AVisconsin. He did not identify 1 The investigations here reported were carried on in part under the guidance of Prof. R. A. Harper of the department of Botany of the Uni- versity of Wisconsin. When Prof. L. R. .Tones assumed the chair of Plant Pathology at Wisconsin in February 1910, the %^ork was contin- ued as a special patholcgical problem under his immediate direction. From each of these the writer has received helpful criticism and sug- gestions. 208 Wisconsin Research Bulletin No. 18 the fungus ])ut stated that the sclerotia resemble those which he had found on Silphiuni, Helianthus, etc. It was found at once that the fungus agrees with European specimens of Sclerotium rliizodes Auersw. growing on Phalaris arundinacea, and on Calmnagrostis arundinacca. (Sydow, Myco- theca Germanica Nos. 298 and 299.) Later an examination of the exsiccati in the Ellis collection now in the herbarium of the New York Botanical Garden showed that the fungus Sderoiium rhi- zodes has been distributed as follows : De Thiimen, Fungi Austriaei, No. 1096, on Poa pratensis, 1873. De Thiimen, Mycotheca Universalis, No. 199 on Poa pratensis, 1873. Eriksson, Fungi Parasitiei Scandinaviei, No. 82, on I'oa praicKS'is ]882. Rabenhorst-Winter, Fungi Europaei, No. 3,199, on Phala- ris arundinacea, 1884. Sydow, Mycotheca ^Marchica, No. 1339, on Poa leaves, 1887; No. 2698, on Calamagrosiis neglccta, 1889; No. 2996, on Digraphis arundinacea, 1890; No. 3297, on Poa irivialis, 1891; No. 3298, on Anlhoxanthum odoratum, 1891. Krieger, Fungi Saxonici, No. 550, on Phalaris arundina- cea, 1889; No. 600, on Agroslis, 1890; .No. 1397, on Brachy- podium silvaiicum, 1896; No. 1398, on Calamagrostis HaUcriana, 1898; No. 1399, on (a) Holcus lanalus, 1891; on (b) nolens mollis, 1897. Allescher and Sehuabl, Fungi Bavarici, No. 200, on Phalaris arundinacea, 1891. The exa.minations of thes^e specimens established beyond question the identity of the Wisconsin fungus on Calamagrostis canadensis with the European species, and the writer has sup- plied specimens so named for distribution in Fungi Columbiani. The first publication and description of this fungus was un- der No. 3232 in Klotzsch's Herbarium Yivum INIycologicum, a review of which appeared in the Botanische Zeitung Vol. 7, p. 294, 1849. The description reads as follows: "1232 Sclero- iimn {Sarcidiuni) rhizodes Awd. Mspl. Suhglobosum, ])rimum allxtvillosuni. mox ola1)rinsculum, nigreseens rugulosum, fibrillis albis seriatim insideus. Auf liliittern von ralamaqroslis Epig- pios schon vor deren Entwicklung. " A SCLEROTIUM DlSEz\SE OP COMMON GRASSES 201) Fi-iink (188], p. 545, and 1896, vol. 2, p. 511; ciuite ade- (luately describes tJie syniptorns of this sclerotial disease of grass leaves. He records that the disease appeared as an epi- demic in 1879 in the vicinity of Leipzig on Pkalaris arundinacea and Daclylis glomcrata and that a large part of one meadow appeared dry and white as a result of the attack of the fungus. He states that no conidia from the mycelium or fruiting bodies from the selerotia had been observed. Sorauer (1886, p. 300) quotes briefly from Frank's descrip- tion of 1881 but adds no new data. Saccardo (1899, p. 1154) lists Sclerotium rhizodcs with the fungi having sterile mycelia. He repeats the brief descrip- tion quoted above. Otherwise he makes no mention of the oc- currence of the fungus as given in the various exsiccati listed above. Tubeuf (1897, p. 266) records this fungus with selerotia of unknown affinity and mentions, only, that it occurs on living plants of Pkalaris arundinacea, and Calamagrostis; also on dead leaves of Dactylis glomerata. This summary of the literature pertaining to Scleroiium rhizodi/cs indicates clearly the meagre and incomplete knowl- edge concerning its life history and its relations to its host plants. In the region about ^Madison, Wis., the fungus is most abund- ant on Calamagrostis canadensis, although as explained later it occurs on various other species of grass. ]\Iy studies have been made chiefly with material from this one host and all the statements which follow are to be so understood unless otherwise clearly specified. Detailed Description of Syjiptoms "When the leaves of infected plants start to unfold, their tips remain more or less convolutely rolled as in the bud and soon become white, dead, dry, and rigid. The whitened tips, over badly infected areas, are conspicuous in mass effect and at first sight, as already stated, resemble frost injury. Exam- ination, however, reveals the presence of a thin but dense felt of mycelium which is most marked on the inner surface of the infected leaves and along the inner edge of the leaf, especially when only a part of the leaf is rolled laterally. (See Figures 210 AViSCOXSIX rjESEARCir lilLLETIX Xo. 18 1 ;iii(l ;i. ])()in1s marked )ii.) In the outei'inost leaves, the tip is usually the only portion affected and often only a lateral half of the blade is invadid. As the portion of the leaf at- tacked dies and fails to unroll, the tip of the leaf next in order is often caught and firmly held in tin- roll and it in turn holds in the same manner the leaf next in order. .Meanwhile the growtii of the hasal i)ortion of t.hc leaves together with the elongation of the internodes tends to separate the lowei' halves of the h-avcs whose tips are thus held together and to l)roduce j/iM-uliar and characteristic crooks as shown in Figures 1, 2 and '.'). AVIk n the fungus develoi)S vigorously, the inner leaves are completely ])ejietrated hy the mycelium which also extends into the culm below the g.r'owing point. In this ease the death of the tei'iiiina] bud icsults. During a season of vig- orous development the majority of the infected culms never gi'ow to he nioi'c than twelve inches high, while large numbers are less than six inches high. JM]tii'e groups (if culms arising fi'om a i-hi/onie ai'e often thin. pun\'. and dwarfed. h'rom these conditions it is cvich n1 that the mycelium first becomes vii'ulent within the bud whence if spreads through leaf after leaf and that the indi\i(liud leaves are thus infected befoie they uiu-nll fi'oiii Ibe ])U(1. The mN'celium l)ecomes most eons])icu()Us on a haf in tlie lowi st part of its region of growtii. The unaffected basal pai't of the blade becomes liattened out and .)ust h(>low the i>;iint where the next leaf in or(h'r emerges fi-<»m the roll an area is usually covered with a felt df white mycelium. Jn case of partial lateral infection a narrow zone with tufts and knots of mycelium appears along the nuirgin of the fold or roll where the dead leaf tissue meets the gi-een tissue. Soon rouinhd bead-like sclerolia are jn'oduced along the infected ])or- tions of the leaf or fi'om the I'elt of mycelium at the l)as(\ (See I^Mgure 1 points niai-ked s). AVIieu nuitui-e. they vai'y in si/e from 1 1(1 .") mm. iu diameter (See Figure li). The sclei'otia are I'Mi'med on the h.ives and are always suiierficial. They are seldom foi'med within the roll and nevei- within the tissues. The ])i'oduction (if the ecnspicuous ci-oeks and the develo]i- iiieiit of the sclei'otia as described, aie features which make certain the identification of tbis I'ungus as it a|)peai's on Cahi- magroslis canadensis. The wiitei- has I'ound this fungus ap- pearing in the region about ^fadison on tiie following addi- A ScLKHD'I'ir.M DiSKASK OF ("oM.MOX GkASSES 211 ti()ii;il yTiisst's : /'liiihirls (ini mil unci ii L. ; (UtlanuKjrosiis iC(jlv((a (I"]liili.) (ijicrtii; /'(xi proh iisis L. ; /'ani(i(laria ncrvnla (Willd.) Kuntzc; I'hhinn prafcnsc L. ; Jfordiiim juhatum L. ; Bronius viliains L. ; Kalonia P( )nisijlfa)iica (DC.) A. Gray; Af/ropi/roH cnninion (L.) K. & S. ; A(/nis-fi<; hDfmaJi. S. IV All of these <\\e('J>t the lifst three ii; ■(! ;i I'e lielO re- l>oi-t('(l as hosts for the first time. 'J'he oecurronee of the fnn- j?iis ui)on sovei-ai lOuropean grass liosts not hero mentioned lias aliiady liet n noted. In general the appearance of the fungus on all of these liosts is similar to that described I'cr Calamagros- (is ca)icid( iisis. In each case the sclerotia produced are identical and infected ])lants of each species have been lomid growing by the side of infected i)lants of ('alamagrostis canadensis. It should be noted that in the case of perennial grasses a bud wiiich grows upward into the air produces leaves only, for one or more years and then teritiiuates its life as an individual culm by producing flowers and seed. There is considerable difference in the habit of growth of these leaf and frniting culms in the ditfei'ent species that serve as host plants. In Pon l)ral(>isis and Panicidaria turvain the vegetative culms, as I shall designate the culms liearing leaves only, are short and the leaves which they beai' arise I'ather close to the gi-ound. Here the infected leaves are not lifted up to the general level of the vegetation and although there nuiy be many infected leaf culms, the general effect is not so conspicuous as it is in the case of Cnhnnagrostis canadensis. The leaves of these two species are normally conduplicate in the bud and the infected leaves re- main thus folded. Sclerotia are ])roduce(l between the folds on the u])per sui-face and aie always superficial. Along the groove of the fold, a thin felt of mycelia develops. On Poa prafen- sis especiall.w the distribution of the fungus is rather irregular and not continuous in a single leaf. During the seasons of 1907. 1908 and 190!). general obseiwa- tions were made in the marsh meadows about ^hulison as to the abundance and the course of development of the fungus. In 1910 and 1911 however, a single marsh meadow conveniently situated for observation was selected for a more intensive stiuly of these problems. This meadow is almost circular in shape with a diameter of nearly half a mile. Although its elevation is but a few feet above the level of Lake ^Monona, which is near by and into which it is drained, it is usually sufficiently dry to 212 Wisconsin Researcpi Bulletin No. 18 l)e f-nt over fo,r hay. Tlie statistical analysis which the writer made of the plant population of this marsh meadow showed that Calamagroslis canadensis was quite generally dis- trihuted over the marsh and that it constituted 18 per cent of tlio entire plant population. The counts were made rather late in a season in wiiich the fungus was not especially vigorous ex- cept in certain areas througli which the transect studied did not pass. The fungus was noticed at nea,rly all points, but no separate count was made of the infected culms. Seasonal Development in the Field The development of. the fungus was further studied in the field in the spring of 1910. It was an early spring. On March 24, the first grass buds were beginning to unfold and many of these showed the typical effects of the fungus. Sclerotia which had been formed during the previous year w'ere found in numbers on the ground but none of them showed any signs of producing fructifica- tions. On April 15, the grass stood from four to six inches high. In certain areas where infection was most general in previous years, the fungus was especially abundant. In some areas of several square rods extent, it appeared that 75 per cent of the culms then unfolding their buds w-ere infected. On April 30, the grass averaged one foot high. The dead tips gave to the regions of worst infection a conspicuous whit- ened appearance as if the tips had been burned or frozen. Many of the infected culms were totally dead. Young sclero- tia were forming in considerable number. Throughout May and June the,re was no apparent increase in the number of culms showing the infection. The majority of culms already diseased had died to the ground. Others con- tinued to grow and put forth new^ leaves which in turn showed the presence of the Fungus in varying degrees of vigor. The season was one of unusual. dr\'ness and, as a result, few sclero- tia were produced. By July 14 there were few areas where llic fungus appeared abundantly on the growing culms. In areas of previous slight infection, healthy culms now stood about two feet tall overtopping the dead culms killed earlier in the season. Still duiiiig the remainder of the season, culms bearing infected leaves could always be found in considerable number. The grass itself made little growth after the middle of July and during the first week in August it was cut for hay. A ScLEROTii-.M Disease op Common Grasses 2V.i Oil Sfpteiiilier 2, the new gi-o\\ili was iicaily one foot high and in this, infected euliiis were searec and the eharacteristic crooks were not well developed. Observations in 1911 The spring ol" IDll was nearly three weeks later than that of 1910. The fungus appeared as in pre- vious years on the fi.rst leaves that opened and the subsequent development was as described above. There was a fair amount of rainfall through April and May and by iNIay 27, it was evi- dent from the number of affected culms that the infection was more vigorous and more general than in previous seasons. Scler- otia were abundant and many were mature on ]\Iay 27. Throughout June the dead and whitened tips of infected grass blades were abundant and conspicuous especially on culms which continued to develop new leaves. There was, however, no increase as the season advanced in the number of infected culms. Areas which were free from the fungus earlier in the season remained free from it. These observations made it clear t.hat there is no spread of the fungus from culm to culm aerially and that a culm which does not show the fungus on its first leaves does not harbor it later in the season. The effect of the fungus is especially marked on Calamagros- tis canadensis. IMany of the culms die to the ground early in the season. The remainder continue to put forth infected leaves, but are decidedly stunted. Very few infected culms produce flowers. Even apparently uninfected culms which arise from the same rhizomes with infected culms, are weak and stunted. Observations made year after year show a decrease in the num- ber of plants of C. canadensis in the infected areas. Throughout the greater part of the marshes C. canadensis is associated with Carcx stricta Avitli which it is more or less in competition. The destructive effects of the fungus seem to be an important factor in favor of Carex striata. Observations made each year indicate that the fungus devel- ops during the spring and early sunnner. It is most vigorous and conspicuous during the period of the most rapid growth of the host plant. It becomes less noticeable as the season ad- vances due to the death of many culms and to the overtopping by unaffected culms and by associated vegetation. The fungus seems unable to make headway on fullj' developed leaves during June and July. Each leaf as it unfolds contains the fungus, but 214 Wisconsin Keseahcii J^illetin No. 18 tke exj)osnro to dry Miid licatcd ;ni- cliccks tlu> di'vclopniont of inyeeliiiiii and selerotia. Abundance and Distribution To ascertain the amount of infection the following nietliod was used upon 1lie marsh selected for critical study. At inter- vals of tlfteen ])aces along a line leading through the marsh a heavy wire hoop tAvelve inches in diameter was dropped down at random. All the vegetation inside the hoop was then cut close to the ground, the infected and the uninfected cidms of C. canadensis were sorted out and counted and the results tabulated. The greatest injury appeared to l)e at the borders of the marsh meadow, especially on an area of about sixteen acres, which is somewhat isolated from the greater part of the marsh by a I'ailroad embankment and a canal which passes through Ihe marsh. This area is well drained and is ahvays dry enough to be mowed by machine. Here the dominant species Tabi-e I Distribution op Fungus on Culumugrostis canadenxin Station Number of infected culms Ilealtliy culms 1 18 19 21 17 49 22 87 14 45 36 21 58 18 0 0 1 10 426 6 2 4 3 6 4 5 8 I) 26 7 31 8. 5 <) 8 10 7 11... 32 12 29 13 6 14 9 15... . ; 7 16 18 17 6 Total 215 are Carer sliicia, CahtDUKiroslis (■■anadensis, Poa 'prafcKsis, (ihjrrria n(rvala and Sparliiia cjinosin-oides. These are consid- erably intermingh'd. I'dh {.ruhnsis is especiaJly abundant along the edges neai- Ihe uphmd. ('. (uniadi iisis is (piite uni- formly dislribuled over tile area except at the extreme bor- der, l^eginning at the margin, data on the distribution of the fungus were taken as described above at seventeen consecutive points fifteen paces apai'1. The i-esults for C. canadensis are given in Table 1. A SciiEKoTUJM Disease oi-' Common Grasses 21.1 From tliose data it appears tliat 66 per eent of the culms oC ('. conadensis ^rowinj? on this area were infected. While the I'lin^Mis was (|iiit(' general in its distribution on this area there were patches of snudl extent, usually from one to two rods in diameter, that were almost entirely free from the fun^is. Often these would he (>ntii-eiy sui-rouiidcd hy infected str'i|)S and in such cases thi' contrast was always mai-ked. Data takt'ii from typical areas show that on ^lay 27 the healthy plants then stood on the average, 30 inches high, while the tallest of the infected ones nearby were l)nt 12 inches, and the majority of them were 4 to 8 inches high. Even the culms that had escaped infection were smaller and poorly developed. These "islands" of unin- fected plants in the otherwise destructively infected regions gave by comparison conspicuous evidence of the damage done by the disease. Data on the abundance of the fungus over the greater i)art of the marshes given in Table 11 were o])tained in the same Table II ADomoxAL Data on Distribution ok Fungus on Calamagrostis canadensis Station Infected fulms Healthy culms Station Infected culms Healthy culms 1 0 0 19 13 24 2 0 9 20 5 12 3 6 3 21 0 6 4 2 o 22 0 13 5 i 7 23 0 0 6 2 18 24 0 0 7 0 0 25 0 4 8 14 7 26 3 12 9 6 10 27 6 19 10 1 2 28 5 28 11 8 22 29 0 0 12 1 37 30 0 0 13 14 13 31 1 12 14 17 tj 32 1 10 15 7 9 33 0 19 16 8 27 34 0 8 17 27 21 35 6 24 18 8 45 36 19 22 37 4 8 Totals for the 37 stations 195 467 manner. The transect began at the south edge and extended through the center of the marsh as far as the canal and thus coincided with the ti-ansect previously studi(>d in determining the ])Iaiit population of this meadow. For tliis part of the marsh tlie infection averaged 29 per cent, ("ombining all the data obt^iined gives an average infec- tion of 47 per eent for the two transects which represents fairly 216 Wisconsin KESEARCii liri.i.KTiN No. IS' tlu; coiiditions on tliis particular meadow on IMa.y 27, 1911. One transect passed through a well drained marginal area. The other passed from the margin through the wettest part of the marsh meadow. The two give a fair average of the whole meadow, ' ' In the central wet portions of the meadow Carcx aquatilis, Carcx Sariivcllii and Carcx filifot'mis were dominant and C. canadensis was either absent or sparse. It is noticeable that the percentage of infected culms was relatively lower under such conditions. (See data for stations 20-34 in Table II.) The sparseness of the grass evidently gives less opportunity for the infection to spread through the soil. Throughout the entire marsh there were areas usually of small extent that were free from infection. There were also areas with more than 90 per cent of infection. The latter were uni- formly located in areas whe,re C canadensis was dominant. Range of the Disease in Wisconsin During June 1911, the writer spent three days in making observations on the occur- rence of this fungus in the extensive marshes in the townships of Albion and Christiana, Dane county. Wis. Here several square miles of marsh meadow were traversed extending for nearly eight miles along Saunders' Creek. These meadows have for a number of years either been cut for hay or utilized as pasture. C. canadensis and Carcx sfricUi were dominant over most of the area. The former grew in luxuriance over large areas and stood when in blossom 4 to 5 feet high. Carcx aqua- tilis and Carex riparia were abundant in the wetter i)arts of the area and Poa pratensis, PanicuJaria nervata and Poa flava were common near the borders. In several areas Phleum pratense was abundant. Over the entire area visited, the fungus was found to be con- spicuously abundant on C. canadensis, although areas of several acres were found which were nearly free from the fungus. On others the fungus although abundant was much scattered and its effects not conspicuous. Over larger areas, however, the same degree of destruction was seen as has been described for llie vicinity of Madison. For the season of 1911 1lio infection on this entire area was not less than 10 per cent of tlic culms of C. canadrnsis. Besides this there was a less conspicuous and less general injury to various ofher grasses by the same fun- gus. A ScLEROTiuii Disease op Common Grasses 217 In this region several small isolated patches of C. canadensis, located on high land, were found to be infected. In fact the most uniform and complete destruction seen anywhere was on such an area. A nearly pure formation of this species measuring 2i{.x2 rods was found by the roadside bordering a cultivated fu^ld and on high dry land with the nearest marsh three-fourtlis of a mile distant. The undisturbed dead culms of previous years formed a rich mulch. Except for a fringe at the ends of this formation, practically every" culm was infected. Many were entirely dead and tlie short culms with the uniformly white tips appeared in sharp contrast to the surrounding green vegetation. The infected belt extended to the border of an oat field, but no trace of the fungus was found on the oat plants. The occurrence of the fungus in such isolated areas of Cal- amagrostis suggests either that the fungus is a widespread soil or root fungus which does not always show parasitic develop- ment in the leaves, or that it is distributed by spores which cause a rapid infection of roots, stems, and leaves. As noted above I have so far found no spore stage. Marshes were also visited in the region about Fort Atkinson and Lake Koshkonong in Jefferson county, "Wis. In several small isolated marehes, no trace of the fungus was found, but in long stretches of marsh land along Bark River and Rock River, the fungus was abundant. Here again it was often found in roadside patches of C. canadensis on rather high land. The most general occurrence of the fungus found anywhere was near the mouth of Koshkonong Creek. Here a continuous marsh meadow of eighty acres was examined, June 23. A heavy growth chiefly of red top and Carcx stricta covered the higher portions while PJialaris arundinacca stood five feet tall over the lower parts. C. canadensis was abundant over areas in which nearly every culm was infected with the fungus so that the en- tire Calamagrostis population Avas overtopped by Carcx striclu, Agrostis alha or other plants usually of lower stature. The in- fection was so complete that not a single flowering culm of C. canadensis was observed on the entire eighty acres. Phalaris arundinacea was, however, but slightly affected with the fun- gus. From various reports it seems that these conditions prevail throughout the greater part of the state. Dr. J. J. Davis states 218 Wisconsin Research Bulletin No. IS in a letter to the writer: ''I first noticed Schrotium rhizodes on C. canadensis in ]S!)2 and I have seen it on that host every year that I have been in tlie fickl since. I have collected it at hoth ends of the longest axis of the state and at various intermediate ])oints so that I think that it may he said to be generally distri- l)uted through Wisconsin and to ])e a constant and general mem- lici- of the ])ai'asitic fungus tlora of the state." In response to inquiries, (ieorge L. Peltier of the State Cran- berry Experiment Station at Grand Rapids, Wis., writes as follows: "I have made oliservations on Sclcroiium rliizodcs, but have been able to find it only on blue joint {(_'. canadensis.) It is very wich^spread here and I have found it wherever I have looked for it. In a field just west of the station about 30 per cent of the stalks seemed affected. On one of my ti'ips I found a. whole field of many acres where almost every })lant was af- fected. It liad weakened the grass so much that none was able to head out." It appears, however, that this fungus has not been reported in America outside of Wisconsin. This is most singular. Here in Wisconsin it is widespread, abundant and conspicuously parasitic. Its chief host, (\ canadoisis ranges from New- foundland to Alaska south to Xnrlli Carolina, New ^Mexico and Calit'niiiia and other glasses which may serve as hosts in- crease the area in which the fungus may api)ear. It seems ])rob- able that this fungus is ecjually connnon in otlier sections besides AVisconsin, but lias been overlooked. Extent of Infection on Grasses Otiikk Than Calama^jrostis Canadensis The fi'r(|ncncy of llic fungus on oilier grasses bears directly on the (|U('s1i()ii as 1o the mdliod of infection. I'lutUiris (ini .1 (lliHi(( (I Seal tcriug gi'ou|)s of plants were found with the fungus in llic iiiaish meadows along Albion Creek, iioek Kiver uiar V\. Atkinson, and Lake Koslikonong. These were al\va,vs in the iiuiiiediatc \ieiiiit\ of badly infected areas of C. canadi nsis and were usuall\ at the l)order of a Pha- laris formation. On the whole, this species was slightly infect- ed in this region. Tliis grass did not occiii' in the marsh mea- (hiw s st udied at ^ladison. A ScLKUoTii .M 1)isi:ask ok Common Okasses 219 daUiiiuKjrusiis najlivia TIn' I'liiigus \v;is al)iiii(l;iiil. and in- jurious over areas covered witli this species. Jnfcctcd (L cana- densis was always found in llic vicinity and usually the two species were intermingled. Panicularia ncrvala Wherever this species grew inter- mingled wilii infected ('. canadensis a small per cent, of its leaf culms showed infection. On this host however, the fungus did not appear to be sei'iously destrnctive. I'oa prah iisis In the case of this sj)ecies there was rather abundant, antl serious injury especially where infection of C. ca.j (i(l( iisis was vigorous in the innnediate vicinity, but the fun- gus also apiK'ared (juite abundantl.v in liie nearly ])ure forma- tions of I'oa prai< )tsis, which thrive in the border and upland portions of marshes about Madison. Phlenm pmtcnse Vigorous infection of this grass was seen in the border of a nmrsh near Albion. Only a few infected leaves and culms were found and in nearly every case these grew by the side of infected culms of ('. canadensis. Hordcum juhatum At ^Madison this grass was found grow- ing in border areas of the marsh often with its roots inter- mingled with those of infected culms of C. canadensis. It was only under these conditions that culms were observed showing the characteristic effects of the fungus. Bromus ciliafus, Eatonia pennsylvanica, Agrostis hyemalis, Agropijron caHittinn. A few culms of each of tiiese species were found infected with the fungus. Infected plants of C. cana- densis were alwa^'s close at hand. These observations suggest that while C. canadensis serves as the principal host for Sclerotium rliizodes the fungus may spread to various other grasses especially when they are in close prox- imity, a fact which is fully explained when the soil and root relationships of the host are considered. Several species of grasses, especially Agrostis alba, Andropogon furcatus and Spflrtina cijnosuroides, appear to be immune, or at least no evid- ences of the fungus were found on plants of these species which grew within zones of infected C. canadensis. It is of interest to note that on the stems of Vrfica gracilis, which was growing wdthin an area of vigorous development of the fungus on C. canadensis, there were found in one season sclerotia somewhat similar to those on C. canadensis. 220 Wisconsin Research Bulletin No. 18 Conditions Favorable for Development It has already been shown that the fungus usually developed most abundantly in the field during the earl}^ part of the season, but that during moist summers it was also vigorous later in the season. To determine the conditions which favor the develop- ment of the aerial mycelium and the production of sclerotia the following studies were made during April and May 1910. Culms in the early stages of development Avere gathered in the field, immediately placed in sterile test tubes about 4x20 cm. in size and plugged about the stem with cotton. These were taken to the greenhouse and so placed that the cut end of the culm and the mouth of the test tube were in water thus forming a damp chamber of the test tube. Clumps of the plants with infected culms were also transplanted into pots and kept under bell jars. For comparison others were exposed to the air of the greenhouse. In all the latter the fungus developed slowly with- out any conspicuous show of mycelium and the sclerotia began to form in about ten days. Here the development and appear- ance closely resembled that observed in the field. In the case of the infected culms placed in sterile test tubes there appeared within twenty-four hours an abundant mycelial growth which extended £rom the infected portions of the leaves out into the tube forming a cottony mass which often filled the tube for one-half or two-thirds its length. Soon numerous scle- rotia began to form. Many of these were out in the aerial my- celium and were not directly attached to the leaves. At the end of fifteen days the culms and unaffected portions of the leaves were still green, the mycelium was still vigorous and many of the sclerotia were fully mature. There was a less vigorous development of aerial mycelium on potted plants inclosed in a bell jar. Many infected culms died in ten to twenty-four days while noninfected culms remained green. Sclerotia on these plants were mature in about twenty days. These experiments show that increased humidity favors the development of the mycelium on the surface of the leaves and promotes rapid formation of the sclerotia, which agrees with the facts observed in the field as previously discussed. A S('I;I:ko'i'ii M |)isi:-\si; OK Common (iK\ssi;s 221 Source of Infection During the past throe years hundreds of infected culms have been examined in all stages of the di>ease and throughout the entire j)eriod of its ai>i>eai'an('e, l)ut no spores were found. Evid- ently the mycelium is prevailingly stei-ih; as it occurs in nature. The history of many other sclerotia-forming fungi suggests that the sclerotia may dcveloi) ascoca,rps with ascospores. Dil- igent search for germinating sclerotia has been made during each of the past three years. Each year as soon as the snow melted sclerotia were fountl on the ground in the areas of worst infection and they were found and examined in situ throughout the season. No evidence of germination was found. Both sclerotia gathered from the grass in the field and those grown in cultures have been treated in a variety of ways in the endeavor to induce germination, but with no success at present writing. During the season of 1910 almost no sclerotia were matured in the field, l)ut in 1911 they were produced in abund- ance and 1 have at least 500 now planted under a variety of conditions. In th-e case of a few sclerotia gathered from Urtica, germination was secured. 1)ut a discussion of these is reserved until the identity of the fungus is more certain. So far as I can find the sclerotia do not germinate freely and abundantly each year. T1r\v are, in fact, not matured in abundance each year, for by far the greater number dry up and liecome shriveled, while imnuiture. Yet the fungus is abundant j'ear after year. It was noted that the infection re- appeared year after year in the same areas and that patches near by were constantly free from infection. Examination also showed that in many cases the majority if not all of the culms arising from the same ,root stalk were infected. The primary de- velopment in every case seemed to be from within the bud for here the fungus appears when the first leaves unfold. All this evidence suggested that the mycelium may be perennial. To test this the following experiment was made. IMarch 5, 1910, while the snow stood nearly two feet deep on the marsh, rhizomes of the Calamagrostis were dug from an a,rea where the fungus had been abundant during the previous season, taken to the greenhouse and potted in muck soil that had been in use in the greenhouse for some fifteen years. In the potting, the old culms and dead leaves were removed so that 222 Wisconsin Research Bulletin No. 18 the rhizomes and buds only were planted. The soil was kept well watered. i\Iarch 14 the yonng culms were from two to four inches h\\\ and in one of the opening buds the mycelium of the fungus was visible. ^March 25 this culm showed three leaves infected with the fungus and producing the typical crooks. At this date the opening leaves of th,ree other culms showed the presence of the fungus, two culms were dead from the effects of the fungus, and eight were apparently free from infection. The culms were 12 to 15 inches tall on April 1, with as many as five leaves. The infected culms showed the typical development seen in the field". These results are quite conclu- sive that the mycelium is present in the buds during the winter. Relation op the Host and the Fungus Method of Investigation Following the above experiment the infected areas were visited, and leaves, buds, portions of stems and rhizomes of infected plants were fixed in chrom-acetic and picro-formal fixing solutions. This material was imbedded, sectioned and stained with either iron-haematoxylin or Avith the Fleming triple stain. The sections showed that the mycelium is coexistent in and on the leaves, buds, stems, rhizomes, and roots of the same plants. The characteristics of the fungus in these different parts are such that a detailed description of each is necessary. Character of the Aerial Mijcditnn On the matured foliage leaves, the mycelium is in i)art aerial as described above. The mycelium is white in mass. It is abundantly branched, is sep- tate, and the hyphae anastomose to some extent. The ends of the hyphae are often enlarged. The walls are thin and the cyto- plasm is slightly granrular and much vacuolated. Figure 8 B shows the general appearance of the aerial mycelium taken from the surface of the leaf. For a more careful stud^y of the cell structure, the mats of mycelium produced in cultui'es on cooked potato were fixed in 2 This experiment has been repeated with the following' data: No- vember 24, 1911, Mr. A. G. Johnson (chopped from the frozen ground at Madison i-hizomes of Calamagrostis canadensis and Poa pratensis. These were sent to me at the New York Botanical Garden and im- mediately placed in pots which were kept In a greenhouse. As soon as the first leaves of the growing culms unfolded (.January 8, 1912) the fungus was found with typical development in the buds of several culms of both species. A SCLEHOTIL'Al Dj.SHASE OF CoMMOX GuASSES 223 Flciniug's weak .solution and stained ])y either the triple nietliod o,r by iron haematoxylin. This treatment showed that some of the cells of the myeelium were two-nucleated with a reticulated protoplasm as shown in Figure 7, J. I have not found division tigures in any cells of the fungus and I can give no data on the constancy of the two-nucleated condition or its possible origin in conjugate division. A few cells, however, had more than two nuclei and occasion- ally only one nucleus was pi'eseut. The general character of the mycelium ditt'eivd on the various culture media. This will be described later. The Mycelium in (he Leaves Figure 6 A shows a portion of an infected leaf in the condition shown in Figure 3 A the cross section being taken at a point indicated by m. The vascular bundles were all that was left of the leaf tissues in the infected portion at this stage and the fungal filaments ramified through all parts of the bundles except the phloem. All of the cells of the mesophyl had been totally destroyed and of the epidermal cells the outer walls alone remained. A cross section through an inner leaf of the ,roll showed almost complete destruction of the vascular bundle elements as is shown in Figure 6 B. In all of the tightly rolled and shriveled leaf tips (See Figure 3) the dead and dried remnants of vascular bundles are closely bound together by the mycelium which is itself dead, at this point. During the early part of the season the vigorous destruction of the host tissues proceeds until the plant often appears as in Figure 2, when the entire roll of leaves is completely permeated by mycelium and the destruction is so complete that almost no tissues are recognized within the roll. Action of the Fungus on Leaf Cells The action on the in- dividual host cells is apparently rapid. Careful study of many sections perfectly fixed, sectioned and stained, fails to show the presence of hyphae within turgid mesophyl cells. There is some evidence that cells may collapse somewhat in advance of the ac- tual penetration of the hyphae. Here the plasmolysis of cell con- tents preceding actual penetration by hyphae is not marked and may be due to other factors than the direct influence of the fun- gus. At any rate plasmolysis and disintegration of the meso- phyl cells of the leaves occurs so rapidly that the successive stages in the process cannot readily be observed. 224 Wisconsin 1\esearch Bulletin No. 18 Although the mycelium develops abundantly on the leaves and thus extends beyond the point of internal infection, there are few cases of penetration from without into an expanded leaf either through stomata or through the epidermis. It is notice- able that the mycelium spreads in the leaves most readily from the tip toward the base in the direction in which the vascular bundles run. This is because the mycelium advances most rapidly in the mesophyl tissues which are arranged in strips sepa,rated from each other by the fibro-vascular bundles which extend from epidermis to epidermis and across which the myce- lium passes less readily. This is clearly shown in the cross sec- tion of leaves exhibiting a strong lateral infection. Such sec- tions reveal a clearly defined boundary of the invasion which presents for study various stages in the destruction of cells. In this zone of advance through the mesophyl the mycelium is chiefly intercellular. The ends and sides of the hyphae come in contact with the thin cell walls. In the first stages of penetra- tion there appears to be a slight thickening of the host cell wall and a region about the point of dissolution often stains strong- ly. There is no tendency for the fungus to dissolve out the middle lamella. The hyphae simply pass through openings dissolved in the cell walls. Once inside, their work of destruc- tion is rapid. The protoplast is plasmolized and the thin cell wall collapses. Fii'st the cell contents are digested and later the cell wall is also completely digested so that the entire mesophyl tissue soon disappears. In the outer epidermal layer the my- celium often travels from cell to cell destroying everything but the layer of cuticle. In the region of fungal advance the mycelium is less destruc- tive to the fibro-vascular bundles. Filaments soon ramify freely throughout the woody elements and there is a slight disorganiza- tion of the thick woody cell walls. In portions of the leaves that have been infected longer and in the case of inner leaves more vigorously attacked, there is almost complete destruction of the vascular elements as is shown in Figure fi B. Here the woody walls are relativel}^ thin and collapsed, and the phloem distorted. Distribution of the Mycelium in Aerial Buds Longitudinal sections were made of terminal buds exhibiting varying degrees of fungal destruction. In the early stages the mycelium ex- tends over the surface of the rather tightly rolled leaves with A ScLEROTiuM Disease of Common Grasses 225 sli','lit peuetration in the youngest leaves and more or less pen- etration in the outer. Jt iiuiy even iorin a cushion of mycelium over and around tlie embryonic tissue of the growing point and yet not penetrate this tissue. Figure 6 D shows such a condition. Figure 6 E is a somewhat diagrammatic sketch of the same bud from which Figure 6 D was drawn. It shows that the myce- lium is both upon and within the young leaves before they un- roll from the bud. As young leaves develop from the meriste- matic apex they grow into this cushion of mycelium and are in turn coated by mycelium and thus exposed to infection. As noted, the mycelium does not penetrate the embryonic tissue. In culms which are vigorously attacked the mycelium penetrates into the tissues of the stem below the growing point, dissolving them and, if the process is not checked the death of the terminal bud may result. The fungus seems first to become destructively parasitic in the mesophyl of the outer leaves as they are maturing and expand- ing, to which it gains entrance while the leaves are in the bud. Then the destruction extends to leaves within the bud roir and finally to the apical internodes. In many cases, however, the parasitic attack is confined to the tips of the outer leaves. As they develop each shows the dead tip and the felt of mycelium. All degrees in the rapidity and the amount of destruction may be observed. The Mycelium in the Stems Plants were selected which showed abundant fungus infection of the leaves and pieces were cut from the culms at various points. These were sectioned and examined for the presence of the fungus in the successive nodes, internodes, and buds. To the eye the main portion of the stem, with the sheathes of the leaves and the buds enclosed by them showed no felt of mycelium such as has been described for the leaves. The stained sections however, revealed the presence of the fungus in greater or less abundance in all of these parts. Only a few strands of hyphae were found in the tissues of the hardened internodes. In the nodal regions, the strands were quite numerous in the peripheral tissues where they often ap- peared as knotted tangles of intracellular mycelium. Occasion- ally there was penetration to the intemodal cavity. Here there was no evidence of destiiiction of cell walls other than at the points of penetration. In the old cortical cells the cytoplasm is 226 Wisconsin I^esearch Bulletin No. 18 reduced to a lliiii film wiiicli is difficiilt to locate in any of the cells. As a rule few buds are produced on the u])per i)arts of a healthy culm, but late in the season lateral buds often develop from the upper nodes of tall culms whose terminal bud is badly affected with the fungus. The leaves from such buds may or may not be infected witli the fungus. The Mycelium in the Lower Buds and l' ndcrground Stems A careful study was made of l)uds which arise low down on the culms and on the rhizomes and which would not develop into culms or rhizomes until another season. It is readily seen that the presence or absence of the fungus in the case of the buds is a crucial point in determining the life period of the fungus and the source of infection of the unfolding leaves. Horizontal and cross sections were made of buds of different ages and sizes which were variously situated on basal ])ortions of plants whose aerial culms showed infection. In the majority of these buds the myce- lium was found to be present. Figure 6 6r is a drawing from such a bud whieli was situat<'d just lielow the surface of the ground and which would develoj) as a culm in the following season. The successive sections showed, as did the longitudinal sections of other buds, that the fungus was rather irregularly distributed on and through the rudimentary leaves and that it was more abundant near the apex of the buds. Often but one side of a bud was infected by the fungus. When such a bud un- folds, but, a lateral half of certain leaves will be diseased, a con- dition which explains the common partial and lateral infection of aerial leaves already described and illustrated. In these buds the hyphae pavSs through the ci^ll walls freely in the manner shown in Figure 6 G. The cell walls in all cases appear normal but the cell contents have entirely disappeared. The chief difference between the effects of the fungus in the foliage leaves and in the tissues of the leaves of dormant buds is that in the latter there seems to be no alisorption of cell walls except at the i)oints of ])htl>' eiilaiship such as is com- monly understood in which the fungus retains a constant posi- tion in reference to the growing points. In the roots, rhizomes, and stems the fungus attacks the older and hence weakened cells. This is in marked contrast to the action in the leaves where it attacks vigorously the active mcsophyl cells but it is to be noted that the hyphae are unable to penetrate into the growing apex of the buds. In larger roots such as shown in Figure 8 D there was an abundant but rather irregular distribution of the mycelium throughout the cortical tissue, with much the same characteris- tics as are seen in the cortex of the underground stems. The nests of mycelium and the bladders were formed within the cells altliough the latter were not grouped in a belt or zone. The my- celium was both inter-and intra-cellular. In the root cells of this region the cytoplasm forms an extremely thin layer which was difficult to identify even in uninfected cells. The cell walls were of normal thickness and were not collapsed. There was a strong tendency for the mycelium to extend longitudinally through the root yet there was here soiue evidence of its spread in a radial direction also. (See Figure ID.) In the spur roots there is considerable variability in the be- havior of the fungus. Here the m3'celium is confined chiefly to the layer of cells immediately surrounding the central cylinder. It can be traced into this zone from the main root and its dis- tribution is here decidedly in the direction of the growth of the rootlet. Here there is an opportunity to study the relations of the hyphae to the protoplast for they penetrate cells while the cytoplasm is conspicuous and the cell walls are thin. The fungus advances to the extreme tip which is devoid of a pro- nounced embryonic region. AVhen a hypha passes into a living cell in this region it first beeomcs irregularly enlarged or lobed 230 Wisconsin Research BulliEtin No. 18 and is somewhat coiled al)Out or pressed upon the nucleus which in turn becomes rather irregular in shape. The cytoplasm be- comes dense and granular or slightly stringy and stains a deep orange or red. (Figure 7 C, E and G.) Later the nucleus disappears in some cases evidently after fragmenting and the cytoplasm is transformed into irregular, dense, deeply staining particles scattered about within the cell wall. (See IF). In some instances the fungus tilaments within such a cell seem also to disintegrate so that fragments of tlic mycelium are mingled with the debris of the protoplast. In the majority of infected cells, however, the mycelium remains intact while tlie cytoplasm and nuchnis are undergoing disintegration. xVdjoining but un- inf(>cted cells show^ the cytoplasm and nucleus as normal and faintly staining structures. In certain of these infected cells these changes in the pro- toplast are accompanied by the formation of bladders which are intercalary or perhaps occasionally terminal swellings of the hyphae. They are always intracellular and are of various sizes. They begin to appear in the small cells of roots in the region of the advance of the fungas and they may be found fully de- veloped in the older roots and in the basal nodes and the sur- rounding scales. When young they show a finely reticulated structure (Figure 8 F.) Later they are more dense and gran- ular. The greater number found in old roots are entirely empty and possess rigid and unbroken cell walls although in a few cases they appear wrinkled or shriveled. In early spring the vigorously growing roots are not infected. Soon these roots cease their rapid growth and begin to send out short slowly growing side roots. The fungus mycelium probably gains entrance both at the base from the stem and by direct penetration fi-om the soil. It advances in the root toward the growing i)oint. The latcrjil I'oots aic infected directly from the main root soon jil'lcr tlicy jiic (irst formed and as a result are stunted and slightly liypci'l ropliicd. These infected roots may continue to liv(^ into at least tlie second season. Some side roots escape infection completely oi- for a longer time and these elongate in a normal fashion. Summari/ iiis is c(ie\istent in leavi's. stems, buds, rhizomes and roots ol' the same plant, its general distribu- tion in umlei-^i'onnd peii'nnial pai'ts an:! its existence in buds A SCLKROTIIM DiSKASK OI' Co.MMON (iHASSES 231 wliicli jiic to di'vclu]) ill siict^i cdiii^' sc-isoiis make clear tlie jjeren- iiial iiafun' of the iiiyeeliiiiii. Jt is of espeeial signifieance that the iii\ccliuiii was found in j^reater or less abuiulanee in the scales and youii. 116) has also most clearly pointed out the two methods of origin of sclerotia. The sclerotia of Basidiomycetes (Agaricus, Coprinus, Typhula, etc.) start by the iiitei'lacing of branches from a single hypha, while the scle- rotia, of Ascomycetes wliidi he investigated begin as a plexus of several fih^mciits, as previously noted for this fungus. The sclerotia which develop on the host plants are rounded and smooth on llicii- entire surface except on the side which was appressed to the hvif and here the sclerotium is usually flattened and rugose to conform to the ridges in the surface of the leaf. A ScLEROTir.M Disease of Common Grasses 233 AVIicii I'lill.N' mat HIT tlic\- usually drop from the leaf. Attached to many of them are oi'ten short strings of dry and dead myce- lium from which tiie sclerot.ium arose and in which the growing sclerotium is imbedded. These appear on the immature sclero- tia especially somewhat like roots, or rhizoids, and hence prob- ably suggested the specific name given by Auerswald. It seems probable that spores when they are produced develop first saprophytically in the humus soil. Investigations and ex- periments are still in progress to determine more completely the idtimate fate of the sclerotia. As many as 500 sclerotia of the 19] 1 crop are now planted in pots and in bottles which are be- ing handled in a variety of ways. There is some evidence that sclerotia may sju'out vegetatively. Fully mature sclerotia were taken before they had dried, and placed on sterile sand in small pots which were covered with glass lids and kept moist. A thin felt of mycelium developed from the sclerotia and spread over the sand where it continued to thrive for several months. When bits of this mycelium were transferred to various media the typical growth resulted. Other sclerotia were placed directly on media and in the majority of eases developed a growth typical to the various media. Old sclerotia which were tJioroughly dried failed to produce growth of any kind when thus treated. Infection Experiments For these experiments healthy uninfected plants of C. cana- densis were grown from rhizomes transplanted to pots of sand or garden loam. When the culms were from three to twelve inches in height masses of mycelium from various cultures we,re placed on leaves of various ages and the plants were then kept in bell jars. In no ease did the mycelium establish itself on the leaves. The following method was then tried. Vigorously growing pure cultures were grown on hard potato agar and on lima bean agar in test tube slants. A test tube with a culture was inverted, the cotton plug .removed and the open tube slipped over a culm and so adjusted in a clamp stand that the leaves came in contact with the mycelium. The plug was then replaced and thus the unin- jured mycelium was brought in contact with leaves and the whole was enclosed in a moist chamber formed b}^ the plugged test tube. In all, twenty-five experiments were tried on culms 234 AViscoNsiN Keseauch Bulletin No. 18 of various ages and ou leaves in various stages of developnieut. The mycelium would develop ovci- and about the leaves and al- though left for several days, wiien tlic test tube with its cultvire was removed and the culm enclosed in a moist chamber, the mycelium adhering to the leaves died, the leaves continued to develop normally and there was no evidence of infection. Infected cuhns wei'e l)r()uglit in from tiie field and enclosed in a ])ell jar until tliere was abundant development of aerial myce- lium and then placed in contact with healthy leaves. No infec- tion resulted l)y this method. rLxpei'iiiients to test the ])ossibil- ity of the infection of seedlings through th(> soil are now l)eing carried on and have not yet given any definite results. These experiments seem to indicate cleai'ly that the mycelium can not penetrate into sound leaves after they have opened from the bud and that there is no spread of the fungus from culm to culm aerially. Cultural Studies The fungus was obtained in pui'c cultui-e by placing fragments of infected leaves in Petri dishes ])()ui('(l with lima bean agar and with liai'd potato agar. Abundant gi'owth of mycelium with forn)ation of sclerotia i-esulted and ti'ansfers were readily made tf) media in test tubes, 'i'lie fungus has been ke|)t in cultui'c^ since April 1010 and its behavior on various media studied. All cultures wei'c grown in a refi'igerator at a temperature of about 16° C. T/i»ia Bran Agar- On this medium thei-e is luxuriant and rapid growth. At first a h'ne cottduv' Ia\-ei' of mycelium sj)reads over the sui'face of the medium and fi'om this there is a copious aerial growth. Within fi'om ten to twenty days sclerotia begin to foruL When matni'e many of th( se ai'c rounded and similar in si/e to those formed on the host in the Held. l)ut the individual sclei-otia may grow togetliei' to make ;ui ii'regulai- mass, oftrn l.T) cm. in diaiiietei-. Spt for a period of IT) months nnd although the medium hati shrunken to oni^-thii'd of its original \-olume the myceliinii was still vigoi'ous in small patrhes and new sclerotia wei e l;(.'iiig foi-med. :i Lima bean agar: 1000 cc. water, 100 g. ground beans; soak 30 min., then boil :?0 min.; express juice and restore to 1000 cc; add 10 g. agar, cook in steamer, filter, autoclave at 5 lbs. A S('i,i;i;(i'rii M Diskask op Commont Gkassks 235 lldid I'lhilo A(/(ir^ (liowili was less i'iii>i(l nn this incdiiitn that tliat ()l)taiii('(l on lima Ixaii af^'ar. A tVw dctinitt' sderotia were t'oniicd Iml as a rule lar^c iiicyulaf sponji'V masses formed (in tile suifacc of tlic mcdiinu as a crusted stroma. Tlie surface of tiiese remained spongy oi' c1 ion of young soodlings. A SCLKIJOTICM DiSKASE OF COMMON CiKASSES 2:37 ClIAIvACTKlilSTICS OF THE FuNGUS ; DISCUSSION AND COMPARISONS The ruiigus Schroliuni rhizodcs in ils relations to the peren- nial grasses studied exhibits a combination of eliaraeteristics not hithertcj asei'ibed to any one fungus. In the aerial parts of the host tiic liehavior is somewhat simi- lar to that of various smuts, especially as shoAvn by the recent studies of Lutman (1910, p. 1204). He found for Ustilago levis on oats that in the growing plants the mycelium is most abund- ant at the nodes and in the growing points and that many leaves contained the fungus to their tips. The mycelium remains in or near the growing points but is intercellular. McAlpine (1910, p. 10) has pointed out the tendency for the mycelium of Ustilafjo nudn on barley to persist in the under- gi-ound portions of the plants and penetrate into new culms which were induced by repeated cutting of the old culms. The seed fungus of Lolium temulentum as described by Free- man (1903, p. l-l:-16) shows cpiite a different behavior. Here the fungus seldom enters the roots and leaves, but lingers in the growing points and finally enters the nucellus layer of the seed from which it later infects the germinating embryo. There is, however, no destructive effect on the seed ; in fact, Freeman notes (page 5) that infected seeds are larger and better devel- oped, suggesting a symbiotic relationship which Hiltner's (1899, p. 835-837) work seems to show is due to the fixation of atmos- pheric nitrogen by the fungus. ^liss Ternetz (1907, p. 359) in her studies of various mycor- i-liiza finds the seeds of the macrosymbionts infected and raises the question as to the source of this infection. In Andromeda polifolia she observes mycelium in the rind of the older twigs which had reached this point by the coordinated growth of the plant and the mycelium. Thus she finds the mycelium wander- ing from roots into aerial parts, but she notes that it avoids chlorophyl bearing cells. Since she did not find mycelium in the flower stalks she believes that the source of the infection in the flowers is external, yet it seems probable from the positive evidence given above, that this infection of the fruit comes vegetatively from the roots. As to its distribution within the host, Phoma radicis An- dromcdae, as described by Ternetz seems to be intermediate be- 238 WiscoxsiN Heseakcii Bulletix No. 18 tweeu strictly root inhabiting myeorrliiza and >!ch rolluni rltiz- odcs on ('alamofjroslis raiiadf nsis as described. Turning to t.he behavior of Sch rofium rJnzodcs in roots and niidei"ground steius. we find that its inorphob)gical characters and host relations are siicli tliat it might be considered as a myeorrhi/al fungus. 'I'his tei-iu. hi)\v( vei-. has been so loosely applied to root iiihabiting fungi that it, has at present little specific significance. Following the ratlier limited observations of Kamienski (1881 and 1882), Frank (1885 and 1887), and Woronin (1885), the more extensive studies especially by Scldieht (1889), Janse (1896), and Stahl (1900) liave shown that root inhabiting fungi are present in a large number of flowering ])lants. So called myeorrliiza have also l)een found in the tissues of certain cryptogams as has been shown especially by Atkinson (1898), Janse (1896), Nemec (1899), Lang (1899) and Campbell (1908). It has been established that annuals, biennials, and perennials growing in all sorts of soil, belonging to a wide range of fami- lies and exhibiting autotro])ic as well as holosai)i'o])hytic modes of life, may ])ossess root iuhahiliug fungi of i-udotropic or ec- totropic character. Schlicht (1899, p. 26) includes in his list of s])ecies possess- ing endotropic myeorrliiza the grasses //Olnis hnmius, Festuca ovina, Agrosfis raniiiam, Aim ca not famish food to the mycelium outside of the particular cells in which they are situated. ^Fagnus (1900) described the conditions which exist in the non-chlorophyl bearing plant Xeottia nidus avis. In the cortex of the roots and the rhizome he finds an endotropic myeorrhizal fungus. In certain cells the fungus destroys the protoplast and 240 "Wisconsin Research Bulletin No. 18 forms thick walled hyphae and bladder-like structures which he considers to be organs for storage of food. In the outer and inner layers of the zone of infection, however, he finds digesting cells in which the fungus is for the most part digested. Thus he finds that the fungus is locally parasitic in some cells and that in others it 3'ields its substance to the host cells which con- tinue to live. Shibata (1902) finds much the same condition in Podocarpus, Alnus, and Psilotum. In the digesting cells of the last tw6 named the wall substance of the fungus remains as densely stain- ing clumps. In the cells destroyed by the fungus he notes the formation of vesicles which are filled with food. Arzberger (1910) has recently investigated the root tuber- cles of Ceancilms americanus, Ehuagnus argentea, and Myrica cerifera. He finds in Ceanothus that hypertrophied cells and nuclei result from the infection, that the cell walls of host cells are dissolved and that later the cytoplasm and nuclei of these cells are absorbed. Soon the cell content of the fungus dis- integrates, but not until the contents of the host cell are used up. In the root tubercles of the above named species he finds ulti- •mately both the host cell and fungus die "as a result of their previous relationship," but he concludes that the material is in some waj' used by the adjoining health}^ uninfected cells. Regarding the vesicles or "sporanges" which are formed, Arzberger finds them to be terminal, spherical or pear shaped bodies. In Ceanothus and Myrica their contents break up into a few segments by a process which he considers analogous to spore formation in other fungi. Physiological Significance In all of the cases described by these authors, with the excep- 'tion of Schlicht, in some cells at least the fungus is wholly or in part desti-uctive. They all find evidence, however, of a sym- biotic relationship in the digesting cells or in the benefit which the surrounding cells may gain. The anatomical evidence for symbiosis in the case of endotropic mycorrhiza seems to rest on the observed ability of individual cells to recover from the in- vasion of the fungus or in the ability of adjoining cells to pro- fit by the activities of the fungus though it may be destructive of certain individual cells. A ScLEROTiuM Disease op Common Grasses 241 The characteristics of these endotropic fungi within the cells iluring the early stages of infection and the formation of blad- ders as described by Magnus (1900, pp. 10-19), Shil)ata (1902, pp. ei6, 657-660) and most especially by Groom (1895 pp. 333, 334 and 338-340) bear some noticeable resemblance to Sclero- tiuni rhizodes although it is to be noted that the macrosymbionts referred to above range from autotropic to holosaprophytic plants. The later developments of the Sclerotium rhizodes in the roots suggest a somewhat different relationship. I find no conclusive evidence that there are here digesting cells which overcome the fungus and resume a normal appearance after infection. A cell is never entered by the mycelium until it has ceased to divide and until it has reached nearly its maximum size. An immediate result of the invasion of such a cell is the accumula- tion of a large amount of densely staining fine granular cyto- plasm as is shown in Figures 7 C and G. Soon coarse irregular shaped solid bodies appear which are like those described by ^Magnus and Groom and assumed by them to be the disintegrat- ing fungus. Here, however, these bodies appear while the my- celium is intact and healthy. These clumps are plainly the pro- ducts of the host cell; perhaps resulting from its stimulated activity. The healthy mycelium can often be found passing through empty cells in close proximity to cells showing these "clumps." The bladders are often of such size or number that they completely fill the cells leaving no trace of cell contents. There are furthermore fungus filaments which give direct con- nection between the mycelium and the bladders in difl'erent cells. When empty the thick heavy walls of the bladders re- main for the most part rigid and intact. These conditions lead one to the conclusion that in this case the intercalary bladders are formed at the expense of the host cells and that the food thus stored is used by the mycelium in other portions of the host tissue. From the histological and cytological studies which the writer has made of the fungus, it appears that its relation to the in- dividual cells of the roots and the rhizomes is one of mild para- sitism and that the nearly mature cortical cells are invaded and the cytoplasm digested. As already shoAvn the fungus is still more destructive in the leaves. 242 "Wisconsin Research Bclletix No. 18 ]n ccftain areas of inarsli meadows the Calamagrostis plants do not harbor the fungus. At least it does not appear on the leaves and thus far the study made of the roots does not show its presenci' tlici-e. The l)est develoi)ment ot tlie grass is iu these fungus free areas. It has been pointed out 1hat a lai'ge nuiiilx-r of eulms are killed by the fungus especially iu the early spring. The effect of this is juarked in several areas that have exhi])ited general in- fection during the past four seasons. In areas where in 1908 there was a vigorous growth of Calamagrostis with considerable fungus appealing on the culms and in wliicli the appearance of the fungus was still more marked in 190!) and 1910, there was in 1911 a marked thinning out of the grass. The observations in the field therefore agree with the conclusions based on the rela- tions found in the roots: viz. that the fungus is perennial in the roots with injurious effect. The original theory of Kamienski (1881) was extended by Frank to apply to many more cases of fungi in roots. Frank (1885. ]). 88) maintained that the fungus conducts and pre- pares food t'oi- the host and especially that it assimilates humus compounds. He considered that even purely endoti-opic myc- oi-rhizal fungi receive food from external sources and thai the root contained, as it W( re, a tra})ped fungu'-. the host ultimately digesting the fungus (1892 ]). 267). rii'oom's \\vw (1895. ]). 854 and 'AiiG) differs iii that he eon- sidcrs that the fungus in Tliisniia first absorbs fond from the cells, that the fungus is not completely digested by tlie host cells, and that in the outer layers it '"act.ually ])rt)hts by the symbiosis" (1895, ]). 356). Stahl (1900) argues at length in t'aNoi- of the view that botli ectotropic and endotropic mycorrhi/a furnish su])i)lies of elabo- rated organic compounds fi'oni without, lie attem])ts to show that the liosi ]»lants in such cases are weak in photosynthesis and that they also manifest a decrease of ash content on ac- count of this weak power of assimilation. In the case of Sclerotium rhiz()(](s in the roots of C. cana- flosis it is clear that the fungus filaments do pass out into the soil M'hei'c there is i-cason to believe that they can thrive saproi)h\t ically. It is ]>ossible that tlu'se tilamcnts may fur- A ScLEROTiuM Disease of Common Grasses 243 nisli the cndutropii- ])()rti<)ii.s with various luineral salts and organic compounds wliicii may at first be drawn upon by the host cells thus accountiug for the enlarged cells and the dense cytoplasm found in sjjur rootlets during the early stages of infection. Admitting tin- possibility of a benefit of this sort, it is certain tliat the ultimate effect of the fungus both on indi- vidual cells and on tiie entire plant is that of a parasite es- pecially when conditions favor the development of the fun- gus on the leaves. There have ])een numerous investigators who luive noted para- sitic l)ehavior in various mycorrhizal fungi. Reess (1885) studied Elaphomyces on pine and believed that the fungus obtains food from the roots of the pine. Ilartig (1886) opposed Frank's views of symbiosis and considered that the fungi in question were paraijitie. Xadson (1908) found "raycorrhiza" penetrat- ing roots and killing large niunl)ers of oak seedlings. Boulet (1910) reports the presence of endotropic myeorrhiza on al- mond, apricot, peach, cherrj', plum, apple, pear, etc., and states that the fungus lives as a parasite but still has a beneficial effect on the host except when it attacks the roots with unusual vigor. It is such evidence as this that indicates the unsatisfactorj'' state of our knowledge concerning the intimate physiological relations which exist when a fungus resides as is the case in endotropic myeorrhiza within living plant tissues. After all we .judge parasitism and symbiosis by rather gross general re- sults. Possibly the same fungus may under different conditions manifest varying degrees of parasitism or of symbiotic re- lationship. A mycorrhizal fungus on an oak may be at one time symbiotic as Frank maintained and again the same fun- gus may be parasitic as Nadson reports. On this point the behavior of Sdcrotium- rliizodes is espec- ially interesting. Judged solely by its behavior in the roots it might be considered as entering into symbiotic relationship. At any rate it is not here apparently seriously parasitic. It may bring into the root various mineral and organic food substances which may in part ])e appropriated by the root tissues. There is no evidence, however, that the fungus is itself digested. In the leaves, however, it is vigorously parasitic. At the time when the culms have elongated there is no direct 244 AViscoNsiN Research Bulletin No. 18 connection between the mycelium in the leaves and that in the roots and it can not be considered that the vigorous develop- ment of the fungus in the leaves can supply the underground m;}-celium with food. The mycelium is perennial in the roots and probably in the soil, but is rather short-lived in the leaves. The vigor of the parasitic development on the leaves varies considerably with the season and with the species of host. It is certain that the general effects of the fungus on the grasses here discussed is tliat of a parasite. Economic Significance The occurence of this fungus on C. canadensis decreases the yield and quality of the hay obtained from marsh meadows. When the infection is as high as 47 per cent, as was the case in 1911 on the marsh meadow most carefully studied, this loss is considerable. The ability of the fungus to infect and injure such important gi'asses as Phleum prate^ise and Poa pratensis is also a matter of economic significance although the injury ta these two grasses is not serious at present, so far as observed. Cultivation of the soil and crop rotation naturally suggest themselves as a means of control for this fungus. This would be practicable in ease the fungus becomes destructive in up- land meadows. On nearly all the meadows where this fungus has been found tillage is now impossible and until improved drainage makes cultivation possible the fungus will necessarily run its course. The observations made thus far indicate that Agrostis alha is not attacked and if this be true the introduc- tion of this grass into infected marsh meadows might be advis- able. The susceptibility of Calamagrostis canadensis to at- tacks of this fungus makes the use of this species on semi-re- claimed marsh lands of doubtful value. Summary The fungus Sclcrotium rliizodes attacks the leaves of various grasses causing them to become dry, rigid and bent into char- acteristic crooks. Upon the leaves appear felfs of mycelium from Avhieh sclerotia develop. The development on the leaves is most vigorous during April and ]\ray when the death of many entire culms results. A ScLEROTiuM Disease of Common Grasses 245 At ]\Iadison, Wis., the fungus has been found on eleven dif- ferent grasses: Phalaris arundinacea, Calamagrostis neglecta, Poa pratensis, Pamcularia ncrvata, Phleum pratense, Hordeum jicbatiim, Bromus ciliatus, Eatonia pennsylvanica, Agropyron caninum, Agrostis hyemalis, and Calanvagrostis canadensis. It is especially injurious and destructive to Ccdamagrostis canadensis which serves as its principal host. The fungus is vigorously parasitic in the leaves, less so in the buds and the stems, and but slightly so in the roots where it assumes some of the characters usually associated with mycor- rhiza. Fungus filaments extend out into the soil where they live more or less independently. The mycelium is perennial in the soil and in the underground parts of the infected plants. The infection of aerial parts occurs from the underground parts of the plant. The mycelium is sterile so far as has been observed to date. The production of spore bearing structures from sclerotia has never been observed. The fungus is of considerable economic importance. It de- stroyed or dwarfed as many as 47 per cent of the plants of Calamagrostis canadensis on one marsh meadow near Madison in the season of 1911. This fungus is quite generally distributed throughout Wis- consin. It has not been reported elsewhere in America; pre- sumably it is quite widely distributed, however. Its parasitic attacks on the leaves of various grasses has long been known in Germany, Belgium and Scandinavia. Further data are desired on the following points: additional host plants, geographical range of the fungus, economic im- portance, germination of the sclerotia, and infection of seed- lings. 246 "Wisconsin Research Bulletin No. 18 BIBLIOGRAPHY Ai'z])erger E. G. 1910. Fungous Root Tubercles. Ann. Rpt. Missouri Bot. Garden 21: 60-102. Atkinson, G. V. 1893. Symbiosis in the Roots of the Ophioglosseae. Proc. Amer. Assoc. Adv. Sci. 42: 254-255. Also in Bui. Torrey Bot. Club 20: 356-357. Brefeld, 0. 1881. Untersuchungen uebe,r Schiramelpilze, 4. Boulet, V. 1910. Sur les mycorhizes endotroj^hes de quelques arbes fruitiers. Compt. Rend. Acad. Sci. [Paris] 150: 1190-1192. Campbell, D. H. 1908. Symbiosis in Fern Prothallia. Amor. Nat. 42: 151-165. Davis, J. J. 1893. A Supplementary List of the Parasitic Fungi of Wisconsin. Trans. Wis. Acad. 9: 158-188. Frank, A. B. 1881. Handl)ucli der Pflanzcnki'anklu'iten. 1885. Neue Mitteilungen ueber die IMycorhiza der Baume und der Monotropa h.iipop\i]i>i. P)or. lleut. Bot. Gesell. 3: XXVII— XXX. 1887. Sind die AVurzelanscliwcllungen der Erlen und Elaeagnaccen Pil/.gallcn .' Ber. Deut. Bot. Gesell. 5: 50-58. 1887. Ueber neue Mycorhiza-Formen. Ber. Deut. Bot. Gesell. 5: 395-408. 1892. Lehrbuch der Botanik 1. 1896. Die Krankheiten der Pflauzen. Freeman. E. ^I. 1903. The Seed Fungus of Lolium frmuhnfum L., Dai-ncl. PliiL Trans. Roy. Soc. London B 196: 1-27. Groom, P. 1895. On Tliis)iiia Ascroe (Beccari) and its Mycorhiza. Ann. Bot. 9: 327-361. A ScLEROTiuM Disease op Common Grasses 247 Hartig R. 1886. Ueber der syin})iotischen Erscheinungen im Pflan- zenleben. ]^ot. Centbl. 25: 350-352. Hiltner. L. 1899. Uel)cr die Assiuiilation des i'roien atmospharischen Stickstoffs diirch in oberirdisehen Pflanzenteilen b bcii.b' .Myoclion. (\m\h\. Bakt.. 2. 5: 831-837. Janse. J. M. 1896. Les endophytes radicaux de qiielqiies plantos java- naises. Ann. Jard. Bot. Bnitenzorg 14: 53-201. Kamienski, F. 1881. Die Yegetationsorgane der Monolropa liypopitys L. P,ot. Ztg. 39 : 457-461. 1882. Les organes vegetatifs du Monofropa liypopitys L. :\rein. Soe. Sei. Cberbourg 24: 5-40. Lang. W. H. 1899. The ProtbaHns of Lycopodium clavafum. Ann. Bot. 13: 279-316. Lntman. B. F. 1910. Some Contributions to tbe Life History and Cyto- logy' of tbe Smuts. Trans. Wis. Acad. 16 : 1191- 1245. Magnus. W. 1900. Studien an der endotropben ]\Iyeorrbiza von Xcottia Xidus avis L., Jabrb. Wiss. Bot. 35: 205-272. McAlpine, D. 1910. Tbe Smuts of Australia. Nadson. G. A. 1908. Zur Lebre von der Symbiose. I, Das Absterben von Eiebensamlingen im Zusammenbange mit der ]\Iyeorrliiza. Jabrb. Pflanzenkrankbeiten. St. Petersburg. 2: 26-40. Abst. in Centbl Bakt. Abt. 2. 26: 100-101. Exp. Sta. Record 22: 722-723. Nemec. B. 1899. Die ]\Iyk(>ri"lii/a einiger Lebermoose. Ber. Deut. Bot. Gesell. 17: 311-316. Rees. :\r. 1885. L^eber Ebipbomyces und sonstige Wurzelpilze. Ber. Deut. Bot. Gesell. 3: 293-295. 248 Wisconsin Research Bulletin No. 18 Saccardo, P. A, 1899. Sylloge Fungorum. 14. Schlicht, A. 1889. Beitrag zur Kenntniss der Verbreitung und der Bedeutung der Mykorhizen. Inaug. Diss. Erlan- gen. Shibate, K. 1902. Cytological Studien ueber die endotrophen Myeor- rhizen. Jahrb. Wiss. Bot. 37: 643-684. Sorauer, P. 1886. Handbuch der Pflanzenkrankheiten. Stahl, E. 1900. Der Sinn der Mycorliizenbildung. Jabrb. Wiss. Bot. 34: 539-668. Stout, A. B. A Statistical Analysis of the Vegetation of a Wild Hay Meadow. To be published in Trans, Wis. Acad. 17. Ternetz, Charlotte. 1907. Ueber de assimilation des atraospharischen Stick- stoffs durch Pilze. Jahrb. Wiss. Bot. 44: 353- 408. Tubeuf, K. von. 1897. Diseases of Plants. Eng. Trans, by Smith, Woronin, W. 1885. Ueber dir Pilzwurzel von B. Frank. Ber. Deut. Bot. Gesell, 3: 205-206. Explanation op Figure 1 This photograph shows the habit of growth of Calamagrostis canadensis and the general effects of the fungus Sclerotium rliizodes as it appears early in the season. Three of the culms here shown were badly infected, the leaves were shriveled and rolled especially toward the tips, the characteristic crooks were present and the felt of mycelium can be seen in the picture at points marked m. Young sclerotia were present at points marked s. The culm which stands second from the terminal bud of the rhizome had escaped infection. The investigations show that in such plants as this the fungus is present in the leaves, stems, buds, and roots and that the mycelium is also present in the soil and on the roots. A ScLEROTiuM Disease op Common Grasses 249 Explanation op Figure 2 A life sized picture of a rapidly growing culm of Calama- grostis canadensis which was vigorously attacked by the fun- gus. The appearance of the crooks and their method of forma- tion are here well shown. The tips of the unfolding leaves were penetrated by the mycelium which held them together in a roll and hence as the leaves developed they arched upward. Often the growth tensions are such that a portion of the crook is twisted and folded as is the case here. Death of the entire culm soon results from such a vigorous development of the fungus. Explanation op Figure 3 A. Upper portions of two older culms of Calamagrostis cana- densis showing conditions somewhat different from the previous figures. In tlie left hand figure at point marked m is seen a. partial lateral infection of a leaf. This is a common phenomenon. The right hand figure shows a complete series of crooks, five in number. The tips of the leaves are held at the points lettered. These figures illustrate the fact that every leaf of an infected culm possesses the fungus and that the mycelium gains entrance to the leaves when they are in the bud. While the tips of the leaves here shown are completely de- stroyed there is a less vigorous development of the fun- gus than is shown in Figure 2. This condition is seen during the middle of the season, especially during a rather dry period when few sclerotia are formed. Such culms may grow for some time but seldom produce flowers and seeds. B. Portions of C. canadensis culms and leaves showing typical matured sclerotia as they are produced in the field un- der favorable conditions. Culms such as are shown at left hand side soon die from the attack of the fungus. (Natural size) Explanation op Figure 4 A. A typical culture of ScJerotium rliizodes on hard potato agar showing abundant growth of mycelium and the crusted pseudo-sclerotia. Culture is ten months old. 250 Wisconsin Research Bulletin No. 18 B. Culture on lima bean agar nearly ten months old. The mycelium is less abundant than in A but the selerotia are more perfect. C. Development of the fungus on bean pods with a few well formed selerotia and abundant mycelium. Explanation of Figure 5 A. Two cultures, 15 months old, on hard potato agar. The mycelium is still quite vigorous and the large compound irregular shaped selerotia are mature. B. A pure culture of the fungus on peaty marsh soil upon which infected Calamagrostis canadensis had been growing. The soil with various grass roots included was sterilized and fragments of the mycelium from test tube cultures were placed on the soil. At the time the photograph was taken this culture was ten weeks old. The picture shows the mass of white mycelium that covered the surface while toward the bottom may be seen clusters of the mycelium which also spread cpiite generally through the soil mass. These soil cul- tures demonstrate that the fungus can live as a sapro- phyte in the soil. Explanation of Figure 6 All (li'awings of this Figure are from CaJaHiaf/rostis canadensis. A. Part of a cross section through such an infected leaf roll as is shown at m, Figure 3 A. Tlie drawing is fi'om the outer leaf of tlie I'oll. The mesnpliNl of the leaf is en- tirely destroyed. (X 100). B. Drawn from tiie center of a haillx- infeeted i-oU showing desti'uetion and eoll;i])se of tlu^ tissues of the bundle. A and B show the conditions which prevail in the outer and inner portions of an infected roll of leaves. (X140). C Fi'om the same cross section as A, l)ut from the ])order of the infected area showing the mycelium advancing through the tissues. (X 100), A SCLKUOTHM DiSKASE OP COMMOX GRASSES 251 D. Distribution of the niyct'lium ahuut the growing points of a terminal l)ii(l of a eulrn the expanded leaves of which sliowed infection such as seen in Figure 1. (X 100). E. Tile 1)U(1 from wliich I) was drawn. Sliowing the distri- I)ution of the mycelium. P. From a cross section of a stem near a node. This particular oidm was two feet tall. Its leaves were infected. A node situated imnu'diately below the surface of the ground witii its ])ud was fixed during the first week in August. At that time the culm was mature. The bud was one which would develop during the next year. The mycelium was present in the outer layers of the culm as shown in tlie segment drawn. Drawings similar to this could be shown for the various nodes of infected culms. (X 100). G. Portion of the bud mentioned above drawn in its relation to the stem F and from the same cross section. The mycelium is present in the tissues of the bud leaves as well as between the leaves where it remains over winter. (X 100). Explanation of Figure 7 All from CaJ(nnagrosfis canadensis unless otherwise stated. A. Portion of a basal node showing the extent of penetration by the mycelium and the vesicles which are formed. (X 100). B. Drawn from a longitudinal section through a basal node and its bud. a is a part of the node and h is from the outside leaf of the bud. The mj^celiura is here shown passing from stem to bud. ^Material was collected dur- ing the summer. (X 100). C. From a longitudinal section of a fibrous root. Shows con- dition of cells soon after penetration by the mycelium ; the nuclei are in various stages of degeneration and there is an accumulation of dense material in the cytoplasm. (X 140). D. Cross section of an older root (the main root of Figure 8 D) showing mycelium and vesicles in empty cells. (X 100). 252 Wisconsin Research Bulletin No. 18 E. Cross section of a spur o,r lateral root. Shows the tendency of the fungus to develop in the innermost layer of the cortex. (X140). F. Mycelium in practically empty cortical root cells. The densely staining granules are the remains of the proto- plasm of the host cell. Mycelium is intact. (X 140), G. From a cross section of a lateral root of Calamagrostis neglecta. Mycelium is here seen on the surface of the root. Formation of vesicles, accumulation of dense protoplasm and abnormal nuclei are here shown. (X 140). H. From root of Poa pratensis. Mycelium in the cortex. (X 100). I. Early stages of vesicle formation. Nuclei and cj^toplasm of host cells nearly disappeared. Later stages are shown in Figure 8, F and G. Figure 7, C, E, F, and I, and Figure 8, F and G present a scries showing the stages in the development of the mycelium and vesicles, to- gether with the accompanying disintegration and dis- appearance of the protoplasm of the host cell. (X 140). J. A few cells of the mycelium grown on cooked potato show- ing the prevailing two-nucleated condition. (X 630). Explanation of Figure 8 A. Drawing (X 140) of the mycelium as it develops on cooked potato. The oidium-like cells do not function as spores, but are similar to the cells in a young sclero- tium. B. Characteristic mycelium from the surface of a leaf of C. canadensis (X 140). C. Drawing (X 140) of mycelium which developed from the cut end of an infected root. D. A portion of an infected root of C. canadensis is here shown (X 10) with the enlarged spur rootlets and the myce- lium as it exists in the soil and on the surface of the roots. The large bladder-like vesicles as they appear in the soil are here shown. E. A small portion of a rootlet (X 100). The mycelium is here shown adhering to the surface of the rootlet and pene- trating into the root tissues at two points. A ScLEROTiuM Disease of Common Grasses 253 F, Typical vesicles as they appear within cells of the cortex of roots of C. canadensis (X 175). G. A single spherical vesicle within a cell of a root of C. neglecia (X 175). H. Diagrammatic cross section of a leaf which is laterally in- fected. The dark portion represents the distribution of the mycelium. The tip of a second leaf is shown within the roll. Wisconsin Reseakch Uim.ktin No. 18 254 Figure 1. C. canadensis. sl;owin;r tlio sroniM-al symptoms of tlie Sclerotium disease. A ScLKKoTii M I)isi;a>k ok Common (iUASsKS Figure 2. The characteristic "crooks" resulting from a vigor, us attack of the Sclerotium WisroxsiN Kkseaucii Bulletin No Is 250 to U - 4^ - 01 1 Illustrations of various leaf symptoms associated witb the Sclevo- tium disease A Sci.i;i;ici'ii .\i Diskask ok ( 'o.m mox (iit.vssEs 2")7 Figure 4. rure cultures of ScUTotium rhi7.ii(l<'s on various uiedia Wisconsin Keskaucii P>ii.!.i:ti.\ Xo. IS 2r)S KiLiiirc of tlio Sclerotium ; A shuwinir the lar:^e sclcr.itia : F> t'.ie mycelium in soil A ScLEROTiuM Disease of Common Grasses 259 Figure 6. The mode of tissue invasion of grass leaves and stems by the Sclero- tium mycelium A SchKKOTII .\l DiSKASK OF CoMMON ( J KASSES 261 0 ll li f^ I /( // 11 • B. ^J \A \:^ y ^ ft .^^^^ r^^ M]^ \( H. Figure 8. A-C, mj-celium of the sclerotium; D-G, infected roots; H, infected leaf Research Bulletin 31 ' ' March, 1914 The Control of Damping-Off Disease in Plant Beds JAMES JOHNSON AGRICULTURAL EXPERIMENT STATION OF THE .UNIVERSITY OF WISCONSIN MADISON. WISCONSIN CONTENTS lacportiiMoe of Moil OTgant^iins 20 Cnusnl orgTRnisiniM of dampiug: off SHB^ 30 Host plnnift 34 Mot[?jiods «»f «oH troalnieiit 35 Fornialiu ex peri men tw 3J> .^ft.<4celtitnecu!!i treatmeat 4'j l'-.vi»f rhnc-uts imrter field oondItion.s 40 Cultural reniedies 51 1 iieekJiiK the OiHenKe 50 Siiminar J- 58 Literature cited 60 The Control of Damping-Oft In Plant Beds' JAMES JOHNSON Introduction The widespread occurrence and economic importance of soil organisms is at present becoming fully recognized. How to treat the soil in such a Avay as to destroy the undesirable soil organisms without injuring the productivity of the soil has be- come one of the great problems of the science of agriculture. The subject is complicated by the fact that, in addition to the harmful organisms, the soil contains a number of organisms which are desirable and frequently essential to the permanent productivity of soils. Most soils harbor fungi, insects, bacteria, and other lower forms of living organisms which are injurious to one plant or another. Many of these pests are obligate parasites and occur only M'here the host plant has been pre- viously grown. In such cases crop rotation is, in most cases, a long but sure remedy. Some of the organisms, however, in- eluding the common damping-of? fungi, are peculiar in that they are parasitic upon a wide range of hosts, besides being capable of existing saphrophytieally upon the vegetable matter in the soil. Crop rotation is hence practically valueless as a remedy against these fungi. The most noticeable occurrence of the damping-off disease is in plant beds and forcing houses, where large numbers of plants are grown closely together. The amount of injury arising from this intensive system of plant culture requires a rapid and efficient means for the control of the disease. The last two dec- ades have witnessed the recommendation of a number of meth- ' The writer is greatly indebted to Prof. L. R. Jones of the Department of Plant Pathology of this Station for many helpful suggestions re- ceived during the progress of this study. '^^ Wisconsin Research Bulletin 31 ods for the control of damping-off, many of which are reported in popular literature. The purpose of the experiments des- cribed in this paper has been largely to compare these methods. The work on this disease was at the outset primarily conducted for the purpose of con trolling damping-off of tobacco seedlings. Other plants were used in some cases to facilitate the work, but the principles apply equally well in a general way to all plants grown in seed beds aud subject to the disease. A large portion of the experiments have been conducted in the forcing house, and the results are applicable therefore not only to damping-otf in plant beds but also, in a large measure, to the disease in forcing houses. Tobacco growing in Wisconsin is ail industry Avhieh returns approximately $6,000,000 annually to the growers of the state. The damping off disease is of fre- quent occurence in the plant beds and causes a considerable loss in the destruction of seedlings. Indirectly, such losses result in delayed transplanting, transplanting of inferior plants, and sometimes in reduced acreage, making the control of the disease of considerable local importance. Experiments on the control of damping off by several investigators on various crops and soils have indicated some variations in results, possibly due to differences in soil. Sandy loam and clay loam soil containing considerable humus have been used in the experiments cited in this bulletin, and hence the results apply particularly to those types of soil. The studies have been mainly confined to damp- ing-off by Pythiuni deharyanum (Hesse), though the same dis- ease produced by Rliizoctonia has also been studied in this con- nection sufficiently well to warrant the assumption that from a ])ractical standpoint the couti'ol methods apply equally as well in both cases. r>(fore taking up tiie subject of control it is Avell to mention ])rief^y some of the characteristics of the casual organisms. The Caus.vl Organisms rijlhium deharjjainim (Hesse) was first described and named by Hesse in 1874 (12) at which time he pointed out the destruc- tiveness of this fungus as a parasite of 'seedlings. As early as 1881 De Bary (6) believed it generally present in garden soils of Europe and extended the list of host plants named by Hesse. , Ward (32) in England, found it common in the garden soils of that epuntry and studied its reproc^uction and relatioji to the Ct)NTROL OP Dam PING-OFF IN i^LANT BeDS 31 host tissue. More recently, the econoinic importance of this fungus lias been emphasized in the United States by Atkinson (1) as causing the damping off of seedlings. FIGURE I. TOBACCO PLANTS ATTACKED BY PYTHIUM DEBARYANUM. Various stages of the disease are shown. The attacks of this fungus are very largely confined to young plants. The fu.ngus may, however, under exceptionally favor- able weather conditions kill the embryo in the seed or grow up- on old plants. The zone of first attack is practically always confined to the hypocotyl, either just at, or below the surface of the soil (P'igure 1.) Under favorable conditions, however, the disease rapidly spreads np the stem of the plant and into the midrib of the leaves. The characteristic appearance of a seed 32. Wisconsin Research Bulletin 31 bed attacked by damping-off is a bending over and Avilting of the diseased plants. Pythium dcharyanum is quite readily distinguished in diseased areas from most other damping-off fungi. The coarse, non-septate, highly granular, irregularly branching mycelium (Figure II,) can be easily observed on the interior and exterior of the host tissue when bits of the freshly infected tissue are teased apart and examined under the microscope. Ordinarily, the diameter IP FIGURE IT. MICROSCOPIC CHARACTERS OF PYTHIUM DEBARYANUM. A, mycelium: b, conidia; c, oospores; d, germinating conidia. of the hyphae will be found to vary considerably, the older hyphae usually being the larger. The very young portions of hyphae are frequently finely granular, but become coarsely granular with age, while the oldest h^T^hae are usually emptied of their contents. The mycelium penetrates the cells of the young and succulent ti'^sucs readily, seeming particularly adapted to luissing through coll walls and obtaining nourish- ment from the interior of the cells. Miiereas in tissues damped off by Ehizoctonia the liypluic ;ii-c more generally found be- tween the cells. AVhere they pass through the cell wall the diameter is characteristically much reduced. Reproductive bodies are not ordinarily found in damped off plants, though they may be produced quite frequently in certain Control op Dami'ing-off in Plant Beds 33 species of plauts. For the study oi" reproductive bodies it is however, more couvenient to obtain the fungus in pure culture upon various media. Thaxter's potato hard agar in petri dishes yields an abundance of conidia when inoculated with Pythium debaryanum. ■ Oospores were found to be produced in great abundance upon sterile slices of green cucumber in petri dishes, or upon very young tobacco seedlings grown upon fil- ter paper in petri dishes and inoculated with Pythium soon af- . ter germination. Zoospore producing bodies were not obtained upon any of various media tested for that purpose. The im- portance of the reproductive bodies in spreading the disease, or in carrying the fungus over unfavorable periods, seems com- paritively unimportant, but this matter requires further study before anything definite can be said upon it. Khizoctonia was first described by l)e Condolle (7), in 1815, as a root destroying fungus. Since then it has been described as a root parasite upon a wide range of host plants in Europe. It was not until the early nineties that the fungus came to be recognized as a destructive parasite in field crops and plant beds in this country. Pammel (16) described Ehizoctonia as caus- ing a root-rot of beets in 1891, and Atkinson (2) mentioned it as causing a damping-off of cotton in 1892 in Alabama and of various seedlings in New York (1) in 1895. Duggar and Stewart (8) have treated in some detail the occurrence of this fungus upon a large number of hosts in this, country. The genus Rhizoctonia includes several forms of sterile fungi which are recognized by certain characteristics of the mycelium and the manner of growth in pure culture. (Figure III.) The young branches are more or less narrowed at their point of union with the parent hyplia and grow characteristically in a direction almost i3arallel to each other. A septum is also formed a short distance from the point of union with the par- ent hypha. The young hyphae are strongly vacuolated but this condition disappears in the older hyphae which are fre- quently yellowish in color. In pure culture on potato hard agar the growth is at first loose and white, but later becomes short, tufted, and brownish. Under the microscope these tufted growths are seen to consist of short cells, sharply con- stricted at the septa and irregularly branched. These short- ened cells are capable at times of functioning as conidia. In older cultures brownish sclorotia are formed which, when 34 Wisconsin Research Bulletin 31 placed under favorable conditions, are capable of producing a new mycelial growth, and hence serve to carry the fungus through unfavorable periods. In certain forms true spore*; are said to be produced. FIGURE III. MICROSCOPIC CHAR.4.CTERS OP RHIZDCTOXIA. A, young mycelium showing habit of branching; b, old, brown hyphae, usually empty; c, short tufted growth, characteristic in culture; d, cells resembling and functioning as conidia. Host Plants J'jjtldam dcbarydn 11)11 and. Rhizoctonia sp. are now known to attack a wide range of jtlanis' when favorable climatic condi- ' The following lists, thoug4i probably not complete, serve to illustrate the types of plants upon which these fungi are parasitic. Pythium de haryanuin has been described as parasitic upon white clover (Trif oleum repens), European millet (Panicum miliaceum). false flax (Camelina sativa), shepherds purse (Capsclla bursa pastoris), white pine (Pinus strobus), beet (Beta vulgaris), orchid (Stanhopea saceata), egg plant (Solanum melogcna), spurry (Spergula arvcnsis), corn, (Zca mays), cress (Lepidium sativum), pigweed (Amaranthus rctroflcxis), white mustard (Brassica alba), cucumber (Cucumis sativus) and certain species of Viscaria, Lobelia and Gillia. Rhizoctonia has been described by Duggar (8) and others as parasitic in this country upon bean fPhaseolus vulgaris), beet, (Beta vulgaris), carrot, (Daucus earota). celery (Apium gravcolens), cotton (Gossypmm herbaccum), lettuce (Lactuca saliva), potato (Solanum tuberosum), rad- ish (Raphanus sativus), rhubarb (Rheum rhapbnticum), ornamental as- C'oKtkol of I)a.mpi.\(j-okf j.\ i'LANT Beds 35 tioiis for tlu'ir gi-owtli occur. Some i)lcUits are, however, very iiiuch iiioi'c siispeetible than others, owing to certain physiologi- cal oi- luorphologieal factors. The amount of injury also de- pends markedly upon the manner of cultivation and the stage of development of the plants. During my studies upon damping-oft' of seedlings I have ob- served Fylliium (Ifharyanum as occurring upon cress (Lepidium satii'um), tobacco (Nicotiana tahacum), lettuce (Laduca scitiva) , endive (Cichorium cndiva) carnation cuttings (Dianthus caryo- phyllus), cucumber (Cuciimis sativus), turnips (Brassica rapa), cabbage (Brassica olcracea), squash (Cucurhita maxima), cauli- flower (Brassica olcracea var. hotrytis)', pumpkin (CurcurbUa pcpo), mangle-wurzel (Beta vulgaris var macrorkiza) , balsam (Impaficns halsamina), kohlrabi (Brassica olcracea var. caido- rapa), tomato (Lycopcrsicum csculcntum), beans (Pliaseolus vidgaris) and geranium cuttings (Prlargonvum liortorum) . Rhizocionia has been observed to occur parasitically at Madi- son upon cress (Lcpidlum sativum), tobacco (Nicotiana taha- cuni)j lettuce (Lactuca sativa), clover (Trifoleum repens), sun- flower (Helianthus annuvs), radish (Raphamis sativa), egg plant (Solanum melogcna), pea (Pisiim sativum), celery (Apium gravcolens), tomato (Lycopcrsicum esculentum), violet (Viola affinis), beet (Beta vulgaris), geranium cuttings (Pclargo}tiu)n horlorum) and sweet Willam (Dianthus harhatus). Historical The practice of heating soil by surface firing was probably practiced long before its actual sterilizing value was known. The benefits known to be derived "were largely those of an im- proved i)hysical condition of certain soils, the added fertilizer from the ashes of the burned material, and a reduction of weeds in the soil. Atkinson (1) in 1895 pointed out several factors Avhich intlu enced damping-oflf by Pytliium in cutting-beds, suggested sev- paragus, (Asparagus sprengeri). China aster fCallistephus hortensis), carnation (DinntMis caryaphyUus), sweet William (Dianthus barbatus). violet (Viola odorata). lambs quarters (Chenopod\iim album), tumble weed (Amai-(inthus albus). pigweed (Amaranfhus rctroflexus). white pine, cucumber, begonia, coleus, verbena, hydrangea, hardy candytuft, mammoth sage, phlox, pyrethum, snap dragon, raspberry, squash, clover, lucerne and peas. 36 Wisconsin IIesearch 13ulletin 31 eral cultural means for its coutrol, and recommended steam sterilization in severe eases. Stone (29) in 1900 emi)hasi/ed the value of partial steriliza- tion of forcing house soils, and its i^raotical applicatio]i for pro- tection against the lettlice drop disease. Selby (23) in 1906 recommended the use of 1-160 to 1-200 for- malin (2-2yo pints to 50 gallons of water) for the coutrol of bed rot of tobacco i)lanTs caused by Khizoctonia. His experi- ments using this strength fornmlin gave partial control of the disease. Poor results, hoAvever, were ol)tained in some in- stances. Jones (13) in 1906 working in Vermont, found that the use of 1-100 formalin reduced damping-off in pine seedlings from 90 per cent to TVo per cent, and 1-200 formalin reduced the dis- ease to 9 per cent. Spaulding (26) in 1908 recommended the use of sulphuric acid (1 ounce to one gallon) against damping-off of coniferous seedlings, and obtained poor results with 1-100 formalin. Gilbert (11) in 190!) using various methods of soil treatment for the control of root rot of tobacco (Thielavia hasicola), rec- ommended steam sterilization. He obtained poor results with 1-100 formalin in most cases. Scherffius (21) in 1911 working in the Transvaal, used five different methods of soil ti-eatment against damping-off in to- bacco plant beds, viz., (1) open lire; (2) boiling water; (3) steaming; (4) roasting; (5) formalin. All factors considered, he obtained the best results with the open fire method. Roast- ing gave almost as good results as the first and was closely fol- lowed by steaming. The formalin treatment w-as rated as fourth, and boiling water gave results but little better than un- treated plots. Russell and Petherbridge (21), in 1912, used carbon bisulphide and toluol on greenhouse soils and reported marked increase in plant growth due to the elTcrt o!i soil organisms, but did not state definitely the effect on lii<- damping-off fungi. Giff:'ord (10) in 1912 compai'cd steam sterilization and forma- lin treatment for the prevention of damping-off of coniferous seedlings by Fusarium in jiarticulai-, and reconunends the use of one per cent forinalin for its control. Various other stalements are to be fouiul in litei-ature recom- mending the use of potassium sulphide, iiotassium ])ermanga- CuNTKUL OK I).\M1'1NU-0FF IN PLANT BeDS 37 nate, lime, Bordeaux niixturt', and potassium nitrate for the control of the damping-ofil: diseases. The status of tlie problem is such, following the work of the preceding investigators, that the subject appeared to need fur- ther study, especially along the following lines: — 1. The strength of lormalin reciuircd for complete control of damping-off as with formalin 1-100 or weaker, previous investi- gato)-s secured contradictory results, only partial control being obtained in most cases. 2. The relative value of other fimgicides sometimes recom- mended for control. 3. The value and jji-acticability of steam sterilization of soils in the field. Preventives Experimental Methods The experimental data which is at hand upon the efficiency of various methods of soil sterilization against damping-off in plant beds is small and contradictory. This is due largely to the fact that the experiments have been carried on principally out of doors, where, owning to the influences of weather condi- tions, the data may in some instances be misleading. Further- more, the control methods have been directed against the damp- ing-off fungi as a whole which may again lead to complica- tions from a scientific staiidpoint, though it may appear rela- tively unimportant fi'om tho standpoint of practice. Reinfec- tion is also, of course, an important consideration and must be considered at all iiuK's where attempts are made to eliminate certain organisms from large exposed areas out of doors. In the following experiments it has been attempted, as far as possible, to eliminate these factors, and to furnish conditions most favorable for the development of the disease. In a num- ber of preliminary experiments begun in 1909 on forcing house benches, results were obtained which were indicative, but not entirely reliable because of changing atmospheric conditions and occasional reinfections. The results of these experiments will not, therefore, be given in detail. They served to show, however, that sprinkling the plants with dilute solutions of potassium sulphide, formalin, copper sulphate, ferrous sul- 3^ Wl.SCUNSlX Ri-:8EARC1J BiLLETIN 31 phatfc, and aimnoiiia at strengths below tliat toxic to plants was inefificient in checking the disease after it had once started. Preliminary experiments in the fall of 1910 indicated that a satisfactory tempei'atiiiv and liuiiiidity for damping-otit' could he maintained in a small damp chamber, and that the possibility of reinfections tlirough the air was practically nil. A large damp chamber IS feet long and ;5.5 feet wide was then con- structed on one of the forcing house benches. (Figure IV.) The top was made of hot-bed sash placed about 2.5 feet above the bottom of the bench. Tlie sides of the chamber wi-re also partly glass, allowing sufficient light to enter from Ihc sides so as not to cause the seedlings to grow spindly. The unit area used for soil treatment in this chamber consisted of a forc- ing liouse flat Avliicli measured 21 inches by IG inches aiid o inches in depth, holding 28 to 30 pounds of soil. This chandjer 1 1 olds 23 flats of this size. In earlier experiments the control was aimed at both Pythium and Rhizoctonia in the same soil, and no inoculations were made as all the soil used was found to contain these fungi naturally. It was obsei'ved that under certain conditions, i)ar- ticularly of temperature, Phizoctonia would be the nmin casual organism of damping-off, while uiuler other-conditions Pythium seemed to ])7'edominate. It ^vas also thought ]iossi])l(' that a control measure might eflfectively restrain one organism and not tlie other. This complicated uuitters to such an extent that in tlie latci' cxpci'inM'iits the soil and flats were all sterilized, and inoculations made in pure culture in the following manner: The flats wei'e filled Avith a soil composed of three pai'ts of compost, two pai'ts of garden loam, and one ])ai't of sand. They were then set one on top of the other, with one fourth inch space between each, and a large, tight box inverted over ilicm. Steam Avas rnn u'lder tliis box from the forcing hou'-<^ heating system for one and one-half hours, which was suffi- cient to sterilize the soil so far as the damping-off fungi were concm-ned. ]>y this method a temjx'ratni'e of pi'actically 100 degr"es Centigi-ade was maintained for one hour. The soil and flats, while still warm, were placed in the damp chamber,, whicli had l)een ]H'e\ iously flushed out with water and sprayed with a strong solution of copper sulphate. About six days ])revious to this, pure cultures of Pythium ami Ivhizoctonia Averc- started on corn meal. After the flats had C(>NTK(JI. OF I).\.MlMX(i-OKK IX PLANT JiKDS 39 cooled sufticiciitl.N , 1h<,' coi-n inc.il cultures were spread in small pieces ovoi- the sui-fMCcs of the soil. The flats Avei-e then cov- ered witli i)late glass which had been fumigated Avith formalde- hyde gas over nighl. After foui- or five days the fungus had grown i)r')fusely into the soil from the corn meal. The soil was now worked up in order to obtain as even a distribution of the fungus through it as possible. The fiats were then ready for treatment with the fungicides. In the experiments described "sterile" checks are fiats which were left uninoculated, and "untreated" checks are those which were inoculated with the disease, but not treated with any fungicide. The checks received the same amount of w^ater as the treated flats. The amount of liquid fungicide used on all flats was 2500 cubic centimeters which was sufficient to saturate the soil to the bottom. After treatment the flats were again covered for 48 hours v.'ith the glass ])lates, which reduced evaporation to a uiinimum, and gave the volatile fungicides, especially, time to act. The covers were then removed and placed vertically betAveen the flats preventing contamination from one flat to another. After four to six days the flats were sufficiently dried out to work the soil again, after which the seed was sown. The fempei-ature of the chamber varied from 23 to 30 degrees Centigrade. Tn the earlier experiments tobacco plants w^ere used, but OAving to the length of time required for germination and growth, garden cress (Lcpidium sativum) was substituted in the later trials. This plant serves admirably for this purpose because of the readiness with Avhich it damps off and the short time required for its germination and growth. Cress is to be looked upon as a sensitive indicator of the presence of any living damping-oflp fungus and as a measure of the efficiency of the fungicide under conditions favorable to the fungus. Experiments with Formalin The tj'ials made in treating the soil Avith formalin at various strengths Averc carried on in connection Avith the other treat- ments for comparative efficiency. On account of the import- ance of the formalin treatnunt, hoAvever, the flats treated Avith this chemical, Avith their checks, have been grouped into Table T. Formalin (40 per cent formaldehyde) of the best purity Avas 40 Wisconsin Research Bulletin 31 used in the exporiments. The experiment was carried on in four different series. In Series I and II, the casual organisms were both Pythium and Rhizoctonia, with tobacco seedlings used as host plants. In Series III and IV Pythium only was present in the soil, and cress w^is used as the host i)]ant. In Series III, only the per cent of plants killed above ground Avas taken into consideration, while in Series IV the per cent of plants killed both above and below ground was considered. The formalin strengths used are given in parts by volume. Table I. Effect of Formalin Soil Treatment t)N Dami'ing off PKECENT OF PLANTS DISEASED Series I'n- treated soil Strength of Formalin t'sed sterile 1-200 1-150 1-100 l-7b 1-50 soil I .-)n (to 7-") 10 10 90 SO 75 0 0 ir SO 10 20 20 SO 10 40 65 0 0 Ill* v, 20 X.') 90 75 0 0 ■ 0 0 0 0 IV 95 fi5 75 35 40 0 0 *Api)areiitly low per cent of disease in untreated soil is due to fact that the larsre per cent of plant,s Icllled Ijelow the surface were not included in this estimate. It will be seen IVoiii this dat;i th;it rornialiii treatment at the strength recoiiiinriKlcd by various authors, i. e. 1-100 to 1-200 is practically no better than no treatment for dami)iiig otf under the conditions of these experiments. Series II shows that 1-100 formalin will not control Rhizoctonia, ns mici-oscojjical exami- nation showed this fungus to be the main cause of the damping- off in tills cnsr. The wide runge of results in Series II is due largely to Hie Ihitv being affected by different temperatures, a factor Avhich was better controlled in the later experiments. In Series III find IV th-- flats treated contained only Pythium, which was not destroyed by 1-75 formalin and lesser strengths, but was completely controlled by 1-50 formalin (Figure V). It is possilile thnt a strength somewhat less than used to control P;^- vi° 3 rr " &3 5 O 2. '^ Hi w 5' X CONTliOL OF DaMI'ING-OFF IN PLANT BEDS 41 lliiuiii will dcsti'oy Uic inytM'liLiiu oL' tlie sterile fungus, Rhizoc- toiiia, but thai is immaterial I'roin a practical standpoint, as a strength sufdeient to destroy both fungi must be used in order to be certain of results. A strength of 1-50 formalin may be slightly in exeess ol' what is actually i'e(|uired to kill Pythium, but it it a coitimon pi-aetice in using fungicides to work with strengths somewhat beyond tliat which is actually required, provided no injury to the host occurs. Selby (23) recom- mends the use of 2-2 J/^ pints of fonnalin to 50 gallons of water, (1-160 to 1-200,) one gallon oi the solution per square foot of seed bed, applying it at intervals of a few hours, as most soils will not take up this amount of water at one time. According to the writer's observations, two quarts of the 1-50 solution applied to each square foot is sufficient, especially if the bed is covered with some material to hold iil the fumes. This is on the principle that a strong solution acting for a shorter time is more efficient than a weaker one acting for a longer time. The stronger formalin will also penetrate down- ward into the soil further, and the soil can be worked sooner than when twice the amount of solution is added. The prac- tice should then be to use four quarts of formalin per 50 gal- lons (one barrel) of water, applying at the rate of two quarts per square foot, or approximately one barrel to one rod of seed bed six feet in width. Ari interesting observation of the effect of formaldehyde up- on the fungus has arisen in this connection. If we separate the amount of injury caused by Pythium on cress seedlings in- to the percentage of disease above ground and percentage of disease below ground, we tind that the percentage of seeds killed before germination in the untreated checks is very high. Table IT illustrates the efTect of formalin of different strengths upon the relative rate of development of the fungus. Table II. Influence of Formalin on Dami'ing-Off Fungus Troatnunit Strength Percent disease lielow surface Percent disease :il)Ove surface Untreated check. .. xo 0 0 0 0 0 10 Formalin... i-ioo 1-100 1— 7.i 1 — 50 93 Formalin Formalin 40 Formalin 0 Steam sterile '.. .. 0 42 Wisconsin Research Billetin 31 According 1o this data, ]-150 foriualiu ctjfectively prevents the fungus ij-oni attacking the cress seed in the soil. This has also been observed to be the case in 1-200 formalin. On the soil ti-ealctl with l-JilO formalin, tlu' i)lants damped off rapidly, however, soon after they appeared above ground. The l)ercentage of dauiping-otf up to a certain time decreased with the increase of strength of formalin, the weaker strengths of formalin (l-7o to 1-200) having a checking influence upon the fungus which disappears in time. This may explain the fact that relatively good residts have been obtained with the lower strength of formalin. If the climatic conditions favorable for the development of the fungus come on a time before the checking influence of the formalin has passed off, good results may be obtained, but if these favorable conditions appear some time later, the treatmcnit with the foi"malin at the lower strengths may be practically valueless. This checking influ- enee of formalin below 1-75 will usually pass oft' in from 10 to 15 days after treatment, and is '-oiisequently of little value, es- pecially to those plants requii-ing longer i)eriods in the plant beds. There can be but little doubt that 1-50 formalin kills the fungus as the treated soil remains apparently sterile for week?, i)roviding infection from the outside does not occur. The partial control obtained by Si'lby and Gift'ord, using for- malin at strengths of l-lOO and b.elow, can be explained by this checking influence of the fungicide together with unfavorable weather conditions for the i)arasite after the foi-malin had left tlie soil. Miscellaneous Soil Treatments Dui'ing the course of the experiments with formalin, sev- eral other fungicides Avere tried for the prevention of damping- off. The strength of the fungicides used, especially that of the non-volatile ones, is limited by the toxicity to the plants. Sev- eral preliminary experiments were carried on in order to de- termine the percentage strength of the fungicide which could be used Avithout too much injury to the plants. These pre- liminary experiments indicated to some extent the uselessness of the application of many of these substances, but further trials were continued in connection with the formalin controls Control ok DAMi'isd-oi-K ix Plant Tinns 43 and sonic data oldainiMl, wliitdi is ^roiiix-d in tlic following table. Taui.k III. Pkr Cknt ok Dise.^skh Plants in Misceli-aneoih Soir, Treat.uents Experiment I. Experiment 2 Treatment Per cent h.v weight Treat- ed Un- treated check ."Sterile check Treated Un- treated check Sterile clieck o.s 1. 0.8 1. 0.12 0.2 0 8 1. 0 4* 0.5* 1. 0..J 0.12 1.3 20 0 I'otassium sulphide 95 70 85 85 0 Liniesulphur+ :. iS 20 15 0 0 0 Copper sulphate 5 90 0 Coppei' acetate 18 20 0 Copper acetate 35 85 6 Coi)per nitrate... 20 10 20 20 20 20 0 0 0 '25 Hordeaux mixture* Sulphuric acid' 0 Sulphuric acid 65 92 85 85 0 Fotassiuui permanga- nate 0 Potassium nitrate 10 60 20 20 20 6 0 0 Mercuric chloride Hot water 85 85 0 *l'er cent by volume. +Sherwiii Williams Lime Sulphur Solution, sp gr. 1.2491. J5— 5— 50 formula. S.-ip. gr, 1.84. A comparison of the percentage of disease in treated and un- treated soil in Table III sho-\vs little or no benefit from the application of these substances as fungicides to the soil in amounts which will permit of seed germination and plant growth. These figures do not, however, indicate the true state of affairs in all cases. Certain of the chemical agents appeared to produce some inhibitory action upon the development of the fungus for a short time after the application, but this action apparently disap- peared too soon to be of any value. This inhibitory action on the fungus was most conspicuous as effecting the percentage of seedlings killed below the surface of the ground. In the case, for instance, of treatment with 2% lime sulphur fully 75 per cent of the plants appeared above ground before damping-off, as com- pared with only 5 to 10 per cent in the case of untreated soil. Treatment of a steam sterile soil with lime sulphur (2 per cent) showed retarded germination and growth of cress, but a perfect stand. Similar iiiliilutory action was also observed to occur in the potassium sulpiiide, copper sulphate, and mercuric chloride treatments. The inhibitory action of these fungicides upon 44 Wisconsin Research Bulletin 31 fungus development was correlated with an inhibition of plant growth, and consequently little hope for their value as soil fungi- cides is maintained. The essentials of a good soil fungicide ap- pear to be that it must first possess strong fungicidal action and secondly that it must be volatile in order that it may be freed from soil before the seed is sown. Covering the soil with screenings of sphagnum moss gave in- dications of checking the disease to some extent, and is worthy of trial on a small scale. Sand covers gave poor results in some cases and are of doubtful value for checking the disease. Steam sterile surface soil two inches in depth, allowed to re- main on the surface soil from 6 to 7 days before the sowing of the seed, appeared to be of no value. A three inch layer of ster- ile surface soil, however, showed very little disease except that which was later determined to have arisen from outside infec- tion. Under practical conditions a three to four inch layer of sterile surface soil should in most cases effectively prevent damping-off, at least during tbe period when the seedlings are most susceptible. The use of such substances as lime, ammonia, toluol and carbon bisulphide have also given negative results in the control of damping-oft" so far as they have been tried in these experiments. Sterilization by Heat Several methods have been devised by various experimenters for the sterilization of soil by heat. The writer found that forc- ing house fiats such as those used in the experiments could be cheaply and efficiently sterilized under an inverted box into which streaming steam was run for one and one-half hours from the forcing house heating system. Fourteen flats were sterilized at a time, or an equivalent of 28 to 30 square feet of soil. This method, though impracticable out of doors, is a good one for a small greenhouse where seedlings and cuttings are started in flats. The methods used for sterilizing soil by heat, which lend themselves to out door seed bed conditions, fall undci" three heads, (1) steaming, (2) surface firing, (3) roasting. The most practical method yet devised for the stci-ilization of &eed beds by steam is the "inverted pan method". This method was first used by Shainel for stoi-ilizing nematode infested soils C'oNTltUL OF Dam PING-OFF IN 1*LANT BkDS 45 in Florida. A galvanized iron pan, six feet by ten feet and six inches deep is inverted over the area to be sterilized after it has been prepard for seeding. As the edges of the pan are sharp they can be pressed into the soil an inch or more, thus fomiing a tight compartment under the pan into which the steam is run 30 — GO minutes from a boiler at a pressure of 80 — 150 pounds. The time of steaming depends largely upon the type of soil and its moisture content and compactness. Loose sandy, moist but FIGURE Vr. STKRILIZIXG SEED BEDS BY THE '-IXVERTED PAN" METHOD. not wet soils, are more easily and rapidly steamed than heavy and wet soils. The pan used at this station (Figure VI) in tobacco seed-bed experiments Avas made 8 inches in depth, which is prob- ably more desirable in some cases. The pan could be made 12 feet in length without loosing any efificiency where large boilers are used. The handles should preferably be placed on the sides instead of on the ends so that the pan could be moved from one section of the bed to another without the operators walking on the sterilized soil. The weight of such a pan is approximately 400 pounds. A one inch steam hose should be used to connect it with the boiler. A traction engine such as is used for thresh- ing is most convenient to furnish the steam. 46 "Wisconsin Keseakcii Bulletin 31 Other methods of steaming are the ' ' sub-surface pipe ' ' system and the ' ' steam rake ' ' method. The former consists of an under- ground system of li/o inch perforated pipes, 18 inches apart, running lengthwise of the bed, G to 8 inches below the surface. The perforations, about Vi iuch in size and 6 inches apart, should be on the under side of the pipe. Steam is run into these pipes at a pressure of 80 to 100 pounds for a period of about 1 to 2 hours. The method is practicable oidy when beds are located for several years in the same place. The steam rake or harrow method consists in forcing steam into the soil through hollow perforated pipes resembling the "spikes" of a harrow. The results are the same, but the method is usually regarded as less efficient than those previously mentioned. Surface firing has been quite popular in the past, but it is giving way to other methods, owing largely to the scarcity of combustible material in many places, and to the labor connected with the process. The method consists simply in producing a hot fire for an hour or more over the section of the bed to be sterilized after it has been fertilized and fitted for seeding. Any combustible material may be used such as brush, straw, or dry wood. Roasting or "pan-firing'' requires tlie removal of part of the soil from the seed bed into the pan and returning it after being sterilized. The pan may be simply a large piece of sheet iron, resting upon supports, underneath which is a fii-e. This i)an should be set on the bed and the soil to a depth of 6 inches on one side shoveled upon it and roasted for an lionr, being care- ful not to let it become so dry that the vegetable matter is burned out. Then a like area on the other side of the bed is sterilized and ill the meantime the soil uiidcnirath the pan has been suffi- ciently lieated, and tlie pan may be moved to anotlier section of the seed-])ed and the operation repeated. Experiments under Field Conditions In the spring of 1912 the "inverted pan" method, surface firing, and formalin treatment (1-100 solution and two quarts per square foot) were compared on tobacco beds. Steam was I'un into the inverted pan for various lengths of time ranging from 25 to 45 minutes, at a pressure varying from 100 pounds down to 25 pounds in some cases at the end of the application. Control ok I) am ping-off in Plant Beds As IK) clainping-oft' occurred in tlic uiisti'i-ili/.cd plots, tlic com- parative value of the different periods of heating could not be determined, and the experiment will not be given in detail. Some interesting results were obtained, however, upon the effect on weeds and growth of the plants. The weed seeds were all killed at as low an exposure as 25 minutes, starting at 80 pounds and running down to 25 pounds pressure. (Figure IX,) The plants in tlic uiisterili/ed plots were not one-half as large as tliosc in the sterilized plots one month after sowing the seed. 3 ^->^.< .r/< v>MS.^^' >^'^:^/^' FIGURE VII. STEAMING NOT ONLY PREVENTS DISEASE BUT ALSO KILLS WEEDS. A, Soil steamed. B, Soil not steamed. (Figure YIII.) Tlie cost of one weeding of the untreated plots was equal to the cost of steaming an equal area. This together with an assurance against bed-rot and other pests, and an increased growth in the plants, made the process an economical one and well wortliy of use on all tobacco farms. The sterilizing of tobacco vseed beds by the "inverted pan method" can be most cheaply done by a few neighboring farmers in cooperation, or better still by a person who owns a steam trac- tion engine who will do the sterilizing by contract at so much per rod of seed bed. The surface fired plot was second best, being much better than the untreated plots, and no weeds were present. The for- malin treated plots came third, the principal effect being a con- siderable reduction in the number of weeds, although this effect 4:8 Wisconsin Research Bulletin 31 is not usually claimed for it. Considerable stimulation in the growth of the plants was also noticed. The practical value of steaming tobacco beds was further demonstrated in the spring of 1913, when the writer steamed approximately ten thousand square feet of soil for ditferent growers in Rock County, Wisconsin. The results were so mark- edly successful in destroying weeds and producing ^nrlicr and l)etter plants, that these and other growers who saw the uprno'^- strations are preparing to steam all their seed beds hereafter. The results of field experiments have shown on the whole that it probably does not pay to steam for damping-off alone, since its occurrence is comparatively infrequent. The other important re- sults of soil steaming however, sum as increasing the growth and uniformity of the plants, killing the weed seeds and injuri- ous insects which may be in the soil, and preventing infection with other diseases make it an economical, if not a necessary i:)rocess. Secondary Prefects of Soil Treatments The treatment of soils by heat and liquid fungicides to des- troy the undesirable organisms results in a number of secondary effects which have been the subject of some study in recent years. These effects are primarily perceptible upon the plants grown in the treated soils. It has been found that the physical, clicmi- cal, and biological nature of the soil is altered by heating. Tlie following tabulation will illustrate these effects upon the germination and growth of seedlings as obseiwed fi'om flats treated by heat and by various chemicals used at the strengths mentioned. The unit area used for treatment was tliat of a forcing house fiat holding from 28-30 pounds of soil. Twenty-five hundred cubic centimeters of the solntion was ai)|)li(Ml to each flat in every case. As we are interested in tlie secondary effect of sterilization by heat and formalin principally, we will consider these facts briefly. The first effect that can ])e noticed in a heat sterilized soil is a change of physical condition which icsults in a more rapid drying out of the surface layer of soil. According to Lyon and Bizzell (15,) this appears to be due to a loosening of soil, which aids more rapid percolation downward. This effect cannot be said to he particularly disadvantageous except when CoNTRor. OF ])ami'ing-off in Plant Beds 49 the seedlings are very young and need moisture at the surface, when it is readily overcome by two or three additional water- ings. Taii[;I': IV. IOfkects of Soih Tkhat.mknts on Plants Treatment Per cent hy weight Germination Growth after 15 days Steaminff 100° ('. 14 lirs Formalin. 1 -JO i.sowiiit,'' after 3 flays) Formalin. 1 .")0 (.sowinj; after 7 (lays) Noetfpct Ret arded Copper siilpliate 1.2 0.8 0.12 1. 1. 1.* 0..J* 1. 0..J 1. No effect Slight inhibition Coppei- nitrate No effect Inhiliited Ammonium ropper carlionale lMarl»edlr letai-ded. Slight retardation.. Marked retardation Ketardt»d No effect Slight retardation... No effect Inhibited Slight inhiljition No effect Potassium suli)lilde Lime sulphur sp. trr. 1.2491 Sulphuric acid c. p I'o assium permanganate No effect Slight inhibition Potassium nitrate No effect I'"'errous sulphate No effect. No effect Hot waiei- No effect Mercuric chloride .12 Retarded Surface firing- * Percent In' volume. Gifford (10), however, reports that in the case of coniferous seedlings the loss of moisture in the upper layers of soil as a result of steam sterilization is a disadvantage of sufficient im- portance to prevent its general adoption against damping-otf. Pfeiffer and Franke (19), on steaming soil for three hours at a pressure of one atmosphere, found that the heated soil pro- duced a crop of greater dry weight and higher nitrogen content than the unheated. Schulze (25) noticed that the sterilization produced injurious effects upon plant growth in the early stages of development, but that as the plants grew older, they grew more vigorously in the sterilized soils. Pickering (18) found that heating moist soil to temperatures from 60 degrees to 150 degrees Centigrade, retarded germina- tion of seeds, and in some cases decreased the total number of seeds germinating. He also obtained increased growth of plants in heated soil, which was attributed to an increase in soluble nitrogen. Koch and Luken (14) heated sandy soil with steam for two houi*s at two atmospheres pressure, and found that the soluble plant food was increased. Tliey also found that the immediate 50 Wisconsin Research Bllletin 31 results of heating were injurious, but that it Avas unimportant and that tlic later yield was improved. Russell and Hutchinson (2*0) heated soils to 98 degrees C. and found that an increased production of ammonia followed for at least 150 days. This increase in ammonification is attrib- uted to an increase in the number of ammonifying bacteria iii the soil as a result of the destruction of the larger organisms, especially protozoa, Avhich destroy the ammonifying bacteria, and are hence injurious to the productiveness of soils. Lyon and Bizzell (15) steamed different soils under a pressure of two atmospheres for periods of two and four hours and found a marked injurious effect resulting, as shown by the growth of plants in the soil. The time required for the soils to recover from the treatment was usually in the same order as their relative productiveness. These experiments indicate that the injurious effect varies with the amount of steaming and the type of soil. These results are not quite applicable to partial soil sterilization against damping-off since the amount of steaming is much less and the recovery consequently more rapid in the ordinary soil treatment. Russell and Petherbi'idge (21), using carlxm bisulphide, toluol, and steam sterilization, report a retardation in germination, but a greatly increased yield as a result of sterilization of green- • house soils. These results, as a whole, are in accoi'd with tlie writer's ob- servations of the effect of steam treated soil ui)on seedlings of cress and tobacco. The very marked increase in gro^^i^h of tobacco plants on heated soil makes it appear that this practice sliQuld be generally adopted hy toliacco growers. It is recom- mended particularly to those growers who have yearly difficulty in obtaining an early stand of vigorous plants. The heating of soil does not appear to affect all soils similarly. According to Stone (30), soils low in oi-ganic matter may be injured by heat- ing, and this fact should be considered before preparing to heat certain soils. In the writers experience with nearly a score of Wisconsin soils, licating has never resulted injuriously. The effect of formalin sterilization upon soils has not been the subject of as much experimentation, and the real nature of the effect has not, to the writer's knowledge, been determined. If the seed is sown before most of the fumes of formaldehyde have • (.loiNTiioi; or l)A.\ii-iN(i-()Ki' IN Plant 1',i:ds 51 left tlu' ground, gt'fniinatioii is rotardod. If the soil is allowed to dry out for a week, however, this eiTect disappears. It is interesting to note, also, an increased vigor and growth of seed- lings grown upon formalin sterilized soil. This effect with the strengths of formalin used in these experiments occurs usually sooner, but not to as great an extent as in the case of steam sterilization. This increase is again possibly due to an effect upon the chemical nature of the soil or, more probably, to the ett'ect upon its bacteriological content. It has also been su])posed that tlie plants may take up small amounts of formaldehyde directly, resulting in a stimulation of growth. The amount of weeds was also much reduced in formalin treated soil, an effect also noted by Clinton (3) in treating soil for root-rot of tobacco. Cultural Remedies Although prevention by such measures as have been described is the only certain protection against damping-off, much can be done in a cultural way towards obtaining a relatively small loss from the disease. Infecird sails;. As an increased amount of the causal organ- ism in the soil will naturally be present in a soil where damping- off has previously occurred, later seedlings in the same soil "will be more likely to damp off when the weather conditions become favorable to the fungus. It is, therefore, a wise precaution to avoid soAving seed in the same beds where the disease has pre- viously occured, unless the soil is sterilized by heat or formalin. It lias been found possil)le by repeated sowings of cress in dis- eased soil to fill it so com])lptely with the causal organisms that ]iractically every seed in the later sowings is killed before it germinates. A charge of seed bed is, therefore, desirable and necessary in case of bad infestations. New ground, especially fall plowed, old blue grass sod has shown itself in many cases to contain very little, if any, of the damping-off fungi, besides being relatively free from weed seeds, thus making it a good seed bed if the soil is otherwise of the right nature. Thiclx sowing. Although the damping-off fungi are capable of attacking plants standing singly under especially favorable weather conditions, no method of culture is more conducive to damping-off in the seed beds than thick sowing of seed, with the resultant crowding of plants. The following table illu§- 52 Wisconsin Kksk.vrcii Bulletin 31 trates the relation between the amount of seed sown and the ])ci-f'('ii1air(: area danijicc' off. Tlie experiments were cni'ried on with tobacco ill Hats in the greenhouse where favorable condi- tions for (]ami)ing-off were maintained. Table Xo. \. Kffk( r ok Thick Sowing ox Pkk( kxtagf, of Diskaskd Plants WEIGHT OF SEED SOWX. Plants 1 Per flat. (iiams. 0.1 ().■> 0.3 0.4 O.o 0.6 0.7 0.8 0.0 • 1.0 Per 100 sq. ft. Ouni'ps. O.lfi 0.3.S 0.49 O.Rfi 0.X3 0.99 1.16 1.33 1.49 l.fiO cii>eaM (1. Per cent. 0 2 0 3 4 X V, 3.0 6 80 X 0 SO 10 m The first signs of the disease occured in flat No. 10 very soon after the plants had obtained their first two leaves. The dis- ease started last in fiat No. 3, but not until the plants were of eonsiderat)]e size and were crowded. This record illustrates quite markedly that the amount of damping-off varies directly witli tlie thickness of sowing. (Figure X). This is due to sev- eral factors, the first of which is probabl\- tlic increased humidity cf the atmosphere around the base of plants when crowded, as a result of lessened air circulation and increased shading. Crowd- ing also affords easier transfer of the fungus from one plant to another and is instrumental in producing more rapidly grow- ing and succulent stems, which are more susceptible to attack. Thick sowing is especially likely to occur in sowing small seed. One ounce of tobacco seed witli a germinating capacity of 90 to 95* per cent sliould be made to cover 5 to G rods of seed bed six feet in widtli, whereas in practice it is frequently sown on two rods. Soil Types. Certain types of soil are particularly favorable for the damping-cff fungi, and should therefore be avoided. Such are the soils which contain a high percentage of undecom- posed vegetable matter, and those which are poorly drained. Tn an experiment using mixtures of varying percentages of manure, IXCKKASKS VMiOK I'LA.NTS. KKilKK IX. STEAMING ELIMINATES THE COST OF WEEDING, a, stenm sterilized; b, no treatment. FIGURE X. THICKNESS OF SOWING FAVORS DAMPING OFF. Figures on boxes show amount of seed In grains, sown in eacb flat. CoKTUUI; of l).\.\ll'IMj-UKF IN i'LANT liKDS 53 compost, garden loam and sand, sterilized and inoculated with rythiiim, the most rapid spread and prolific growth of the fungus was obtained in a mixture of 50 per cent manure and 50 per cent garden loam, and the least growth in garden loam and pure sand. As the fungus lives oji the organic matter of the soil in the absence of tlie host plant, it is reasonable to suppose that the fungus content of soils rieli in organic matter is greater than that in soils poor in organic matter. This is one reason why a layer of pure sand on the surface of the bed should afford a considerable protection against damping-off. Yet experiments show that the application of a layer of sand does not prevent damping-off'. This may be partially explained as follows: The damping-Oft* fungi attack a plant just at or above the surface of the soil probably because the fungus is aerobic. Therefore in a perfectly aerated soil the disease would be more likely to occur further down on the hypocotyl. A layer of sand or moss, how- ever, still allows considerable aeration down to the surface of the soil proper and hence permits the fungous development and consequent disease below the artificially applied layer. Some ad- vantage in sand and moss, however, is undoubtedly supplied by its mulching qualities, on account of wdiich the application of less water to the beds is reciuired. A lower average percentage of moisture in the surface soil is thus maintained. Soil Moisture. The following experiment was planned for the purpose of determining the relation of the development of Pythium in the soil to the moisture content of the soil. Dry sand was graded into various sizes by means of 20, 40, 80, and 100 mesh sieves. Ten grams of each grade was put into each of six test tubes. To each tube was then added enough dilute nutrient solution to make up to the calculated percentages of moisture given in Tal)le VI. The tubes were then sterilized and inoculated with Pythium. After 4 days when considerable growth had been made in most of the tubes the following results were noted on the 20 mesh sand. The same general relations as regards the surface and sub-surface growth existed in the other grades of sand. This data indicates that the fungus remains below the sur- face in a well drained soil when the moisture content is low. "When the moisture content of the soil is high, however, or when it is saturated, as frequently occurs in poorly drained seed beds, 54 "NYiscoxsiN KKSKArtni Uilletin 81 the fungous growth is forced to the surface. This is presumably on account of the lack of oxygen, since the fungus is aerobic. At the surface of a saturated or nearly saturated soil it finds ideal conditions for development, both as regards oxygen and moisture, and if i)lants are present which are susceptible to the disease, damping-ofi' is almost certain to occur. In well drained soils, however, the fungous development is most active some small distance beneatli the surface, where there is both sufficient mois- ture and oxygen. This zone may be, and frequently is, below the base of the hypocotyl, in which case the plants may escape attack from the disease. Table VI. Influence of Soil Moistuke on Growth of Pythilm. Number Moisture Amount of surface giowLli Amount of sub-surface growth 1 Per cent 0 10 1.^ 20 '■*■) 30* None None None Slight Fair N'ery markecl V Sli"-ht .S Slight 4 ISIarked <) (' jSand saturated Effect of Transplanting Diseased J'lants. In the spring of 1908 tobacco seedlings were transplanted to the field from a bed l)adly infested with damping off. Many plants were set out which had a trace of the disease upon the stem, tliough the badly diseased plants were discarded. Soon after transplanting sever- al plants died in the field, and examination showed tliem to have damped-off, nndoulitedly from the disease transferred from the l)lant bed. Later in tlie season when the plants were about half grown, several gave tlie appearance of being stunt(»d and yellow, as if from m;iliiiit rition. ICxamination showed in all cases a full or partial girdling of the stem below the surface of the soil. (KigureXI). At the time of the maturity of the crop about two per cent of 1lie pl;in1s. appaivntly normal in their growth, were broken down by a h(>avy wind storm. Practically all these plants shoM'cd a break just below the surface of the ground, where the stem was wholly or partially girdled, to all appear- ances due to a continuation of the disease started in the plant Control ok Da.mimxg-off in Plant Beds f;-". bed. In 1909 licjiltliy jiiid infected i)lants Avere set out in the field in separate rows, the infected plants, with few exceptions, made a normal, growth, and examination Avhen they reached maturity showed no injuiy or only sears where the disease had apparently cojitimied, ])ut later healed over. The recovery of the infected plants was attributed to the relatively dry season. FIGURE XI. GIRDLING OF TOBACCO STALK DUE TO DAMPING OFF DIS- EASE. A type of field injury which results in the affected plants being stunted in their growth and readily broken down by wind storms. FIGURE XII. INJURY FROM PYTHIUM DE BARYANUM IN THE FIELD. The disease frequently extends a con- siderable distance up the stalk and passes into the leaves causing them to droop and die. The season of 1912 was again very wet, and a large amount of of this field injury due to dampiug-off took place. In a plot not over one-half acre in extent, over two hundred plants were counted which were diseased. The injuries on the whole, how- ever, extended further up on the stalk than in the previous years. Ordinarily, the stalk was blackened a distance of 18 to 36 inches above the surface of the ground, j)roducing a condition of the nature of " blackleg '\ The disease also frequently entered the leaves, causing them to droop and tuni yellow from lack of food. (Figure XII). Selby (22) observed a similar trouble in Ohio tobacco and considered it to be associated with damping-ofif due 56 Wisconsin Research Bulletin 31 to Rhizoctonia. A similar disease is also reported from Java, being due to Phyioplithora nicotinianae (Breda de Haan), and a "blackleg" of tobacco in Japan has been reported as due to Bacillus nicotinianae by Uyeda (31). Clinton (4) has observed a "canker disease" in Connecticut which is without doubt identical to this, and he suggests that it may be a bacterial dis- ease. At this station Pythium debaryanum was repeatedly ob- served on diseased plants at the margin of the diseased areas and taken in pure culture therefrom. Pythium debaryanum in this ease appears to possess an extraordinary degree of par- asitism, being usually known as attacking only succulent seed- lings. These observations indicate that it is poor practice to transplant seedlings with any signs of the damping-off disease upon their stems. The persons who pull the plants from the plant beds for transplanting should therefore be required to discard all plants showing browned or blaekend spots on the stem. The possibility of direct infection in the field seems most remote. Checking the Disease The control measures considered u^) to this time, that is, those of prevention and culture, are means which must be taken before the seed is sown, and cannot be applied thereafter. More often the practical plant grower does not consider these factors and desires a remedy after the disease has begun to do considerable injury in the plant beds. This is an impossible task providing the weiather conditions remain favorable to the disease; hence the importance of foresight in the use of preventive and cultural methods. However if the weather conditions are not continually favorable to damping-off, much can be done towards checking the disease in plant beds by a proper regulation of the tempera- ture land moisture conditions in hot beds and cold frames. Temperature Bclations. Providing other conditions are fav- orable,, damping-off is aided by relatively high temperatures of the soil and surrounding atmosphere. The optimum temperature for growth of Pythium is about 33 degrees C. (92 degrees F.) while that of Rhizoctonia lies at about 25 degrees C. (77 degrees F.). The growth of either below 16 degrees C. (61 degrees F.) is very slow. "When damping-off has started among the plants Control of Damping-ofp in Plant Beds 57 the grower should endeavor to keep the temperature as far be- low the optimum as possible. Presupposing the presence of both fungi as actively causing the damping-ofP, this will be almost an impossible task owing to the wide range of temperature which they cover. However, when damping-off occurs and the tem- perature of the glass or canvas covered frame is above 70 degrees F., it is important to remove the covers to reduce the tempera- ture as much as possible. Moisture Relations. As w^as pointed out before, a relatively high percentage of moisture in the soil favors damping-off. This is also true of the humidity of the atmosphere surround ing the plants. The best aerial growth of the fungi is only ob- tained in an atmosphere saturated with moisture. Where aerial growth of the fungi occurs, the spread of the disease from plant to plant is very much more rapid than when it has to make its way through the moist soil or through parts of plants in contact with each other, because of the lessened resistance offered by passage through the air. The humidity of the atmosphere of covered plant beds is usually much higher than that outside. Good ventilation, provided by partly or entirely removing the covers, will quickly reduce this humidity and dry out the surface soil more rapidly, hence effectively decreasinsr or entirely checking the spread of the disease. "'^I)'' j Infected areas. Damping-off frequently starts at one point in tho seed bod and spreads from this point as a center of in- fection. Where only a few^ of these centers of infection occur, and the w^eather promises to be favorable for the disease for some time, much can be done towards checking the trouble by removing the infected plants and soil from the bed, together with the surrounding area somewhat beyond the last signs of disease. In case of the damping-off of valuable plants wdiere only a few are grown, it is best to prick out the healthy plants and transplant them separately into other soil. Spraying the plants with potassium sulphide. Bordeaux mixture, formalin, ammonia, mercuric chloride, lime or sulphur in an effort to check the disease after it had once started gave negative re- sults. 58 Wisconsin Research Bulletin 31 Summary Damping off of seedlings in ])lant beds in Wisconsin is com- monly caused by one of two fungi, Pylhium deharyanum or Jxhizoctonia. These fungi are capa])le of attacking a large variet}^ of dif- ferent plants, as well as of living upon the dead organic mat- ter of the soil, and are hence extremely persistent when once present in the soil. Th.e disease is favored jjai'ticularly by certain weather con- ditions, such as excessive moisture and high temperatures, and very little can be done to check the disease when such condi- tions prevail. Therefore this necessitates the adoption of meth- ods which kill the fuiigi in order to prevent the disease. The preventive methods are of such a nature that they must be applied before the sowing of the seed, since the methods used to kill the fungi would also kill the seed. A number of chemical agents have been tested as fungicides against damping-off, but of these formalin alone has proven of any value under conditions favorable to damping-off. Treatment of the soil with formalin at strengths of one part formalin to one hundred parts of water and lesser strengths, as frequently recommended, does not kill the fungus. Although it may hold the disease in check for some time, it will allow it to develop later if weather conditions permit. The value of formalin at these strengths is therefore dependent, in a large measure, upon the time of the appearance of weather condi- tions favorable for damping-off. Treating the soil willi 1-50 formalin at the rate of two quarts per square foot of soil, will kill the fungi which cause damping- off, and will hence effectively prevent damping-off under the most favorable weather conditions for fungous growth. Forma- lin soil treatment is also somewhat beneficial in stimulating the plant growth and in killing some weed seeds. The chief ob- jections are the cost of the formalin, the time required for it to act, and the time required for the soil to dry out. Sterilization of the soil by heat has proven the most satis- factory method of preventing damping-off from- all stand- points, excepting that under certain conditions it may be more expensive than the formalin treatment. Steam sterilization by Control of l)A.Mi'i.N(i-uii' in I'lant Beds 59 the ''inverted pan" method is especially reeoniinended where a steam traction engine is on the farm or can be obtained in the neighborhood. Aside from preventing damping-oft', there are several bene- ficial secondary effects of soil sterilization by heat. These are principally the killing of all weed seeds and insect pests of the soil, and greatly increased size and vigor of plant grown on such soil. As a cultural control of dampiug-off, growers should avoid infected, poorly drained soils and thick sowing of seed. The only means of checking the disease after it has occured in the plant beds is to remove the covers in order to reduce the temperature and the moisture of the soil and of the air immed- iately above the plants. GO Wisconsin Research Bulletin 31 LITP]RATURE CITED. 1. /tkinson, G. F. Damping-off. Cornell Expt. Sta., Bui. 94 2. Atkinson, G. F. Some Diseases of Cotton. Ala. Expt. Sta., Bui. -n, pp. 30 to 39 (1892.) 3. Clinton, G. P. Root Rot of Tobacco. Conn. Expt. Sta., Ann. Rpt. 31, pp. 3G3 lo 368 (1907.) 4. Clinton, G. P. Notes on Fungous Diseases for 1906. Conn. Expt. Sta., Rpt. 1906, p. 325. 5. Colligne, W. E. Root and Stem Rot. {JRJiizoctonia violacea ' Tul.) Rpt. on Econ. Biol., 2, pp. 46 to 47 (1912). 6. De Bary, A. Untersuchungen liber die Peronosporeen und Saprolegnieen und die Grundlagen eines natiirlichen Systems der Pilze. Abd. Seuk. Geo., 12, pp. 225 to 370. 7. De Condolle. Mem. d. I\tus. d'bist. Nat. (1815). 8. Duggar, B. M. and Stewart, F. C. The Sterile Fungus Rhizoctonia as a Cause of Plant Diseases in America. Cornell Expt. Sta., Bui. 186 (1901). 9. Duggar, B. M. Fungous Diseases of Plants, pp. 444 to 452 (1909.) 10. Gifford, G. ]\I. The Damping-off of Coniferous Seedlings. Vt. Expt. Sta., Bui. 157 (1911). n. Gilbert, "W. W. The Root Rot of Tol)acco Caused by Tliidavia hasicola. U. S. Dept. Agr., Bur. Plant Tnd., Bui. 158 (1909). 12. Hesse. Pytliium de l)a?"riainnn ein ('ii(1()i)liytis('her Schmarot- zer. Halle. (1874). 13. Jones, D. R. The Damping-off of Coniferous Seedlings. Vt. Expt. Sta., Rpt. 20, pp. 242 to 247 (1906-07). 14. Koch and Luken. Uber die Veriinderung eines leichten Sandbodens dui^ch Sterilisation. Jour. f. Landw., 55, p. 161 (1907). 15. Lyon, T. L. and Bizzell, J. A. Effect of Steam Sterilization on the Water Soluble Matter in Soils. Cornell Expt. Sta., Bui. 275 (1910). 16. Pammel, L. H. Iowa Expt. Sta., Bui. 15 (1891). 17. Darbishire and Russell. Oxid^ation in Soils and its Relation to Productiveness. Part TT. The Influence of Partial Sterilization. Jour. Agr. Sci. 2, p. 305 (1907). ^ Control of Damping-off in Plant Beds 61 18. Pickering, S. U. Studies on Germination and Plant Growth. Jour. Agr. Sci., 2, p. 411 (1908). 19. Pfeiffer and P'rnnke. Tieitrag zur Frage der Verwertung elementaren Stickstoffs dureh den Senf. Landw. Ven^uchs. Stat., 46, p. 117 (1896). 20. Kussell and Hutchinson. The Effect of Partial Sterilization of Soil on the Production of Plant Food. Jour. Agr. Sci., 3, p. Ill (1911). 21. Russell and Petherbridge. Effect "f Partial Sterilization of Soil. Jour. Bd. of Agr., London, 18, p. 80.3 (1912). 22. Selby, A. D. Tobacco Diseases and Tobacco Breeding. Ohio Expt. Sta., Bui. 156, pp. 97 to 99 (1904). 23. Selby, A. D. Soil Treatment of Tobacco Plant Beds. Ohio Expt. Sta., Circ. 59 (1906). 24. Scherffius. W. H. and Taylor, H. W. Sterilizing Tobacco Seed Beds. Agr. Jour., Union, South Africa, 2 (1911). 25. Schulze C. Einige Beobaehtungen iiber die Einwirkung der Bodensterilisation auf die Entwiekelung der Pflanzen. Landw. Vers. Stat., 65, p. 137 (1907). 26. Spaulding, P. The Treatment of Damping-off in Coniferous Seedlings. U. S. Dept. Agr., Bur, Plant Indus., Circ. 4 (1908). 27. Stone, E. J. Methods and Results in Soil Sterilization. Mass. Hort. Soc. Rpt. 1902, p. 52. 28. Stone and Smith. Sterilization of Soil in Greenhouses for Fungous Diseases. Mass. (Hatch) Expt. Sta., Rpt. 1902. pp. 74 to 85 (1902). 29. Stone and Smith. Further Considerations in Regard to the Drop in Lettuce. Mass. (Hatch) Expt. Sta., Rpt. 1899, pp. 149 to 151 (1899). 30. Stone, G. E. The Present Status of Soil Sterilization. Mass. Expt. Sta., Rpt. 24. p. 121 (1912). 31. TH'eda, Y. Bacillus nieotiRnae, Sp. Nov. die Ursaehe der Tobalrvvelkrankheit oder Schwarzbeinigkeit in Japan. Bui. Imp. Cent. Agr. Expt. Sta.. Japan. 1, 39 (1905). 32. "Ward. ]\L Obser\'ations on the genus Pythium. Quart. Jour. Micros. Sci., New Series 22, p. 487 (1883). Research Bulletin 32 June, 1914 Black Rot, Shed Burn, and Stem Rot of Tobacco JAMES JOHNSON AGRICULTURAL EXPERIMENT STATION OF THE UNIVERSITY OF WISCONSIN MADISON. WISCONSIN CONTENTS I Black rot of tobncce Introduction 63 Slsrns of the disease 65 The causal organism 68 Isolation 68 Description 6© Inoculation experiments 70 Factors influencing the disease 72 Infection 73 Moisture 73 Temperature 75 Oxygen 78 Susrgcstlons for control 79 II Shed bum and stem rot of tobacco Introduction SI Investigations 82 Conclusions 84 Summary 86 Black Rot, Shed Burn, and Stem Rot of Tobacco JAMES JOHNSON I BLACK ROT OF TOBACCO ^ Introduction During the fermentation of tobacco there frequently occurs a decay of the leaf which is known among packers as black rot. The disease" is said by packers to occur particularly in the north- em cigar tobacco sections of the United States, including espec- ially the states of Wisconsin, Connecticut, Ohio, and Pennsyl- vania. The importance of this disease to the tobacco industry is illus- trated by the fact tliat the loss in Wisconsin alone, in 1892, was estimated at more than a million dollars. The disease rarely oc- curs to such a marked degree, but conservative packers estimate the loss to the industry as a whole in the United States, in some years at several hundred thousand dollars. The tobacco packers have long been seeking the cause of this trouble as well as suggestions for its control. The writer has had an opportunity to observe this discaso for several years, be- lieving that some definite microorganism must be associated with the disease. In the winter of 1912 an opportunity for a close study of this trouble Avas made possible by the discovery of the disease, during the course of its development, in a packing » The writer is indebted to L. R. Jones, plant pathologist, of this Station, for many suggestions in preparing the manuscript for publication, and to Mr. George E. Gary, of the P. Lorillard Tobacco Company, for valuable observations on the disease under practical conditions. = The term "disease" is extended in tbi? bulletin tg include maladies occurring upon the tissues of non-living matter. 64 Wisconsin Research Bulletin 32 case of tobacco. The slow progress of the studies in the past has been due largely to the peculiar conditions under which the trou- ble occurs. In practice the disease is not usually discovered un- til it has run its course, since there is little or no evidence from the outside as to what is happening in the interior of a box of fer- menting tobacco. The disease, occurring as it does during the fer- mentative process, has been considered by some as an advanced stage of fermentation. Though this is not the case, the close as- sociation with the fermentative process and its complexities, to- gether with the several other factors which may influence the disease, such as infection, moisture, temperature, aeration, pres- sure, and character of leaf, tend to make it very difficult to state definitely as to the possible occurrence of the disease and the amount of damage to be expected. It would be desirable before discussing this disease to take up in some detail the practical observations and theories of a num- ber of packers, the methods of tobacco culture and fermentation, and especially the facts and theories regarding tobacco fermen- tation, since these matters are all pertinent to the question of cause and control. For the sake of brevity, however, the methods of tobacco fermentation and the processes immediately preced- ing it will only be briefly mentioned. In the northern tobacco growing sections of the United States the crop is usually har- vested during the latter part of August and first of September. The plants are ordinarily cut off at the base and hung up in sheds where they go through the curing process, which is a grad- ual drying out of the leaves, and a changing from the normal green color to the chocolate brown. During this process the leaves lose fully from 70 to 80 per cent of their moisture, which passes off into the atmosphere, the leaves becoming thin and dry at the end of about two months. After curing, the leaves require moistening before they can be taken down and handled without mechanical injury. This is accomplished by a day or two of rel- atively moist weather or fog, after which the tobacco is said to come into "case", and is taken down, stripped from the stalks, and packed into bundles weighing approximately fifty pounds. The relative amount of moisture the leaves are allowed to take up in casing is important and will be considered later. The crop is usually delivered in bundles to the packers, by whom it is graded. The leaves of equal quality and len;j;th are tied into small "hands". The hands of the same grade are BiiACK Rot, Shed Burn, and Stem Rot of Tobacco 65 packed under pressure into boxes or cases of about 300 or 400 pounds each, with the butts out and the tops overlapping in the center. The boxes are then stored away for the fermenting or "sweating" process, at temperatures which vary greatly with the time of the year, and the methods practiced by the packer. In case no attempt is made to regulate the temperature in the store room the process is called a natural sweat. In case the tempera- ture is maintained above the normal by artificial heat, it is spok- en of as a forced sweat. A more recently introduced process is known as bulk sweating. In this process the hands are packed regularly into stacks usually of about 2000 pounds or more, in- stead of being packed into boxes. The fermentation goes on more rapidly and actively in this process. After the temperature has reached about 100 or 120 degrees F. the piles are taken up and repacked in a similar ^vay. This may be repeated three or four times, or the hands may be packed into boxes after the first bulk- ing, depending upon the judgment of the packer. The natural sweat is most commonly practiced in the northern tobacco sec- tions, and it is in this method that black rot usually occurs. Signs of the Disease Black rot, as the name implies, is recognized by the dark color taken on by the rotted areas in place of the normal chocolate brown. The disease as it occurs during its progress is charac- terized by being relatively moist, though there is no such decom- position and slime as is found in so-called wet rots. The leaf maintains its form until mechanically disturbed. The black rot as packers know it is quite conspicuously a dry rot, and the af- fected leaves when handled break up into a powder, the cells hav- ing lost all their cohesion. When black-rotted "hands" are shaken the injured portions drop out very readily. (Figures 1 and 2). Experienced packers easily recognize black rot as soon as a diseased bulk is opened, by the characteristic odor which differs from that of normally fermenting leaves. The odor is not obnoxious and would not be recognized as peculiar by the ordi- nary observer. In some instances the rot usually occui's only in the innermost portions of the bulk of tobacco, and especially at the tip end of the leaves where they were overlapped in pack- ing. Where the disease develops only in small areas throughout the case it suggests small, local favorable conditions for its de- FIG. 1.— BLACK ROT UNDER PRACTICAL CONDITIONS A Hand of tobacco, lower end diseased B A similar hand sbaken out, showing extent of injury Black Rot, Sued Burn, and Stkm Rot of Tobacco 67 velopment and it is in such small patches that it usually occurs. In bad cases much or all of the inner pack may be det-ayed, leav- ing but a shell of sound leaves which, owing to the unfavorable conditions for the disease found there, are not injured. Where FIG. 2.— a severe case OF BLACK ROT These tobacco leaves show the location and extent of the injury. the rot has been allowed to run unchecked to "maturity" there usually appear small areas of "black rot soot," filling the small air spaces which sometimes exist between the leaves. The line of demarcation between rotted and sound areas is usually quite 68 Wisconsin Research Bulletin 32 sharp, and the sound portions are still of value if large enough to warrant their separation. The disease may also occur in the bundles if they are allowed to start fermenting before being packed into cases. The Causal Organism Isolation. — Small pieces of freshly diseased tissue were teased apart on a slide and examined under a microscope. On the mar- gin of the bits of decaying leaf there frequently could be seen small pieces of fungous hyphae, which were witli difficulty dis- tinguished from pieces of cell tissue. Pieces of disens; d tissue were then plated out on potato hard agar in petri dishes and in- cubated at a temperature of 35 degrees Centigrade. After from 12 to 24 hours there occurred large colonics of bacterial growths, which were followed in from 48 to 60 hours by a marked develop- ment of fungal growths particularly of a species which was later determined as Sterigmatocystis nigra, V. T. Pure cultures were immediately made of this fungus, and other forms which devel- oped conspicuously on the plates including two species of Peni- cillium and two or three forms of bacteria. Samples of black rot were then obtained from three different warehouses and from several different cases in one warehouse. The rotted portions were carefully removed with sterile forceps to petri dishes and carried to the laboratory where they wci'e plated out separately. Samples of black rot were also obtained from the 1909 and 1910 crops of tobacco and in all cases S. nigra was found present in great abundance. On some of the 1912 crop which had rotted the spores of S. nigra were present in very great abundance on the diseased tissue.^ ' Loew, O., riiysiological studies of Conuecticut leaf tobacco. U. S. Dept. of Agr. Rpt. G5 : 7-57. 1900. See particularly 48, footnote reporting Sterigmato- cystis nigra observed by E. .\. Bessey on a specimen of blacls rot. In correspond- ence Dr. Bessey stated tliat he bad no proof that this fungus caused black rot aside from finding it on diseased tissue. Clinton, G. P., Report of botanist. Conn. Agr. Expt. Sta. Rpt. 1904: ;!2S. Re- ports occurrence of blacl< rot in Connecticut. Frear, Wm. and llibsbam, E. K., Tlie production of cigar-leaf tobacco in Pennsylvania. U. S. Dept. Agr. Farmers' Bui. 41G. 1910. See 23. Mention of occurrence of black rot in Pennsylvania. Rapaics, R. A., Dohany kornios rothJldasa. Magyor Dohanyujog 30, 2-4 1913. Rev. in Mo. Bui. Agr. Intel, and Plant Dis. 4. 659 1913. Sterigmatocystis nigra reported as occurring ui>oii tobacco in Europe. This is said to be the first Euro- pean report of the occurrence of this fungus on decaying tobacco. Black Rot, Shed Burn, axd Stem Rot of Tobacco 69 Description. — The decay frequently occurs without the spor- ulating stage appearing upon the diseased tissues, however. An analysis of the air of several different rooms in a Madison ware- house where black rot was occurring was made by exposing plates of potato hard agar for one and two minutes, and incubating at FIG. 3.—STERIGMAT0CYSTIS NIGRA ON POTATO AGAR The plate was exposed for one minute in an infected warehouse. 35 degrees C. The characteristic development of bacterial col- onies, followed by Stengmatocystis colonies, appeared. In a room where diseased tobacco was being separated from sound to- bacco an agar plate exposed for one minute developed approxi- mately 75 colonies of Sterigmatocystis. (Figure 3.) Air an- alyses were made in the same way of two other warehouses in Madison and four in Edgerton, where black rot was not known to occur at the time. Sterigmatocystis colonies were obtained in 70 Wisconsin Research Bulletin 32 all the exposures but in smaller numbers than in the first case mentioned. Inoculation experiments. — Because of the abundant and gen- eral occurrence of Sterigmatocystis, it seemed likely that this or- ganism, rather than any of the others obtained in culture, was the cause of black rot. Experiments in inoculation were now be- gun using- pure cultures of ^iterigmatocystis nigra. Tobacco leaves were spread out, piling one on top of the other, and from this pile, circles three inches in diameter were cut, avoiding as much as possible the midrib of the leaf. Thirty-fi\e grams of leaves cut in this manner were packed tightly in four-inch petri dishes and sterilized in the autoclave at 245 degrees F. for 30 minutes. Under sterile conditions these leaves were separated and sprayed with 10 cc. of water containing Sterigmatocystis spores from an atomizer which had previously been sterilized. ■Checks were made by spraying with sterile water only. The plates were then subjected to pressure and finally sealed with paraffin and incubated at 35 degrees C. After five days the white threads of mycelium could be seen in portions of the inoculated plates and after eight or ten days the slender conidiophores of Sterigmatocystis stood out conspicuously in large masses around the margin of the leaves as shown in Figure 4. After 15 days, some of the dishes were opened and the texture found to be de- stroyed, whereas the checks remained as sound as the day they were packed. When allowed to dry out, the similarity to the nat- ural rot was at once evident and was pronounced typical rot by several packers. It was thought possible that the sterilization of the leaves might enable the fungus to develop more readily upon them, but this was disproved by inoculating unsterilized leaves, and obtaining black rot. Fully 75 inoculations of tobacco have been made, and in practically every case where sufficient mois- ture and a proper temperature were present, black rot occurred. Inoculations of leaves in the same way with cultures of the other ifungi and bacteria obtained from black rot tobacco failed to pro- duce after two trials any signs of the disease. Some study has been made of the causal organism, microscopi- cally and in pure culture, in order to determine tlie species cor- rectly.* This fungus has been the subject of much micro-physi- * The organism was also determined by R. Tbaxter of Harvard University as Sterigmatocystis nigra. V. T. PIG. 4.— ARTIFICIAL IXOCrLATION A Check B Inoculated with Sterigmatoci/stis nigra 72 Wisconsin Research Bulletin 32 ologic experimentation in the hands of many investigators. The fungus appears to vary considerably in measurements with the various substrata upon which it has been grown, and consequent- ly too much stress cannot be laid upon the measurement of the organism. The following description of the fungus taken from black rot and grown upon hard potato agar is submitted. The mycelium is a hj^aline, septate, and branched thread from 3 to 10 jx in diameter. The conidiophores are hyaline, erect, and continuous, ending in a globose vesicle from 30 to 40 /* in diam- eter, just below which there usually is a slight constriction. The conidiophores measure from 240 to 400 ju, in length and from 10 to 13 /a in Avidth. The sterigmata are fusoid, and meas- ure from 10 to 12 by about 3 or 4 /x ; conidia are blackish l)roAvn, globular, smooth or warty, and about 3 or 4 /x in diameter. Sterigmatocyslis nigra has long been known to play an impor- tant part in various industries,^ in both beneficial and injurious ways, and it is not at all surprising to find it developing in fer- menting tobacco. It is said to be instrumental in the manufac- ture of gallic acid from tannin, and in the manufacture of opium, besides being responsible for the cork disease on cork oak trees. The fungus possesses the power of saccharifying starch and in inducing oxalic acid fermentation. A large number of enzymes have been isolated, especially proteolytic enzymes, and pectinase, the latter of which may account for the complete destruction of the texture of the tobacco leaf in black rot. Factors Influencing the Disease In the fermentation of tobacco there is great variation in con- ditions, brought about by a number of abnormal and artificial factors. The results of fermentation are consequently quite variable and whether a certain case becomes damaged by black rot or not will depend upon tlie presence or absence of one or more factors wliich influence the disease. The greater the num- ber of influencing factors, the more complexities arise in tne study of the disease. The four primary factors in relation to the disease are (1) infection with Sferigviatocystis nigra, (2) the percentage of moisture in the leaf, (3) the temperature of the leaves, and (4) the air supply. In addition to this there are sev- eral secondary factors which influence these primary factors, such as amount of pressure on leaves, character of leaf, time of packing, temperature and humidity of storage rooms, amount of "Lafar, F., Handbook of Technical Mycology, 2, II, 322. 1904. Black Rot, Shed Burn, and Stem Rot of Tobacco 73 infection, and degree of fermentation, which are important to consider in practice. The four primary factors will be con- sidered separately, since the control measures must be based up- on them. In influencing the disease these factors are so very closely interrelated and dependent upon each other that it is dif- ficult to deal with one without considering the effects of the other. Tliis point can be illustrated by supposing the tobacco packed with a percentage of moisture too low to allow Sterigma- tocystis to develop. As fermentation proceeds tobacco Icses from 10 to 15 per cent of its weight." Three-fourths of the loss is water and the chemical activity resulting may liberate sufficient moist- ure in local portions to allow the fungus to grow. The rapidly developing fungus, however, may soon exhaust its oxygen sup- ply and check its own development. A marked rise or fall in temperature, or a succeeding diminution of moisture, may bring about the same result. In general, favorable and unfavorable conditions may arise intermittently and in close sequence where such a number of primary and secondary factors are involved, rendering the occurrence and percentage of damage from the disease very problematical. Infection. — The possibilities of infection of tobacco with Ster- igmatocystis nigra under ordinary conditions is very great. The fungus is found to occur almost universally and has been isolated from soil and various kinds of fodder, as well as tobacco. Black rot sometimes occurs in bundles before they are taken to the ware- houses for packing, which indicates that infection may take place on the farm. In tobacco packing houses where the disease has once occurred, however, the spores are very much more abund- ant, more infection takes place, and more black rot is likely to occur. Moisture. — The percentage of moisture in tobacco at the time of packing varies considerably. There is some skepticism among practical men as to the influence of moisture on black rot. owing to the fact that comparatively wet tobacco is sometimes un- affected by black rot. As will be pointed out later, this is por- haps primarily a temperature relation influenced by moisture. In order to obtain some definite data on the matter of moisture relations, the following experiment was carried out. Thirty-five rolls of tobacco leaf weighing about 100 grams each were made by cutting leaves into strips two and one-half inches ° Jenkins, E. II., Slirinkage during fermentation. Conn. Agr. Espt. Sta. Rpt. 1896 : 285. 74 Wisconsin Research Bulletin 32 ill Avidtli. These rolls were raised to different percentages of moisture and infected by spraying with water and Sterigtiiato- cystis spores from an atomizer. After lying loose in a large moist chamber for a day in order to let the moisture permeate the leaves uniformly, the strips are rolled up and weighed quickly to a hundredth of a gram. The rol' was immediately put into an ordinary glass tumbler which was inverted into a half-petri dish. Melted paraffin Avas then poured into the space between the tumbler and the rim of the petri dish forming a very satisfactory sealed chamber. (Figure 5.) These chambers were incubated at from 32 to 35 degrees C. for approximately two months. The rolls were then taken out and weighed quickly, after which they were placed for three days in a drying oven at 90 degrees C, and weighed again for dry weight. The loss due to destruction of dry matter by black rot lowered the dry weight of the rotted samples in such a way as to render it impossible to get the actual percentage of moisture in each sample. The loss of weight during incubation as shown by weighing before dry- ing was correlated with the percentage of disease. In the rolls where no rot occurred there was only an average loss of 0.36 per cent, whereas in the considerably rotted rolls the loss of weight ran up to 1.57 per cent, and in those badly rotted up to 6.4 per cent of the total weight. These losses represent largely the percentage of dry matter destroyed by the black-rot organism and to a small extent that due to fermentation. The loss of weight during incubation was added to the dry weights and the average dry weight of the rolls determined, which was used as a basis of arriving at the percentage of moisture in each roll. The results of this experiment are summarized in Table I. Tajilk 1 Relatiox of Amount of Black Rot to Moisturk Percentage in Leaf 32 35 Dkgrkks C. IMoisturo, per cent Tests Amount of Black Rot None Consider- able rot Badly rotted 26-32.. 11 18 11 11 0 0 8 6 0 SS-Sfl 0 40-46 a The data show quite conclusively the increase in black rot with the increase in percentage of moisture in the leaf. Though Black Rot, Shed Burn, and Stkm Rot of Tobacco 7.> no black rot occurred in these experiments when there was less than 32 per cent of moisture present, this does not mean that black rot will never occur at lower percentages of moisture. At higher temperatures and correspondingly more active fermenta- tion, it is possible that black rot may occur even with 4 or 5 per cent less moisture. Sufficient moisture should of course be main- tained in the leaf to facilitate handling without mechanical in- jury and to allow fermentation to take place. According to "Whitney and Means'' this should be about 23 or 24 per cent in Florida leaf, and, according to the same authorities, the presence FIG. 5.— black rot MOISTURE DETERMINATIONS This illustration shows a simple and satisfactory sealed chamber used in moisture determinations. of 26 per cent moisture is likely to induce danger during fer- mentation. There is undoubtedly a considerable difference in the percentage of moisture at different points in the same case, which may account for the frequent small local areas of black rot which are to be found under warehouse conditions. Temperature. — Relatively high temperature is a predisposing factor for black rot. The fermentation of tobacco is accom- panied by a rise of temperature. Black rot is a disease occurring during fermentation and, as far as is known does not occur on tobacco except during the fermentative process. This at once leads to the conclusion that the growth of the fungus Sterigma- tocystis nigra is favored by relatively high temperatures. In or- ' Whitney, M. and Means, T. H., Temperature changes in fermenting piles of cigar-leaf tobacco. U. S. Dept^Agr. Rpt. 60:7-28. 1899. 76 Wisconsin Research Bulletin 32 der to determine the relation to temperature conditions, equal weights of pieces of tobacco leaves were packed in each of six petri dishes after having been inoculated and brought to an equal percentage of moisture, favorable for the black rot. They were then sealed iwth paraffin and two placed at each of the following temperatures, from 18 to 20 degrees C, from 31 to 33 degrees C, and from 44 to 46 degrees C. Black rot occurred immediately at from 31 to 33 degrees C. after 10 days, but did not occur at all in the higher or lower temperatures at the end of 28 days. In order to determine further the relation of growth of Sterig- matocystis nigra to temperature, fourteen tubes of potato hard agar were inoculated with spores of the fungus from a pure cul- ture and placed at seven different constant temperatures ranging from 10 to 46 degrees C. The growths of the fungus at the same temperature were remarkably equal, and were estimated in per- centage of surface area of agar slant covered at the end of the 24, 48, 72, and 96 hours as shown in Table II. Table II Rklative Growth in PehcentagI'; of Sterigmatocystin nigra at Vakeous TkMI'ERATURKS Temperature, C. 10-12° 0 0 0 0 0 0 0 0 l.j-lfi^ 0 0 0 0 0 0 0 0 21 23° 3 3 30 30 f)0 60 80 SO 28-29° 40 ■40 9.1 100 100 31-33° 40-42° 44-46° 24-liour cultuic I U .'50 r)0 100 100 30 30 90 90 100 100 0 0 JS-lioiif culture I 0 li 0 72-bi)iir culture I 0 11 0 :96-hour culture T I[ According to this table the most rapid growth of Sterigniato- cystis nigra took place at a temperature between 31 and 33 and at from 40 to 42 degrees V. the growth was being checked by heat. Unfortunately, no constant temperature chamber at tempera- tures between 34 and 40 degrees was at hand, but the table indi- cates that the optimum temperature must lie somewhere between 34 and 40 degrees according to this data or approximately 37 or 38 degrees C. — 98-100 degrees F. According to some authors* sLafar. F., Handbook of Technical Mycology. *2, II, 322. 1904. Black Kot, Shed Burn, and Stem Rot of Tobacco 77 the oi)ti!mini teinperatui'e for Sterigmatocystis nigra is approxi- mately 40 degrees 0. and the luiuimiini 7 degrees C. From this table can be drawn the conclusion that black rot will only occur between temperatures of about 20 to 44 degrees C. — 68 to 111 degrees ¥. — when all other factors are favorable. It is very probable that the limits of temperature in which black rot occurs are much more restricted than this, since it is favored by active fermentation which carniot bo said to occur until a temperature of at least 30 degrees C. — 86 degrees F. — is reached. The temperature of fermenting tobacco may run up above 44 degrees C. — 111 degrees F. — under some conditions, espe- cially in bulk fermentation, and even in ease fermentation, by either natural or forced sweat, in Avhich instance the develop- ment of black rot is impossible. In order to ferment tobacco properly it is impossible to avoid the temperatures which are most favorable to black rot, but in this connection it is very important to consider the length of time the tobacco remains at this temperature. If we compare the height of temperature and the comparative rate of rise of tem- perature in natural, forced, and bulk fermentation in practice, we find the following to be true. In natural sweat, the tem- perature rises ordinarily to from 30 to 40 degrees C. — 86 to 104 degrees F. — in a period of one to two months, depending upon the outside temperature. In the forced sweat a similar tempera- ture is ordinarily reached in from one to three weeks, depend- ing upon how much heat is applied from the outside. In bulk fermentation, however, a temperature of from 40 to 60 degrees C. — 104 to 140 degrees F. — may be reached in from 3 to 6 days.^ In natural fermentation, then, the tobacco leaves lemain at a temperature favorable for black rot throughout the entire time of active fermentation, and if sufficient moisture is present in the leaf, damage is almost certain to occur. Forced fermenta- tion is little better unless it raises the temperature above the maxinuim — about 111 degrees F. — in less than 5 or 6 days, as laboratory experiments have shown that considerable damage ' For more detailed information along this line the reader is referred to the following publications : Whitney, M. and Means, T. II., Temperature changes in fermenting piles of cigar leaf tobacco. U. S. Dept. Agr. Rpt. 60. 7-2S. 1899. Jenkins, E. H., The fermentation of the tobacco crop. Conn. Expt. Sta. Rpt. 1899, 291 to 297. McNess, G. T. and Massey, G. B., Tobacco investigations in Ohio. U. S. Dept. Agr., Bur. of Soils, Bui. 29 : 7-3S 1905. 78 Wisconsin Research Bulletin 32 occurs in 5 or 6 days with all conditions favorable. In bulk fermentation the tobacco passes through the dangerous tempera- tures rapidly, usually in 4 to 5 days, and the bulk then is re- packed before the rot starts, in which process considerable moist- ure is lost, diminishing to a large extent the poss'bility of black rot in the next bulk. FIG. 6.— BLACK ROT PRODUCED BY ARTIFICIAL INOCULATION WITH 8TERIGMAT0CYSTIS NIGRA These hands of tobacco were Inoculated and placed under conditions favorable to the development of tlie fungus. It completely destroyed them. Oxygen. — The amount of oxygen present has unquestionably a bearing on the extent to which black rot occurs. In the interior of a tightly compressed case of tobacco the amount of air pres- ent must be quite small, and readily exhausted by any growing organism before air can diffuse in from the exterior to take its place. More frequently in cases affected by black rot the disease occurs in small local areas, which suggests that it has been checked by some factor. This may frequently be due to a change of temperature or a reduction of moisture, but it is also possible that it is sometimes due to an exhaustion of the air supply, which Black Rot, Shed Burn, and Stem Rot of Tobacco 79 may prevent the growth of the fungus until the factor of temper- ature or moisture has changed sufficiently to prevent its r* sump- tion. Owing to the apparently very unfavorable conditions of air supply under which it grew it was at first suspected that the fungus causing black rot must be anaerobic. Attempts were then made to grow the fungus in the absence of oxygen, using in turn the method of pyrogallol-potassium hydrate, the hydrogen re- placement, and the paraffin oil covering. In all cases growths of Sterigmatocystis were obtained, but much slower and less vig- orous than those in air. Owing to the partial development of •certain other known aerobic forms of fungi under the samie conditions, it was concluded that the oxygen was not removed sufficiently or rapidly enough to prevent growth. The experi- ments, however, indicated that the fungus was capable of grow- ing slowly in the presence of a much reduced oxygen supply such as possibly occurs in the interior of a bulk of fermenting tobacco. The absence of oxygen is probably especially common in the tobaccos packed relatively wet, and may help to explain why w^et tobacco sometimes will not black rot, although this can usually be explained by the fact that relatively wet tobacco fer- ments more actively and hence develops a temperature unfavor- .able for black rot more quickly and maintains it longer than relatively dry tobacco. No importance, however, can be given 10 the matter of oxygen relations in the consideration of control measures, owing to the manifest impossibility of controlling it in practice. Suggestions for Control The most certain method of controlling the disease would be 'the destruction of the fungus either by killing the spores or by preventing their germination. There is little hope of doing this after the tobacco has once become infected, without possibly in- juring the quality. It is possible, however, that satisfactory methods of sanitation and fumigation may be developed which will serve to hold the disease tinder control. The sorting and packing rooms of a warehouse where black rot has once oc- •curred is a constant source of infection. The destruction of the tspores of Sterigmatocystis in these places would no doubt aid in a large measure to prevent a serious recurrence of the disease. This would not be a complete safeguard, however, because, as has ibeen pointed out infection may take place before the tobacco 80 Wisconsin Research Bulletin 32 reaches the warehouse. A thorough cleaning of the entire ware- house, removing and burning all the refuse, thoroughly sweep- ing and possibly washing the floors and walls, followed by paint- ing or whitewashing are secondary means which should be re- sorted to. The possibility of -formaldehyde fumigation of the warehouse is worthy of consideration by packers who are willing to practice strenuous methods in attempts to check this disease. The most essential point to consider in preventing black rot are the matters of percentage of moisture in the leaf at the time of packing and the temperature of fermentation. The growers should use all means possible to have the crop properly cured before freezing weather in the fall, and should not let the tobac- co come into too high case before packing in bundles. When the packers obtain relatively moist tobacco, there are several courses open to them to reduce this moisture before packing. 1. If only a few crops are in "high case" — have a high per- centage of moisture — they may be allowed to remain in the bundle until the latter part of the packing season, when they will have lost considerable of this moisture. 2. Bulk fermentation, for reasons already stated, may be sub- stituted for case fermentation with those crops which ontain a dangerously high percentage of moisture. 3. One "bulking" may be made before packing in cases, the extra handling removing considerable moisture. Where war- ranted, two bulkings previous to casing may be made. 4. The tobacco may be packed into cases and transferred im- mediately to a room temperature of from 40 to 42 degrees C. to insure the tobacco rising above a temperature of 44 or 45 degrees C. — fromllO to 113 F. in three or four days. Unless the temper- ature in the case remains above about 110 or 113 degrees F. for some time this forcing may be more injurious than beneficial, and a higher room temperature in some instances may be neces- sary. 5. Pack the tobacco directly into cases, and store it in rooms where the cases will not start active fermentation until late spring or early summer, when the tobacco should have dried out sufficiently to prevent rot from occurring during fermentation. This method probably has the disadvantage in being more likely to bring on the trouble known as "must" or "white mold". 6. It is i)0ssible that a drying machine or "evencr" may be used to I'egulate the moisture content of all the tobacco passed through it to a percentage that would be just sufficient to let it BrACK Rot, Sited Burn, and Stem Rot of Tobacco 81 ferment properly, but insufficient to allow damage to result. The tobacco should pass through the ''cvener" after the sorting process and immediately preceding packing. II SHED BURN AND STEM ROT OF TOBACCO Introduction Shed burn is a term applied to a decay of tobacco which oc- curs on the leaf tissue during the curing process. A similar de- cay on the midrib of the leaf during the same process is known as stem rot. These troubles are of frequent and general occur- ence throughout the tobacco districts of the country and espec- ially in sections where the natural curing process is practiced. Shed burn and stem rot are the cause of a very considerable less to the tobacco growers when conditions are favorable for the de- velopment of the diseases. Shed burn, pole burn or pole sweat as it is variously called, is usually recognized in mild cases by the development of small dark spots upon the leaf and a resulting loss of elasticity of the affected parts. In severe cases these spots may run together, destroy the texture of the entire leaf, and render it valueless for wrapper purposes. Stem rot is usually characterized by the development of whit- ish or pinkish molds upon the midrib of the leaf, which may cause decay to such an extent that the leaf drops from the stalk or the leaf tissue separates from the midrib. These shed diseases are known to appear under practical conditions as a result of relatively high humidity and temperature in the curing shed due to similar climatic conditions, poor ventilation, or too close hang- ing of the plants in the shed. According to Sturgis^° shed burn is a bacterial decay following a partial destruction of the leaf tissue by fungi. Examination of shed burned leaves in Connecticut, in 1891, showed the pres- ence of two forms of bacteria, one belonging to the genus Bac- terium, and the other being a Micrococcus. A fungus belonging to the genus Cladiosporium was invariably found associated with the decay and apparently always preceded the bacterial decay. In the same report Sturgis ascribes the stem rot disease to Boiry- 1* sturgis, W. C, Preliminary report on the so-called "pole burn" of tobacco. Conn. Agr. Expt. Sta. Rpt. 1891 : 168-173 and lS4-lSr). 82 Wisconsin Research Bulletin 32 iis longibrachiata and suggests sanitation and fumigation with sulphur for control, together with regulation of the temperature and humidity of the shed. A short time later Behrens", working in Germany, attributed the decay of curing tobacco to two closely related fungi, Sclera- tinia libertiana Fcld. and Botrytis cinerea P. According to his observations when these fungi attack the midrib of the leaf the effect is a stem rot and when they attack the leaf tissue the shed bum disease is produced. Sturgis^-, continuing his observations in 1899, found a species of Altei^naria ordinarily present, and believed this organism was the cause of the primary decay of the particular specimens of shed burn examined in that year. Plating out species of diseased tissues brought out several species of bacteria and other fungi, which occurred in such small numbers, however, as to preclude the possibility of their being instrumental in the destruction of tissue. Decay of curing tobacco has also been attributed by others to a species of Pleospora, Botrytis vulgaris, Fr. and to two different species of Mucor. In the fall of 1910, observations on the shed diseases of tobac- co were begun in Wisconsin, in order to determine in a prelim- inary way the causal organisms in this state. These observations were continued in the autumns of 1911 and 1912, in a general way, for the purpose of corroboration. Investigations Search was first made for specimens of diseased leaves in var- ious sheds during the curing season, and in this way several sam- ples of stem rot were obtained from different tobacco sheds in different localities, and a few specimens of shed burn were also collected. The specimens were taken to the laboratory and the causal organisms generally determined directly, if they were iibundant and sporulating. In some cases, however, the pieces of diseased tissue were plated out on potato hard agar, and the fungi obtained in pure culture from which their morphology was more carefully studied. Some inoculation experiments were also made by placing short pieces of green and thoroughly cleaned tobacco midribs in test tubes on a moist plug of cotton, " Rehrens, Dr. J., Trockcue und nasse Fiiule des Tabaks. "Der Dackbrand." Ztschr. fiir Pflanzenkrank. 3 : 82. " Sturgis, W. C, Further notes on the pole burn of tobacco. Conn. Agr. Expt. Sta., Rpt. 1899 : 2G5-2G9. Black Kot, Sued Burn, and Stem Rot of Tobacco 8' and inoculating with pure cultures of the fiin^i is!)latcd. Al- though infection was slow in all cases, and did not take place until the cells of the leaf had apparently become weakened from starvation, the tests served to show that the fungi obtained were capable of producing stem rot. Another method of obtaining the causal organisms was to place a number of mature tobacco leaves, taken directly from the field, in a tight, sterile chamber where the developments of mi- FIG. SUED liLKX 01' TOliACCO The early growth of Fusariitm upon the tissue of a tobacco leaf, removed with alcohol to contrast diseased area. Color of leaf cro-organisms could be watched daily and the source of iufec- tion from the curing shed was eliminated. In three or four days the beginning of fungus growths could usually be observed, es- pecially on the wounded portions of the midrib. As the leaf tis- sues began to change color in spots, indicating the death of the cells, it almost immediately decayed and became covered with mycelial threads of various fungi. The midrib proper, being the last to change color, usually resisted the attaclcs of fungi for the longest time but finally became completely overrun with them and frequently rotted down to a slimy, putrid mass. Pure cult- 84 Wisconsin Research Bulletin 32 ures were made of decaying tissues from these chambers iii the same way as those taken from the sheds. By these methods the writer has collected fourteen different forms of fungi from decaying tobacco during the last three years, many of which, however, occurred only rarely, and only three or four of which can be said to be of general occurrence un- der ordinary field conditions. A species of Fusarium was found of the most frequent and widespread occurrence on decaying to- bacco. Occurring quite frequently, but not to so great an extent, was TricotJiecium roseum and Penicillium brevicaule, and in a few instances a species of Alternaria and one of Botrytis. In the small curing chamber, Fusarium and species of Penicillium and Mucor seemed to predominate, and were later followed by bac- terial decay. Conclusions Sturgis has previously suggested that the initial decay in shed burn may be due to one or more of various fungi. The results of Bchrens observations together with those obtained here point toward the conclusion that no particular organism can be uni- versally ascribed as the cause of this disease. Moreover, it is to be expected that dead or dying plant tissues, such as occur in curing tobacco, will become substrata for various saprophytic organisms when conditions favorable for their development oc- cur. Furthermore, it seems most reasonable to suppose that such organisms as sometimes exhibit considerable parasitic action should be the first, and the ones most likely to produce the decay of curing tobacco. For this reason we may expect that the pure- ly saprophytic fungi should occur more rarely, if at all. The occurrence of species of Fusarium, Botrytis, Sclerotinia, and Al- ternaria seems to indicate that these organisms exhibit some par- asitism upon the slowly dying leaves. These observations indicate that the causal organisms of shed burn and stem rot vary in different sections of the country, and that it is in all probability a matter of the organisms with which the leaves happen to become infected, either from the field or from infectious material in the curing shed. The organism actually producing the decay may also vary with the tempera- ture and humidity as well as with the stage of curing, and con- sequently we may find a considerable variance in the different sheds and in different years in the same locality. Sturgis' con- Black Rot, Shed Burn, and Stem Rot op Tobacco 85 elusion that the fungi are not the principal organisms of decay, but that the bacteria which follow are mainly instrumental in causing decay has not yet been definitely verified, but the differ- ence here is probably a matter of the comparative amount of de- cay. When exceptionally favorable conditions for the develop- ment of microorganisms occur, as when the leaf is saturated with uncombined water, the development of bacterial decay is marked, but such conditions are comparatively rare in tobacco sheds. The ordinary shed burn, as far as my observations go in Wisconsin, seems to be due entirely to fungi. Sturgis attributes stem rot in Connecticut to Botrytis longi- hrachiata alone, while Behrens attributes stem rot in Germany to a Scleroiinia and a Botrytis, which also cause the shed burn disease. In Wisconsin, Fusarium particularly seems to cause the decay known as stem rot and shed burn equally well, and for this reason it appears that the two diseases may be caused by the same organism and that they differ mainly in the point and time of decay. The control of these diseases by fumigation of the shed before hanging the tobacco seems impracticable because of frequent in- fection of the leaves in the field as shown by the decay of leaves when hung in sterile chambers. The practice of sanitation, es- pecially the removal of refuse from the sheds where the diseases have once occurred, will no doubt aid in diminishing the amount of infection. For the present, the growers must look forward to the proper construction of curing-sheds with the object of con- trolling the humidity by proper ventilation and shutting out the moisture during damp weather. The regulation of temperature and relative humidity by fires in the shed when there is danger of the disease, or the entire substitution of the use of artificial heat for the natural curing method now in use in Wisconsin are methods of control which could be benofitially resorted to in this state. ^^ Summary black rot Black rot is a disease of leaf tobacco occurring during the fer- mentation process, especially in the northern cigar tobacco pro- ducing sections. "For practical discussion of control sec Johnson, J., The control of diseases and insects of tobacco. Wis. Agr. Espt. Sta. Bui. 337 : 7-34, 1914. 86 Wisconsin Research Bulletin 32 The loss from this disease is sometimes very great and conse- quently a consideration of control measures is important. The controls recommended are based on the determination of the causal organism of the disease, and the factors which influence its development. The rot has been shown to be due to the development of the fungus Sterigmatocijstis nigra V. T. upon the leaves in the pack- ing, causing them to blacken and decay in spots. In order for black rot to occur, tobacco must first become in- fected with Sterigmatocj/stis nigra after which a moisture con- tent of 26 per cent or more, a temperature of from 30 to 44 de- grees C— 86-111 °F. — and proper aeration must be present. The amount of damage occurring in fermentation will vary principally with the actual amount of moisture in the baves at the time they go actively into fermentation, the length of time a favorable temperature is maintained, and the amount of infec- tion present. The control of the black rot disease under warehouse condi- tions must depend on the regulation of the moisture percentage in the leaf, and the temperature o:^fermentation. Where the dis- ease has previously occurred sanitary methods should be used, and in extreme cases fumigation with formaldehyde may be of much value. SHED BURN AND STEM ROT The shed burn and stem rot of tobacco in Wisconsin may be due to one or more of several different saprophytic fungi. A species of Fusarium has been found to be most commonly the cause of these diseases in this state. The difference between shed burn and stem rot appears to be primarily one of location of disease rather than a difference in the causal organisms. If the fungi attack the midrib, the result- ing decay is called stem rot. If the leaf tissue is attackt>d the de- cay is called shed burn. The prevention of the infection of tobacco with the spores of these diseases by fumigation of the curing sheds with sulphur, as suggested by Sturgis, seems impracticable owing to the large amount of infection from the field. In controlling the disease the growers must rely upon regulation of the temperature and humidity of the curing shed by proper ventilation or the appli- cation of artificial heat. Research Bulletin 37 August, 1915 Germination and Infection with ^the Fungus of the Late BHght of Potato (Phytophthora infestans) I. E. MELHUS AGRICULTURAL EXPERIMENT STATION OF THE UNIVERSITY OF WISCONSIN MADISON. WISCONSIN CONTENTS Page Introduction 1 The effect of external influences on the spore germination 2 Method 3 Experimental data on spore germination of Phytophthora 4 Influence of temperatures between 0 and 30" G. on germination 4 Extent of germination as determined by count 7 Effect of constant and intermittent temperatures on germination.. 10 I nfluence of temperature on period of motility of zoospores 12 Relation of temperature to germination of zoospores.^ 14 Effect of low atmospheric moisture on the conidia of Phytophthora 1 5 Light in relation to germination 16 Method of germination in the field 18 Toxic action of leaf juices 18 Effect of frost 19 Effect of sugar on germination 19 Oxygen as a stimulus to germination 20 Discussion and conclusions , 21 Indirect germination of the conidia of Phytophthora 22 Evidence from earlier field observations that low temperatures favor epidemics of late blight 22 Geographical distribution in relation to spore germination at low temperatures 24 Gomparison of Phytophthora infestans with other Peronosporaceae 24 Effect of low temperature on the germination of the spores of other fungi 28 Intermittent temperature and germination 28 Germination of zo6spores after liberation from the sporangium... 29 Direct germination or development of a germ tube 30 Toxic effect of various chemicals and fungicides on the germination of the spores of Phytophtora and Plasmopara 32 Method 33 Experimental studies.... 34 Toxicity of copper salts 34 Toxicity of certain copper mixtures 35 Fungicidal value of certain sodium salts 37 Fungicidal action of several polysulphic' s 39 Discussion and conclusions 40 Gopper salts and Bordeaux mixture 40 Polysulphides 44 A study of infection of Dotato foliage with Phytophthora 46 Method .'. 46 Experimental studies 46 Influence of temperature on infection 47 Does chilling increase the susceptibility of the potato plant? 49 The rate of development of Phytophthora infection 50 Infection by direct conidial germination 51 The difference in susceptibility between the lower and upper surfaces of the leaf 52 Discussion and conclusions.?: 54 Summary 57 Spore germination of Phytophthora 57 Toxicity of certain salts and fungicides 58 Infection studies of potato foliage 59 Bibliography 61 Germination and Infection with the Fungus of the Late BHght of Potato (Phytophthora infestans)^ I. E. MELIIUS INTRODUCTION Spore germination and infection are the preliminary stages in the destruction of host tissue by most fungous parasites. It is well established that these early stages are dependent, in a large measure, upon environmental conditions and the interrelations of host and fungus. How and to what extent they react on the spread of any disease is not well understood and affords a fertile field for further contributions to plant pathology. A specific example illustrating these general statements is offered in the case of the potato disease caused by Phytoph- thora infestans. It has been recorded many times^ that the advent of favorable weather conditions coupled with the presence of the fungus and abundant host tissue, invariably leads to an epidemic. The control of this disease, or any other which develops in a similar manner, depends chiefly upon the elimination or destruction of these early stages. In other words, the fundamental principle underlying disease control, whether it be by rotation, sanitation, or spraying, involves the prevention of spore germination and infection. Hence the desirability of detailed information concerning ' This publication is based on investigations made in the laboratory of Plant Pathology at the University of Wisconsin, in collaboration with the Office of Truck Diseases, Bureau of Plant Industry, U. S. Department of Agriculture. I wish to express my indebtedness first to the University of Wisconsin and second to the U. S. Department of Agriculture, for providing financial su|)port to make this investigation possible. My sincere Ihanks are due to Professor I,. R. .Jones for many helpful suggestions during the pro.gress of this study and the preparation of the manuscript, and also to I'rofessor H. A. Harper, formerly of the University of Wisconsin. 2 Ward, H. M. Diseases in plants, pp. 149-151. 1901. Lutman, B. F. Twenty years' spraying for potato diseases. Vt. Agr. Expt. Sta. Bui. 159: 225-247. 1912. 2 Wisconsin Research Bulletin 37 not only the morphology of germination and infection, but also the retarding and accelerating external influences, becomes apparent. Much as Phytophihora infestans has been studied since its first appearance in Europe in 1842, our knowledge of the early stages of its life history has been extended little since the writings of Schacht (1856) and De Bary (1863). At that time the important morphological facts were carefully pictured and correctly described; but other important problems relating to external influences or environmental conditions were merely suggested, not solved. The purpose of this publication is to record and discuss data and observations collected during the last three years on these early stages of the development of Phyfophthora infestans. The subject matter falls chiefly under three heads : (1) the effect of external influences on spore germination; (2) the toxic action of certain chemicals and spray mixtures on germination at optimum temperature conditions; (3) infection and its relation to environmental conditions. These will be discussed in the order named. The Effect of External Influences on Spore Germination In a previous paper on Cystopus (Melhus, 1911) it has been shown that the methods used by the earlier students of the Peronosporaceae, while adeciuate for learning the morpho- logical facts connected with spore germination, were too crude and inaccurate to be followed in a more extended study dealing with the effect of environmental influences. This fact led me first to work out a method of germinating the asexual spores in abundance, at will. Following this, such questions as the minimum, optimum, and maximum temperatures for germination were determined. Other problems relating to germination were taken up subse- quently, such as germination out-of-doors, time requirement, percentage germinating, effect of temperature on period of molilily, and germination of zoospores. The studies then undertaken with Cystopus have been extended to Phyfoph- thora. Laii, Hi. 1(1111 oi' l*()'iAr()i:s Method Essenlially llic same mcUiocl as llial used in Llie experi- ments with Cystopus just referred to has been followed in these later studies and hence it needs little explanation except as to source of spore material and treatment of the cultures. Viable spores were obtained from two sources, pure cul- tures on potato blocks, and infected potato foliage. From the potato-block cultures the best spore material was avail- able from the fifth to the ninth day after inoculation. On the foliage the best spores developed two or three days after the first visible evidence of infection. The spores were removed from the pure cultures with a sterile needle and immediately stirred into a drop of water on the microscope slide. Spores were removed from the infected foliage (1) by carefully scraping them from the leaf with a thin-edged scalpel and (2) by placing a small drop of water on the portion of the leaf bearing the spores and then, with the dropper, immediately collecting it again. The latter method yielded more spores and was the one usually employed. Spores were never taken from leaf tissue that had begun to soften, as it often does when infected leaf or stem tissue is placed in a very moist atmosphere. The best spore material was obtained when newly infected plants were placed in an atmosphere sufficiently dry to prevent the tissues from thus softening materially. Clean slides were used, each bearing a drop of tap water which was later impregnated with spores. These slides or cultures, as they will be called in this bulletin, were now placed in a damp chamber, made by placing three layers of moist filter paper in the bottom of a six-inch petri dish and subjected to known temperature conditions.^ The number ' For temperatures below 14° C. an ice box was used with an ice capacity of 400 pounds. In order to know the continuous temperature during any and all of the tests, a Richard's self-registering thermometer was kept in the ice box. With proper care it was possible to maintain a fairly uniform temperature for twenty-four hours — the variation not exceeding a degree. In this manner it was readily possible to obtain temperatures as desired between 8 and 14° C. but for temperatures lower than 8° C, e. g., from 2 to 6° C another device had to be used. A caloric cooker of 3-gallon capacity was filled with water of the required temperature and placed in the ice box. In this way it was again possible to maintain temperatures ranging from 2 to fi° C;. for a pericJd of from eight to ten hours and for a longer time by adding a few pieces of ice every three hours. Temperatures between 0 and 1° C. were ob- tained by placing cultures on large blocks of ice in the ice box. For temperatures between 14 and 21° C, another caloric cooker was used in the way described above, except that it was kept in a room having a comparatively even temperature. For temperatures between 24 and 31° C, an electric incubator was used. 4 Wisconsin Research Bulletin 37 of tests made on the same date varied from one to nine, the usual number being two. Tests were always made at two or more temperatures at the same time. The duration of the various experiments varied considerably, ranging from 45 minutes to 48 hours, depending upon the object of the experiment. Examinations of the cultures were made at intervals of from one to six hours in some cases; in others no examination was made until the experiment was ter- minated. ExPEmMENTAL DaTA ON SpORE GERMINATION OF Phytophthora As is well known, the conidia of Phijlophthora infestans may germinate either by the liberation of zoospores or by the production of a germ tube without the intervention of zo- ospores. When the conidium becomes a sporangium and liberates zoospores, germination may be said to be indirect and when the conidium produces a germ tube instead, germi- nation may be spoken of as direct. This distinction will be made throughout this paper. Following the method just outlined, a study has been made of the relation of temperature to conidial germination of both types, and also of the behavior of the zoospores after their discharge from the sporangium. Other external influences such as light, dry air, oxygen, and density of solution have been considered in turn. The Influence of Temperatures Between 0 AND 30° C. on Germination Having previously learned that relatively low tempera- tures exert an important infhience upon the germination of conidia of Cijslopus Candidas, it seemed desirable to deter- mine whether or not a similar correlation of temperature and germination exists with Phijtophlhora infestans. To this end many tests were made at various temperatures between 0 and 30° C., as indicated in Table I. From one to nine cultures were made at a time, the usual number being two. This is not apparent in every case in the data given in Table I, because of the grouping together of like tests in the table. The duration of the tests varied Late Blight of Pot.\toi:s 5 from ;> lo 1 ~~ ^' / / V '\ ^. ^' ,*' N k / -f^^ rSs 1^ -^ ,y — — '' ^ \ N \ / 0£G/f£:£^S CeA/T/6/fAO£- FIG. 1. — RELATION OF TEMPERATURE TO GERMINATION Curve A represents the nuniher of cultures in which germination occurred at temperatures between 0 and 30° G. The ordinate represents percentage germina- tion and the absissa temperature. Curve B shows the variation in time reciuired for germination between 2 and 25° G. The ordinate here represents time. Extent of Germination as Determined by Count In Table I is shown the percentage of the total number of cultures held at a given temperature that did or did not show germination. It is plain that this is not as specific and direct as might be desired. It was therefore thought advisable to learn the percentage of germination in each culture by counting the spores in a given field. This was done and the results are shown in Table II. The counts were all made with a Leitz No. 4 eyepiece and jobjective with the tube length adjusted to give a magni- fication of about 124 diameters. Three different fields were counted in each culture and the sum taken as representing the prevailing condition. Three things were sought in the counts made: (1) the number of spores that germinated 8 Wisconsin Research Bulletin 37 indirectly, as shown by the empty sporangia, (2) the number that germinated directly, (15) the number that did not ger- minate at all. Table II. — Direct and Indirect Germination op Conidia of Phttophthora Inpestans* 6 6 6 6 6 6 6 3 6 6 6 6 6 6 To- tals Per Hours 24 24 24 16 20 16 24 20 24 24 19 23 28 18 5-6° C. 0 144 98 0 82 152 0 226 4 0 96 52 0 30 21 0 78 176 0 70 93 0 726 596 0 "iS Not germinating 4'i 10-13° C. Direct Indirect Not germinating 4 199 89 8 214 42 3 178 65 10 224 155 1 506 46 3 208 20 1 196 91 2 446 82 1 496 119 1 541 127 0 582 60 0 189 24 4 26 16 3 60 40 41 4065 976 0.8 80 19 17-18° C. 2 26 48 24 22 86 3 25 259 5 6 22 5 60 36 67 12 27 2 48 0 10 32 54 26 72 126 144 303 658 13 V 59 20-21° C. Direct Indirect Not germinating 14 24 122 31 22 62 15 84 119 39 0 172 90 9 84 62 25 248 13 5 23 40 6 61 46 60 210 24 64 289 18 2 105 34 15 215 82 48 214 36 47 76 40 69 217 584 480 2217 18 14 67 22-23° C. Direct 36 3 62 39 2 176 10 3 204 8 0 6 5 2 43 20 0 89 85 2 9 203 12 589 ?5 Indirect 1 4 73. 26-27° C. 34 0 50 40 0 181 5 0 165 8 0 29 0 0 39 0 0 143 0 0 89 48 0 97 135 0 793 14 0 85 28-30'^ C. 3 0 224 0 0 152 (i 241 1 .0 362 0 0 146 0 0 286 0 0 296 0 0 144 0 0 406 0 0 202 6 0 2459 0 •?. m * The figures indicate tlie behavior in each culture as determined Ijy counting the nuniljer of spores in a given area. In this series of experiments seven temperatures were used between 5 and 'M)° C. b^orty-two cultures were subjected to temperatures of 5-6° C. on seven different dates and three fields were counted {)er culture, a total of 126 different counts. The cultures at the other temperatures were treated likewise. Table II, shows the results of these tests, and at the foot of each column the total counts and percentages of germination are recorded. It is evident from these data that 5-6° C. is too low for the highest percentage of germination and that temperatures above 20° C. are too high. This is clearly shown in Figure 2, Curve A, and conlirms the results shown in Table I. That Late Blkhit of P()tatoi:s 9 temperatures between 10 and 13° are the most favorable is again shown in this set of data involving observations of many thousand spores. It must be emphasized that Curves A in Figures 1 and 2 have the same general outline, and this fact indicates, in my judgment, that the minimum, optimum, and maximum have been (piite accurately fixed. Curve R in h'igure 2 is worthy of note in that it has its beginning too \/' y y \ - -- 60 k '> fij^ y \ / y / \ 1 / / > \ 30 / f \ \ / f \ \ s / / ''n ^N ^., / ^^ -- ^ \ > \ O } e-i ^^0' .^ s \ > \ 0£6ffe£^ C£'/^T/Gfir'^C>£ FICi. 2.— TEMPERATURE IN RELATION TO DIRECT VS. INDIRECT GERMINATION Curve A shows Lhe percentage of indirect germination that tooi< place at various temperatures between 0 and 27° C. Curve B shows the porcontage of direct ger- mination between 6 and 30° C. very near the point considered the optimum for indirect germination and reached its highest point at 23° C, a temperature very near the maximum for indirect germina- tion. As shown in Curve B, the maximum for direct germ- ination is at or very near 30° C. The proportion of spores that germinated even at the optimum temperature, ranged from 13 to 99 per cent. That variation should take place is only to be expected considering the extreme susceptibility of the spores to other external influences besides temperature. Some of these will be dis- cussed in turn later in this paper. 10 Wisconsin Research Bulletin 37 These data show clearly that (1) an average of only 80 per cent of the spores of Phytophthora germinate, but the pro- portion may vary at the optimum temperature from 13 to 99 per cent; (2) the maximum for indirect germination lies between 23 and 26° C; (3) direct germination begins at or near the optimum for indirect germination, gradually increases up to or nearly to the maximum for indirect ger- mination, and approaches zero at 30° C. The Effect of Constant and Intermittent Temperature on Germination With the fact established that low temperature materially favors germination, the question arose as to the effect of constant and intermittent temperatures on the germination of the spores of Phytophthora infestans. It was thought probable that the favorable effect of the low temperature was due to the temperature change. Extra care was exercised in the selection of conidia for the study of this question. They were taken from pure cul- tures on potato plugs which were growing vigorously and producing spores in abundance. The water in which conidia were placed was raised to the temperature at which the cul- tures were to be incubated. At the same time another lot of conidia was placed in water at 20° C, held at this tem- perature for five minutes, and then incubated at a lower temperature, as shown in each case in the table. The time required for germination w^as used as a basis of determining the beneficial effect. In order to help establish the time at wliich germination first began, several extra cultures were carried in eacli experiment to be used in establishing the shortest time. By so doing, it was not necessary to disturb all of the cultures under investigation. It will be noticed in Table 111, that germination took place in shorter time in the cultures at constant temperature than in those held for five minutes at 20° C. and then placed at the lower temperature. Thus llie lime required in the seven different experiments averages .')(> minutes when constant temperatures were used and 00 minutes at the intermittent temperatures. The spores in two of the tests at the inter- mittent temperatures failed to germinate. The difference in Lati-: Hi.K.in oi' l\)rAr()i:s n time is so slight that it may well be clue lo variation in Ihe spore material or inaccurate observation, but the two negative results strengthen the evidence in favor of the constant temperature condition. At any rate, it seems safe to con- clude that a short period at high temperature has no bene- ficial effect. It was next thought desirable to learn whether a change from low to high temperatures was more favorable to ger- mination, and whether subjecting spores to a low temperature for a few minutes would be sufficient to cause germination. The tests recorded in Table III show that subjecting spores to low temperatures for short periods and then to high is not conducive to germination. Table III.— ("onidia ok Phytophthoka Infestans Subjected to Intermittent Tempkhatlres Culture? Cul Early cond experic tures at intermittent temperatures Cultures at Date itions of nent Later conditions of experiment constant temperatures Temp. Time Temp. Time Ger- mination Temp. Time Ger- mination Feb. 7 7 8 8 10 9 14 2 2 2 2 5 3 20 20 20 20 20 20 20 Min. •5 5 .5 .5 5 5 5 9 9 13 13 10 12 14 Min. 60 60 60 65 65 50 60 + + + + + 9 9 13 13 10 12 14 Min. 60 45 50 60 60 60 60 * + + -1- + -1- + -i- Dec. 14 16 21 3 3 2 12 12 12 30 30 30 25 2.5 25 120 120 120 - 12 11 12 60 120 80 + + Feb. 7 8 2 2 1 1 30 60 9 13 90 80 + + 9 13 120 60 + + Sept. 11 13 17 2 2 2 1 1 1 60 300 90 10 10 10 990 1380 96 ■ + 10 9 10 120 120 60 + + + * The plus sign {-{-) indicates germination and the minus sign ( — ) no germination. Ten other cultures were first subjected to 1° C. for a period varying from 30 minutes to five hours and then restored to temperatures more favorable — between 9 and 13° C. Ger- mination resulted in 6 of the 10 cultures. There was no indication that this low temperature for the periods indi- cated in the table was in any way strikingly favorable. 12 Wisconsin Research Bulletin 37 The Influence of Temperature on the Period of Motility of the Zoospores In connection with the studies of conidial germination, opportunity was afforded for observing the behavior of the zoospores. According to De Bary (1861), Farlow (1875), Ward (1887), and others, their motihty is supposed to vary from a few minutes to two hours. It was noticed that in some cases, in fact in a great majority of the cultures held at 13° C, the optimum temperature for indirect germination, zoospores continued motile longer than is recorded by earlier students of this problem. It was noticed too that the presence of foreign matter in the culture and the time required for the formation and liberation of the zoospores from the sporangium, also exerted an influence on the period of motility. Table IV. — Influence of Temperature on Duration of Motility of Zoospores of Phytophthoha Infestans Duration of motility at different temperatures, C. Cultures 5-6° 11-12° 20-21° 24-25° Hrs. Hrs. Min. Min. 3 14 3 60 25 4 24 10 90 10 8 19 8 30 30 9 21 5 30 10 4 22 6 45 20 20 6.4 51 19 Table IV shows the results of a series of experiments in which the conidia were carefully selected as to viability and placed in distilled water to insure absence of foreign matter. Unless 90 per cent or more of the conidia germinated, the material was not used. Every cflort was made to insure favorable conditions for normal i)eriods of motility. Zo- ospores produced at the optimum temperature for germina- tion were taken from the original cultures and stippled in small drops on clean slides and subjected to one of the four different temperatures shown in Table IV. Several pre- liminary tests, not recorded here, were made to obtain a general idea of the i)criod of motility. Table IV shows the results of five different tests, and averages for the periods. Lati-: Blight of Potatoes 13 to which the reader is referred, are placed at the foot of each column. It is significant that the period of motility decreases as the temperature increases, thus showing an inverse ratio. Loss of motility, which is a preliminary stage in the germina- tion of the zoospore, takes place most rapidly, not at the temperature most favorable for indirect conidial germina- 20 V "' ■n /e V \ /€ V \ /f > \ /e \ /o V a \ N € V V. s> <« ^ 'V < z V s» ^ o 1 i — ' — ^ /o /s zs FIG. 3.— EFFECT OF TEMPERATURE ON DURATION OF ZOOSPORE MOTILITY tion, but very near the optimum for direct conidial germi- nation, namely, about 25° C. The relation of temperature to the period of motility' is expressed in terms of a curve in Figure 3. Although only a few points have been determined, the curve seems to suggest quite definitely the response of the spore. It is safe to assume that it would approach zero if the temperature were increased above 25°, probably reaching it at some point in the neighborhood of 30° C, the maximum for direct conidial germination. This curve is of interest from still another point of view. It is well established that temperature influences the rate of respiration within certain limits and that the latter reacts on the rate of growth. Growth in the case of the zoospore 14 Wisconsin Research Bulletin 37 leads first to the deposition of a cell wall about the spore and later to the development of a germ tube. In view of these facts, the curve shown in Figure 3 may represent not only the period of motility, but also the relation of the rate of respiration as influenced by temperature expressed in terms of duration of motility of the zoospores. The period of motility within the limits given depends, in other words, on the rate of respiration, which in turn is determined by the temperature. The Relation of Temperature to the Germination of THE Zoospores It has been shown in the discussion of Table IV that tem- perature influences the rate of loss of motility and this has suggested that the growth of a cell wall about the zoospores is likewise facilitated by a favorable temperature. The development after the zoospore comes to rest by pushing out the germ tube, is also of interest and the data in Table V show the further influence of temperature on this. In this set of experiments the zoospores were treated like those used in the experiments recorded in Table IV, only held for a longer period. The number of zoospores that did or did not germinate was counted on equal areas in three different parts of each culture and the results are recorded in Table V. Likewise, three typical germ tubes per culture were measured and the average was taken as the prevailing condition. The cultures at 5-6° and 11-12° C. were held 20 and () hours longer, respectively, to allow the zoospores to lose their motility. The cultures at the three other temperatures received no extra time allowance to compensate for the period of motility because of its brevity. This probably had not a very great effect on the results. By referring to the averages in Table V, it will be seen that not all the zoospores germinate even after they are liberated from the sporangium. According to the data in Table V, the greatest number germinated at 11-12° C. As has been described by earlier workers, those that failed to germinate enlarged, became hyaline, and finally disin- tegrated. This was observed more often at the temperatures above 20° C. than at those below. The bursting of the T.ATi: Bi.hiht of Potatoks 15 zoospores placed at tlic higher temperatures may have been due to the sudden transition from low to high temperature, the increased osmotic activity at the high temperature being too great for the naked mass of cytoplasm. This fact makes it quite probable that many of the spores which disintegrated Tablb V. — Effect of Temperature on the Germination of the Zoospores of Phytophthora Infestans Cultures 5 5 3 5 3 3 6 Average 32 26 29 32 28 28 34 5-6° C. Number 102 67 43 42 44 80 38 99 71 207 89 89 60 76 80 33 41.6 45 70 Germination, per cent 68 11-12° C. Number 131 195 69 159 195 97 57 201 100 79 121 72 43 185 100 24 147 92 48 196 96 177 Germination, per cent 89-f- 14-15° C. Number 122 195 66 47 208 92 67 122 90 107 190 90 Growth 11 179 Germination, per cent 84 5 20-21° C. Number 39 198 85 63 224 90 48 272 88 127 147 40 71 222 96 49 192 82 Growth /I 204 208 80+ Germination, per cent 23-24° C. Number 50 233 48 33 281 85 42 234 95 60 201 82 Growth /i 239 Germination, per cent 77.5 at the higher temperature were missed in the final count, and if such is the case, the percentages germinating at 20-21° and 23-24° C. would be less than given in the table. Even neglecting this probable error, and accepting the accounts and showing the exact effect on the zoospores of the various temperatures, it is still evident that the lower temperature, 11-12° C. is more favorable for the early stages of zoospore germination than the higher temperatures. The effect of temperature is most marked on the rate of growth of the germ tubes. This is graphically shown in Figure 4, the growth being most rapid at 23-24° C, a point probably not far from the optimum. The Effect of Low Atmospheric Moisture on the CoNiDiA of Phytophthora Widely difTerent statements occur in the literature as to the conditions and time necessary for the spores of Phy- 16 Wisconsin Research Bulletin 37 tophthora Lo be killed when subjected to a non-saturated atmosphere. In view of this fact, experiments were under- taken to investigate this subject and the results are recorded in Table VI. Well-infected leaves with abundant spores were picked from the infected host growing in the same greenhouse, and placed in the laboratory to dry. Some of the spores were 200 ^ ^ ^ / -^ 0 V /oo so / / / / / / /s FIG. 4.— EFFECT OF TEMPERATIIHE ON RATE OF GERM TUBE GROWTH OF ZOOSPORES AFTER THEY HAVE LOST THEH^ I^TOTHTTY taken from the fresh leaves and, by trial, shown to have normal vitality. Germination tests were again made after the leaf tissues were dry and dead, when in all cases the spores failed to germinate. In cases of rapid drying, six hours was suflicicnt to kill all spores. Light in Relation to Germination It has been reported by De Bary (1863); Farlow (1876), and others that light checks the germination of spores of various Oomyceles. It has been shown in another paper, (Melhus 1911), thai if the temperature is kept sufficiently low, Cystoi)us spores will germinate regardless of the amount of light, be it diffuse or direct. It was thought desirable to study similarly the spores of Phylophthora. Cultures were prepared, and placed either in the light or in the dark. Light was excluded simply by placing the cultures in the ice box. The influence of light was studied by placing Latl; Blight oi" 1^o'iat(ji:s 17 cultures in a large glass moist chamber dish, ten inches in diameter and four inches high, in which cracked ice had previously been placed. The slides were laid on a small rack Table VI. — Effect of Dry \m o.n the Viability of the Conidia of I'hytophthora Infestans Cultures Age of spores Spores dried Incubation Date Time Temper- ature Germination Direct Indirect Mar. 1 1 2 2 2 2 3 4 4 4 2 2 4 da. 1 da. 4 da. 3 da. 3 da. 20 hrs.' 20 hrs.2 ehrs.s Hours 24 4 25 22 22 24 24 24 Degrees 12 12 12 12 12 12 11 13 0 0 0 0 0 0 0 0 0 Mar. 8 Mar. 11 0 0 0 0 Sept. 23 , Sept. 23 Apr. 1 0 0 0 1 Dried in laboratory 20 hours at 22-25° C. - .Artificially dried in a desiccator. 2 Dried at 25° C. for six hours on leaves until they were dry and dead. just above the ice with a small thermometer. This dish was then placed in the desired light relations, either in the green- house or outdoors. The direct sunlight naturally tended to raise the temperature within the dish, and the ice had to be renewed frequently during each trial. Under these conditions, it was impossible to keep constant temperatures, but this is not necessary to prove that light does not influence germina- tion. The extremes in each trial are given in Table VII, Table VII.— Effect of Light on the Germination of the Spores of Phytophthora Infestans Date Cultures Cultures in light Cultures in dark Hours Temper- ature Result Hours Temper- ature Result Jan. 31 2 6 6 3 3 3 8 6 10 4 7 ■5 3 Degrees C. 12-16 14-16 15-21 12-18 10-14 8-13 10-15 + + + + -1- + 3 4 3 2 3 3 3 Degrees C. 12 13 10-13 12 13 13 12 + Feb. 1 4- May 2 + May G + -1- + Feb. 4 + the variation being from 4 to 6 degrees, which, of course influenced the time reciuired for germination. It has already been shown that indirect germination is retarded by tem- peratures above the optimum. 18 Wisconsin Research Bulletin 37 Method of Germination in the Field It was next thought desirable to learn something about the type and prevalence of spore-germination under field conditions where the plants are exposed to natural infection. The best time to study this fungus is when the potato foliage is wet with either rain or dew. This means that observations must be made after a rain or in the morning when the dew is present. The latter time was chosen as most favorable. Forty-tW'O such examinations were made on six different dates, from September 7, to October 12, between 7 and 8:30 A. M. The dew drops were taken from both sides of the leaf, from shaded and unshaded leaves, from places w^here there was little moisture, and from others where there was considerable. In no case was anything but direct germina- tion found, and this type was present in every collection. The number of zoospores in the drops examined varied markedly, depending, it is believed, upon the amount of sporulation on the two previous days. It was noticed that when the amount of spore material was limited, the number of zoospores present was materially decreased. II should be noticed that the minimum temperature on the days when collections were made, ranged from 10 to 13 degrees C, temperatures very favorable for the direct germination of Phylophlhora conidia, according lo my laboratory studies. Toxic Action of Leaf .Juices It is well known that potato leaves and stems soften in the late stages of Phylophthora infection. It was noticed that very scanty germination was obtained when spores were taken from such tissues. Tests were made to learn whether this might not be due to unfavorable secretions present in the decayed potato tissue. A series of four different experi- ments, three cultures each, was made, testing germination in such juices from infected leaf tissue, with controls in water. No indirect and only scanty direct germination took place in the leaf juice, whereas the controls germinaled abundantly. The reason, Ihen, for the low pcrccnlage of germination when the spores are taken from decaying tissues, is found not in the spores, but in the presence of leaf juices. It is very pos- LaiI'; Hi.Kiirr oi- P()'i\'i()i;s 19 sible LhaL Lhey contain acids, or other decomposition prod- ucts, that have an inhibiting effect on germination. Effect of Frost It is generally known that the spores of Phytophthora are thin w ailed, and readily killed by frost, but whether they are more resistant than the foliage, is not known. This question was tested by taking spores from the leaves killed by the first "killing frost" in the fall of 1911. Six different tests of hve cultures each were made, but in no case did the spores germinate, indicating that a chill sufficient to kill the potato foliage, also kills the spores. Effect of Sugar on Germination With many fungi, sugar has a tendency to stimulate growth. Whether such is the case with Phytophthora infeslans, is shown by the studies that follow^ A series of dextrose solutions was made up, ranging from one to twenty per cent. Conidia of Phytophthora were placed in these vari- ous dilutions, in the same way as already described when using water. It was found that zoospore germination took place readily in a 1-, 4-, or 8-per cent solution, but more sparingly in a 16-, and never in a 20-per cent solution. The zoospores that were produced in the 16-per cent solution, were more or less distorted, and abnormally shaped, and remained motile only a very short time. This is in accordance with the behavior of certain motile algae in similar solutions. Only direct germination took place in the majority of cases in the 16-per cent solution. It w^as of interest to see if germination was more abundant in a 5-per cent dextrose solution than in water. The length of the germ tubes at the end of 24 hours was also noted in both cases. In the dextrose solution 1,361 spores germinated as compared with 751 in water. The average length of the germtube was 200 ^u in the solution and 197 in water. The difTerence in the growth during the first twenty-four hours was not so great. Whether it would have become greater after a longer time was not determined, but it h probable. It is obvious from these studies, that both direct 20 Wisconsin Research Bulletin 37 and indirect germination can take place in dextrose solutions ranging from 1 to 16 per cent, and that no germination results in a 20-per cent solution. It is also clear that a 5-per cent solution does not facilitate indirect germination, but rather tends to inhibit it, if the amount of germination in water is taken as a standard. Oxygen as a Stimulus to Germination It is claimed by Ward (1887), Klebahn (1909), and others that oxygen markedly stimulates germination. In order to test this with Phytophthora, conidia were placed in hydrogen peroxide of various strengths. Merck's brand, which contains approximately 4 per cent, was so diluted as to give the following percentages: 2, 1, 0.5, 0.125, 0.0625, and 0.0312. Viable spores were placed in a drop of each of these solutions on slides, and subjected to temperatures of from 10 to 13 degrees C. for 24 hours, with controls in water. The controls showed abundant germination after 2 hours, and so did the spores in 0.0312-per cent hydrogen peroxide. A few of the spores in the 0.0625-per cent solution also showed slight germination but none occured in the stronger solutions, even after 24 hours. The following day this expefriment was repeated, and germination again took place in the weakest two solutions. Only direct germination developed in the 0.0625-per cent solution. This experiment was repeated again with a similar outcome, as shown in Table VIII, where the results of the three trials with the weakest solutions are brought together. Table VIII Effect ok 0.0312-Pek Cent Solution of Hyduogkn Peroxidk on the Germination of the Conidia of Phytophthora Infestans Number qf cultures Time, hours Spores in solution Germination of conidia Direct Indirect Conidia not germinated Average length of germ tul)es n Spores in water Germination of conidia Direct Indirect Conidia not germinated Average length of germ tubes /i. May 2 May 3 6 6 24 24 0 (1 115 5i) .S i:!l 17!» (1 (1 411 114 2 49 190 201 May 20 78 82 164 0 119 22 206 \..v\'i: Hi.ic.iir ni- Poiatoios 21 11 will !)(.' noLiced from Ihose losls llial iiol only did a hi,ulicr l^crceiiLage ol" the si)ores germinaLc in water than in the hydrogen-peroxide solution, but also that the germination of the zoospores and the further growth of their germ tubes was retarded by the hydrogen peroxide. Oxygen in a nascent state, according to these results, is apparently not a stimulant, but rather a toxic agent. In several cases the hydrogen-peroxide solution was re- placed by water after 24 hours, but this did not revive the spores. Tests were also made, using a 0.0156-per cent solu- tion. Here the amount of germination was the same as in water, the oxygen present being insuflicient to produce a noticeable efTect. Whether a still weaker solution would act as a stimulant is a question. The data in hand, however, do not suggest that such is the case. Later, spores were placed in water containing practically no free oxygen. For this trial, distilled w^ater was boiled for two hours, cooled rapidly to 13° C, charged with spores, and cultures placed in an oxygen-free atmosphere for 3 hours. Germination occurred in the usual manner in abundance. Other tests were made in the same way except that the spore- charged water was covered with a film of paraffin oil. Spores treated in this manner behaved exactly as those in ordinary water. While these tests are somewhat crude, they indicate that free oxygen in the water is unnecessary for germination, and lead to the conclusion that the conidium contains within itself sufficient oxygen for germination. Discussion and Conclusions The conidia of Phijtophthora infestans can germinate either indirectly, that is, by liberating zoospores, or directly by the production of a germ tube, a fact first emphasized by De Bary (1803). Various theories have been proposed for explaining this dual nature of the Phytophthora conidium, none of which have been generally accepted or supported by any considerable amount of convincing experimental data. It is not my purpose, however, to discuss this question but simply to attempt to correlate the results of earlier investigators with my own as to the effect of external influences upon germination. 22 Wisconsin Research Bulletin 37 Indirect Germination of the Conidia of Phytophthora In an earlier paper it has been clearly shown that compar- atively low temperatures aid indirect germination of the conidia of Cystopus and thereby favor infection. The opti- mum for germination was not definitely determined for Cystopus but the results tend to show that it is about 10° C, with the minimum very near zero, while the maximum, as De Bary has shown, is about 25° C. The fact that Phy- tophthora is closely related to Cystopus makes it especially interesting to compare the behavior of the conidia of the two species with reference to temperature. The experi- mental data with Phytophthora recorded in the preceding pages show clearly that temperatures below 20° C. are more favorable to indirect germination than high. The data in both Tables I and II show that germination can take place at any temperature from 2 to 9° C. and that the optimum is about 13° C. It appears, therefore, that the conidia of Cystopus and Phytophthora behave much alike with refer- ence to temperature. Evidence from Earlier Field Observations That Low Temperatures Favor Epidemics of Late Blight Evidence that low temperatures favor the spread of Phytophthora is found in Jones' excellent observations in Vermont, extending over a period of 20 years. These have recently been assembled by Lutman (1911), so it is readily possible to see the moisture and temperature conditions at which epidemics developed. In 1891, when some blight was present, Jones writes as follows of the weather conditions: "The temperature was low and rainfall slight. The weather became very warm and on August 12 and 14 was followed by a fall in temperature and copious rains, and this was followed by another rise in temperature and more rains about the twenty-first. The conditions favoring the blight began thus about the twelfth." The following year, 1892, he writes: "From August G to 12 almost ideal conditions prevailed for the bhght and reference to the records show that it in reality appeared August 10, and progressed with unprecedented rapidity, so that almost every potato plant was destroyed bv it before August 20." The minimum temperature for Lati-: Blight of Potatoes 23 the period from August 6 to 12 ranged from about 15 to 18° C. Lale blight was also very bad in 189;5 when he again writes: "The later conditions favoring the IjUght l)egan about August 22, and the disease is recorded as under full headway by the twenty-fifth, and as continuing into September with unusu- ally destructive results." The minimum temperature for the period between August 22 and September 22 ranged from about 8 to 20° C, with 10 days above 15f° and 21 days below. Phytophthora was again severe in Vermont in 1902, when we fmd this statement regarding its advent: "The summer was moist and cool and the Phytophthora appeared at the earliest date on record here, July 13. * * * A similar period later in the season, August 18-27, finished up the later varieties, so that a canvass of the vicinity of Bur- lington on August 23 showed the plants in two-thirds of the fields entirely dead and rapidly dying in the remainder." The minimum for the period from July 14 to August 27 fiuctuated from about 10 to 22° C. with 25 days below 151° and 19 above. Similar obeservations by the same author occur in the Vermont reports 1891-1909 and might be cor- related in like manner, but the four years cited are typical. Halsted and Selby (1907) have made similar, but less extensive, correlations in New Jersey and Ohio, respectively. These show that an epidemic may develop when the min- imum temperatures are prevailingly below 20° C, a condition at which, according to our laboratory tests, indirect ger- mination is most abundant. It is common knowledge according to Morse (1909), that an outbreak of Phytophthora infestans in Maine follows a period of wet weather during the months of July, August, and September. A very excellent description of the con- ditions that favor an epidemic of late blight is given by Ward (1901) as follow's: "The now well-known spots * * * were observed during dull, cloudy, and wet weather, cooler than usual, when the temperature was saturated for days together in July and August. The actual amount of rain does not seem to have been excessive * * *. So rapidly did the disease run its course, that in a few days nearly all of the plants were a rotting, blackened mass in the fields." 24 Wisconsin Ri:si:ah(:ii Btlletin 37 Geographical Distribution in Relation to Spore Germination at Low Temperatures It is nol surprising that low temperatures are more favoral)le for the .germination of the conidia of Phytophthora infestans than higher ones, when we consider the latitude and climatic conditions under which the fungus thrives best. In a recent paper by Jones, Giddings and Lutman (1912, p. 14) one finds this sentence: "This disease is common in the northeastern states, especially in northern New England and New York and also in adjacent Canada. Further south and west it is either unknown or more sporadic, unless it be in the northern Pacific coast regions." According to Jensen (1887) the potato fungus can not exist where the mean temperature exceeds 25° C. In other words, it seldom occurs south of latitude 40° in the United States and chiefly north of latitude 50° in Europe. Minor variations in rainfall and altitude of course also influence the distribution. The preci- pitation of these regions is from 20 to 40 inches and they are crossed by the mean annual isotherm of 50° F. (10° C.)- A recent paper by Reed (1912) is interesting in this con- nection and shows the effect of altitude on the distribution of late blight. "It has been noted each year that the disease has not appeared until the advent of weather cool enough to bring lower temperatures during the night. At the altitude of Rlacksburg, Virginia, (2,200 feet) the cool nights bring an abundant formation of dew which apparently gives proper conditions for the germination of the swarm spores of the fungus." He continues by saying that the disease appears earlier at high altitudes and is i)ractically unknown below an altitude of 2,000 feet. These observations are interesting since they are made in a region south of the usual section infested by late blight, where a higher altitude, compensates for the difference in latitude. Comparison of Fiivtophthora infestans with Other Peronosporaceae Other parasites belonging to the same family as Phytoph- thora are similarly influenced by temperature. Pce-Laby (1899) has described the ravages of Peronospnra parasitica on l)rocc()li in southern France, where this plant is grown Late Blight of Potatoes 25 durini» the winter months. lie states that the Peronospora nourishes and is especially vicious during the coldest weather. The same fungus has been reported on cabbages in the south- ern states and southern California only during the winter by Orton (1910). I have noted (1912) that Peronospora parasitica is most prevalent on Lepidum virginicum in the spring and fall in the vicinity of Madison, Wisconsin, which further supports the contention that temperature influences the distribution of this fungus and shows, as well, its close likeness to the potato fungus. Another clear case is offered in Istvanfli and Palinkas' (1913, p. 25) studies of the grape mildew. In an extensive monograph they have recorded meterological conditions in conjunction with the infection periods of Plasmopara in Austria for the season of 1911. Their data show that the greatest number of infections took place in the spring and early summer. From May 5 to June 20 they follow^ed the development of 9 different outbreaks of the mildew, only one less than they studied all of the remainder of the season. Their chart shows, also, that the normal temperatures be- tween the above dates stayed below 20° C, with the mini- mum fluctuating mostly between 10 and 15° C. In view of this fact, it seems probable that a considerable proportion of the germination must have taken place below 20° C. Indeed, it is not improbable that many of the spores germinated in the range of the minimum (10 to 15° C.) w^hen dew might well have existed, although the authors make no mention of this fact. Besides moisture in the form of dew on the foliage of the host plant, it also occurs as rain. I have collected motile zoospores of Ph^'tophthora from potato foliage wet with rain on several occasions and also the zoospores of Plasmopara viticola from the wild grape. Whetzel (1904) in his studies of the downy mildew of the onion, Peronospora schleideniana, reports that he took germinating spores from infected plants in the field during a rain and at no other time did he see the spores germinating. (lerneck (1912) has recently called attention to the fact that heavy general rain storms accom- panied by high humidity and heavy dews facilitate the spread of the grape mildew. Halsted (1889) has on several occasions called attention to the correlation of rainfall and 26 Wisconsin Research Bulletin 37 Ihe developmciil of the Peronosporaceae, but makes no special mention of temperature assuming an important role. In my studies of the downy mildew of the wild grape it has repeatedly been noted that an outbreak of the disease fol- lowed a shower of rain or several days of cloudy weather. The conclusion seems perfectly justifiable considering my data and observations and those of other investigators on species belonging to the same family, that the low temper- ature conditions which prevail either during or following a rain are favorable for the germination of the spores of this fungus. A considerable amount of data is at hand as to the opti- mum temperature and limits of conidial germination of Plasmupara vilicola, which is worth comparing with the limits I have established for Cystopus and Phytophthora. According to Scribner (1886) the most favorable temperature for the germination of the conidia of Plasmopara lies between 25 and 35° C. Millardet (1886) found that germination took place in one and one-half hours at 9° C. Patrigeon (1887) reports that a temperature of 25-30° C. is the most favorable. Viala (1893) states that at 28-30° C. zoospore germination takes place in from half an hour to an hour, but if they are subjected to temperatures below 20° C the conidia do not germinate by zoospores any longer, but push out a germ tube instead. It is very evidcnl that Viala was influenced in his study of direct germination by the results of l)e Bary (1863) with Phytophthora, where direct germination is a common phenomenon, although recently Istvanffi (1913) has de- scribed and figured this type of germination as developing occasionally. Viala states further that it recjuired two or three days for germination to take place at 10-17° C, and that below 5° C, no germination took place, although the spores did germinate when restored to 25° C Recently Ravaz and Verge (1911) have published a short account of si)ore germination of the grape mildew in which they main- lain llial tlie minimum is 6.5° C. and the maximum very near 30° C; the optimum, using time as a basis of comparison, is near 27° C The time gradually decreases as the temper- ature increases up to 27° C, while at higher temperatures it again increases. It is unfortunate that Ravaz and Verge did not describe their method in detail and duplicate their Lati-: Blhiht of P()tatoi:s 27 experimeiiLs, so that their figures might carry conviction. That this is necessary is apparent from a very recent article by Gregory (1912). He summarizes his experience as follows: "Ordinarily germination is brought about by placing the conidia in a suspension of water in a van Tiegham cell. * * * At times, however, the conidia can not be germinated under these or any other conditions, so far as I have been able to determine. At best the germination is rather uncertain." I have experienced little difficulty in germinating the mature viable spores of Plasmopara, although, as stated earlier, no extensive study has been made with this fungus. In an excellent paper dealing with the grape mildew, Istvanffi and Palinkas (1913) report that no germination took place at 2° C. and that at 8° C. zoospores were formed only to a slight extent after 8 or 10 hours. At 14 or 15° C. the formation of zoospores was abundant in 2 or 3 hours. Germination was most rapid, however, at 20 to 22° C., at which temperature only from one to two hours was required. At 28 to 30° C. only slight germination developed after 4 to 10 hours. No zoospores were formed at 35° C. Here we have still another set of limits for the germination of the spores of Plasmopara. I I is hard to interpret these data in the light of the observations of Liistner (1905), Sorauer (1908), Gerneck (1912), and others to the effect that many days of wet weather and the condi- tions in spring and fall are most conducive to the develop- ment and spread of this disease. According to existing data on Plasmopara viticola, the optimum appears to lie some- where between 20 and 30° C., a strikingly high temperature when compared with the optimum fixed for Cystopus and Phytophthora. The question naturally arises as to the pres- ence of moisture on grape foliage, when the temperature ranged between 20 and 30° C., for a sufficient time to allow germination to take place. That a film of water is necessary has been repeatedly demonstrated by many other investi- gators. It seems to me that the question of spore gcrminaton of the conidia of the downy mildew of the grape is one worthy of further careful study in the light of my results with species belonging to closely related genera, and because of its direct relation to infection, spread of the disease, and its control. 28 Wisconsin Research Bulletin 37 The Effect of Low Temperature on the Germination OF the Spores of Other Fungi Low temperatures not only have a beneficial influence on germination and infection with these various Peronos- poraceae but also on a variety of other fungi. It has been known for a long time and has recently been re-confirmed by Hecke (1909), that the optimum for the germination and in- fection of some of the cereal smuts is about 10° C.. In the case of our common wheat rust, Jaczewski has found that both the uredo and accidiospores germinate best at tempera- tures slightly below normal (18° C). Johnson (1912) has re- cently confirmed Jaczewski's (1910) results for Puccinia graminis and extended them to several of the other cereal rusts. It may well be that when we have more accurate knowledge concerning the relation of external influences to germination and infection of other parasitic fungi, the num- ber known to show a preference for low temperatures will be materially increased, especially where there is as close a cor- relation between the presence of the fungus and climatic conditions as exists in the case of the potato fungus. Intermittent Temperaturi-: and Germination Eriksson (189(3) found that intermittent temperature, changing from low to high, facilitated the germination of the accidiospores of Puccinia graminis. Similar tests with the spores of Phytophthora have given me no such results as reported by Eriksson but, on the other hand, it has been found that constant low temperatures are more favorable. Likewise a transition from high to low temperature had no stimulating effect and in fact prevented indirect germination absolutely if the time at the high temperature was over one hour. In this respect the conidia of Phytophthora behave strikingly differently from the spores of various species of Ascobolus studied by Dodge (1912), who proved that sub- jecting the spores to a dry heat of from 50 to 70° C. for from 5 to 10 minutes and then returning to ordinary temperatures materially aids germination. But, of course, the two types of spores are widely different from a morphological and physiological standpoint. The latter have thick walls and live saprophytically, while Ihe former are thin walled and Latk Br.KWiT OF P()'rAT()i;s 20 live parasitically. The spores of Ascobolus have the nature of resting spores while those of Phytophthora are extremely ephemeral, remaining alive in dry air less than 24 hours. Germination of tiii-: Zoospori:s afti-:r Liberation from TMI-: Sporangium A great diversity of statements exists in the literature as to the period of motility of the zoospores. De Bar>' (1863) states that this period varies from 20 to 30 minutes. Farlow (1875) found it to be from 15 to 30 minutes. MacAlpine (1910) reported that they were motile only 8 to 10 minutes and Ward (1887) says that their motility varies from 1 to 20 minutes and that sometimes they give only one little flirt and then come to rest. In the case of the Phty tophthora infecting the Areca palm, the period of motility of the zoospores in the field is from 30 minutes to 2^ hours, according to Colman (1910). The results presented in Table IV and Figure 1 show that the period of motility varies inversely with the temperature within certain limits, ranging approximately from 20 hours at 5-6° C. to 19 minutes at 24-25° C. This variation with temperature explains the diversity of the statements of previous workers and also shows that they worked with temperatures too high for the best indirect germination. That the period of motility should be shorter at 24° than at 6° C. is only to be expected when we consider what happens to the zoospore in its process of losing motility. It gradually loses its lenticular form and deposits a cell wall at its periphery. This is a common and familiar phenomenon described first by Prevost (1807) and corroborated later by De Bary, Ward, and others. The deposition of a cell wall requires metabolic processes, or growth, and the rate of growth is much faster at 24 than at 6° C, because respira- tion and absorption of water increase with the temperature. This is further evident from the varying rate of growth of the germ tubes which develop from a quiescent zoospore (Table V). A similar influence of temperature on the motility of the zoospores of Olpidium viciae has been observed by Kusano (1912) who found the motile period shorter at 27° than at 15° C. 30 Wisconsin Research Bulletin 37 An interesting fact in this connection is the difference in behavior of the zoospore and the conidiiim. The most favorable temperature for the germination of the liberated zoospore is above 20° C. and probably about 25° C, while the optimum for indirect conidial germination is only 13°. The optimum for direct germination and germination of the liberated zoospore is about the same. In both cases my- celium is developed. The optimum for its growth, according to Jensen (1887), is 23° C. His observations were made on the mycelium growing in pieces of potato tubers. This is shghtly higher than that given by Orion (1911) and Jones (1912), who found that the fungus grew best between 16 and 19° C. on sterile potato blocks, but the difference in the meth- ods used by these investigators may well explain the dis- crepancy. In another portion of this paper dealing with infection it will be shown that the optimum for the growth of the mycelium in leaf tissue is probably very near 25° C. The difference in response of the liberated zoospore and the conidium is probably due to the fact that indirect ger- mination of the conidium requires but little energy. No new cytoplasm is formed. Cleavage, it seems, is brought about by the absorption of water and the translocation of the cytoplasm, due to a process of vacuolalion. These processes can go on at low temperatures. The germination of the free zoospore, on the other hand, requires the deposit of a cell wall and the development of a long germ tube, and these phenomena arc accelerated by higher temperatures. This is in accord, also, with the physiological principle that tem- perature directly inlkiences respiration, absorption of water, and growth within certain limits. Whatever may be the explanation of the difference, it is clear, as is graphically shown in hMgures 1 and 5, that while the period of motility varies indirectly, the growtii of the germ tubes varies directly with tlie temperature within tiie range given. DiRi:cT Gi:hmination or Devi:lopmi-:nt of a Germ Tube The conidia of Phylophthora are difl'erenL from those of the other members of the Peronosporaceae in that they can germinate either indirectly, that is by zoospores, or directly, by a germ tube. When the latter type of germination takes place, a germ tube grows out from one end of the conidium. Late BLKiirj or Potatoes 31 This type of germination is not inllucnced by temperature in the same manner as the indirect type, and fewer of the spores in water germinate in this manner. For indirect germination the minimum has just been shown to be 2-3° C. For direct germination the minimum is much higher, lying between 10 and 13°; at least at this temperature a very low percentage germinated directly. At 23-24° C, the optimum, 30.2 per cent germinated. This is somewhat less than the percentage given by .Jones (1912) who found that more than 50 per cent germinate directly at 25° and that this method of germination is exceptional at 10-20° C. It was plain that 30° G. was ver\^ nearly the maximum for germination. This maximum agrees with that fixed by Hecke (1898) for direct germination in a dilute leaf decoction, but my minimum and optimum are not in accord w'ith those given by him. He found the minimum 7° and the optimum 20° C. in potato leaf decoction, about 5° lower in each case than the figures I obtained, using water. It is not strange that there should be this difference considering the medium used in each case. The density of the medium doubtless influences osmotic activities, which in turn probably react on the spores. It is interesting to note that direct germination has its minimum near the optimum for indirect germination and its optimum at the maximum for that type. Hallier (1895) believed that young spores germinate directly and Ih^ older ones indirectly. De Bary (1863) concluded that direct germination was due to abnormal conditions of the conidium. It is interesting to note that Hecke reached the opposite conclusion from Hallier, namely, that the younger spores produce swarm spores and the older ones germ tubes. Farlow^ (1875) is inclined to believe that light influences the method of germination, at least to a certain extent, light favoring direct germination. According to McAlpine (1910) only the first crop of sporangia produces zoospores, all others germinating directly. By Ward (1887) the development of a geim tube by the conidium was thought to be due to the nutrition of the germinating spore, the amount of oxygen present, the number of spores in a given drop, and the comparative maturity of the spore. In an earlier part of this paper I have shown that the addition of oxygen has little influence. 32 Wisconsin Research Bulletin 37 It has been clearly shown by Ilallier that direct germina- tion is not an abnormality, as suggested by De Bary. Hecke has since substantiated Hallier. It is hard to believe with them that the method of germination depends upon the age of the conidium, or with McAlpine that all except the first crop germinate directly. Before it can be established that state of maturity bears any relation to type of germination, a criterion of degree of maturity must be established. This is only possible by making careful histological studies of the various stages in the process of development of the conidium. Livingston's (1900) work on Stigeoclonium is interesting in this connection. He found that osmotic pressure was the controlling factor in determining the form of the plant. When the osmotic pressure was low, zoospores were formed, but when it was high, increased vegetative activity resulted, inhibiting the production of zoospores. It may well be that osmotic pressure is the determining factor in the case of the germination of the spores of Phytophthora. Very little evidence is at hand as to the role played by direct germination in causing infection either to potato foliage or tubers. It has been shown that direct germination, at least, seldom occurs on the potato foliage in the morning dew. Hecke (1898) has suggested that this type of germina- tion functions partly in causing tuber infection in the soil, but presents no direct experimental evidence. He argues that, since direct germination is the only type of germination that can take place in a potato leaf decoction with more than 5 per cent solid matter (a solution comparable with a mixture of soil and water), and since old spores always germinate directly, considerable tuber-infection must result from direct germination. Further study of this problem is in progress and will be discussed in a later paper. Toxic Effect of VAmous Chemicals and Fungicides on THE Germination of the Spores of Phytophthora AND PlASMOPARA In the preceding chapter I have discussed the conditions favorable for the germination of the spores of Phytophthora infestans. It has been shown that temperatures below 20° C. are more favorable than those above and that the optimum is about 13°. The question arose as to the effect of various Lati-: Hi.HiiiT ov l*OTATOi-:s 33 chemicals and I'ungicides on indirecl germination at such low temperatures. Although the behavior of the spores of Phytophthora in a large number of salts and acids is known, no tests have previously been made under optimum condi- tions for germination and no study has been made of the effect of Bordeaux mixture and calcium polysulphide on the germination of the spores of Phytophthora. Moreover, earlier workers put the spores directly into the mixture to be tested, a method not comparable with what takes place in the open when foliage is sprayed with a fungicide. Method The method first used by Burrill (1907) and later elabo- rated by Reddick and Wallace (1910) has been followed in this study. It consists of spraying the compound to be tested on a clean slide and allowing it to dry after which a water suspension of the spores is placed on the slide. The slides that were used were allowed to remain in a cleaning mixture^ for 24 hours. F^ollowing this treatment they were thoroughly rinsed and dried between filter paper. All the water used in making the dilutions was distilled over alkaline potassium permanganate to free it from ammonia. Whenever possible Kahlbaum's chemicals of the highest purity were used. In the case of the copper compounds the dilutions were made in terms of copper and not of the salt as a whole. In all the other chemicals the dilutions are in terms of percentage as usually figured, namely, one gram of the compound dissolved in distilled water and made up to 100 cc. In the case of the poly sulphides, the dry weight was determined and dilutions made on the basis of percentage of solid matter. In Sherwin-Williams commercial lime sulphur, for example, it was found that each cubic centimeter con- tained 0.434 gms. solid matter. The dilutions were sprayed on slides immediately after preparation by means of a DeVilbiss atomizer and allowed to dry.^ The spore material used in these studies was selected with particular reference to securing high viability. Two species were studied, Phytophthora infestans and Plasmopara viticola, the latter being included primarily for purposes of com- < Potassium bichromate, 800 gms.; commercial sulphuric acid, 4,600 cc: water 3.000 cc. ^ The author is indebted to Dr. O. Butler for preparing the solutions which have been studied in this chapter. 34 Wisconsin Research Bulletin 37 parison. Distilled water containing spores was stippled in small drops on the slides bearing the various salts to be tested. The cultures were immediately subjected to 13° C, the optimum temperature for indirect germination of the Phytophthora. Controls were always run with each experi- ment and if the germination in pure water was not normal, the result was discarded. The controls also served as an index as to the proper time to continue the experiment. The time of final examination of the various cultures varied from 2 to 24 hours, but in the great majority of tests the duration was less than 5 hours. No attempt was made to estimate the percentage of germination taking place. Only its presence or absence was noted. Experimental Studies It seemed desirable to repeat some of the earlier studies on the toxic effect of copper salts, making use of the method just described. Because of the different method employed, it was impossible to use the earlier studies of Prevost (1807), Millardet (1887), and Dufour (1890) as guides. They put the spores directly into the solution to be tested, while in the studies that follow, the spores were placed on the dry salt after it had been sprayed on the slides in solution and allowed to dry for 24 hours or longer. Toxicity of Copper Salts Some of the common copper salts were tested as to their toxic effect on germination in order to learn whether any marked difference exists in these. Trials were made with each, in dilutions ranging from 0.06 to 0.00015 per cent of copper. In the following table only dilutions within the limit of toxicity are given, although a large number of tests were made in both stronger and weaker solutions. Five copper salts were tested, cupric nitrate (Cu (NO3). 3II2O), cupric sulphate (Cu SO4 .oILiO), cupric acetate (neu- tral) (Cu (C2H302).H20), cuprammonium sulphate (Cu SO4 4NH2.HiO) and cupric chloride (CuCla .HoO). All the salts except cupric nitrate were used on the spores of the two fungi, Phytophthora and Plasmopara. From one to five tests were made with each of the different strengths of the various salts, and each test included triplicate cultures unless otherwise stated. Late Bi.k.ht of P(nA'roi:s 35 It should l>e noted Ihal all of lliesc coi)j)cr salts are about equally toxic when they contain equal amounts of copper, except in the case of cuprammonium sulphate. Cupric nitrate prevented germination at 0.01 ')(> per cent, and cupric chloride at 0.0233 per cent. Cupric chloride is less toxic than any of the other salts. It is also evident from the data set forth in Table IX that the spores of Phytophthora are Table IX. — Toxic Effect of the Copper Salts on the Spores of Phytophthora and Plasmopara Coj)per Salt With P. infegtans With P. viticola Copper salts Expts. Cult Germ. ures Not germ. E.xpts. Cultures Germ. Not germ. Cupric nitrate Per cent 0.0312 0.0156 0.0078 0.0038 0.0159 0.0079 0.0039 0.0019 0.0012 0.0006 0.0199 0.0099 0.0049 0.0024 0.0233 0.01165 0.00.58 0.0029 0.079 0.00395 0.00197 0.00098 0.00049 Per cent 0.119 0.059 0.0,30 0.014 0.0628 0.0314 0.0157 0.0078 0.0039 0.0019 0.0628 0.0314 0.0157 0.0078 0.0625 0.0312 0.0151 0.0075 0.0314 0.0153 0.0076 0.0038 0.0019 1 1 2 2 3 4 4 4 4 5 2 3 2 2 Cupric nitrate 2 Cupric nitrate 4 4 0 0 1 2 11 14 0 2 3 0 0 0 0 2 0 0 0 Cupric nitrate 3 4 3 4 4 5 1 2 2 2 2 2 2 1 2 2 2 2 2 0 Cupric sulphate Cupric sulphate 0 1 2 4 15 19 0 4 7 7 0 3 4 2 0 0 0 6 18 6 8 4 6 0 0 3 3 0 0 4 1 0 0 6 6 18 12 0 6 9 Cupric sulphate Cupric sulphate Cupric sulphate Cupric sulphate 11 10 0 0 Cupric acetate Cupric acetate Cupric acetate Cupric acetate 9 ' 6 6 3 Cupric chloride Cupric chloride Cupric chloride Cupric chloride Cuprammonium sulphate Cuprammouium sulphate Cuprammonium sulohate 2 2 2 0 2 2 Cuprammonium sulphate Cuprammonium sulphate more resistant to the toxic effect 3f the copper salts than are the spores of Plasmopara, a fact that has previously been emphasized by Wiithrich (1892). Toxicity of Certain Copper Mixtures Having studied the copper salts as such, attention was turned to the various copper mixtures employed for the control of plant diseases. The mixtures studied and their composition, in the order cited in Table X are as follows: (1) Dauphiny mixture, 1 part copper sulphate crystals + 1.84 parts crystallized sodium carbonate; (2) Soda Bordeaux mix- ture, 1 part copper sulphate crystals + approximately 0.33 36 Wisconsin Research Bulletin 37 parts sodium hydroxide; (3) Woburn Bordeaux mixture, 5 parts copper sulphate crystals + 1 part calcium oxide; (4) Bordeaux mixture, alkaline to phenolphthalein; (5) Bor- deaux mixture, 1 part copper sulphate crystals + 1 part cal- cium oxide; (6) Bordeaux mixture, 1 part copper sulphate crystals +2 parts calcium oxide; (7) Bordeaux mixture, 1 part copper sulphate crystals + 1 part calcium oxide — cal- cium carbonate; (8) Bordeaux mixture, 2 parts copper sul- phate crystals +3 parts calcium oxide; (9) Commercial Paris green, copper acetoarsenite; 3 aceto-arsenite of copper. Ac- cording to Merck's Index it should contain about 27.7 per cent copper. Table X.— Toxicity of Variohs Copper Mixtures Used in Practice for the Control of Plant Diseases Per cent copper Per cent copper sulphate Results With P. infestans With P. xAHcola Fungicide No. exDts. Cults, germ. Cults, not germ. No. expts. Cults, germ. Cults. not germ. 1. Dauphiny mixture ... 0.0079 0.00395 0.00197 0.0159 0.0079 0.00395 0.0159 0.0079 0.00395 0.00197 0.00395 0.00197 0.00395 0.00197 0.00395 0.00197 0.00197 0.0079 0.00395 0.00197 0.5* 0.25* 0.125* 0.0314 0.0157 0.0078 0.059 0.0314 0.0157 0.0,59 0.0314 0.0157 0.0078 0.0157 0.0078 0.0157 0.0078 0.0157 0.0078 0.0078 0.0314 0.0157 0.0078 1 1 1 2 2 1 4 2 ? 4 0 5 3 0 1 3 4 3 0 8 0 0 11 6 4 2. Soda Bordeaux mixture 1 1 0 2 2 0 3. Wol)urn Bordeaux mixture 3 3 2 1 2 1 2 2 2 1 2 2 1 3 3 3 0 0 0 0 0 0 0 b 0 (i 0 0 0 6 5 5 7 11 6 9 6 4. Bordeaux mixture 5. Bordeaux mixture 8. Bordeaux mixture 7. Bordeaux mixture 8. Bordeaux mixture 9. Paris green 2 1 2 1 1 0 0 0 0 0 3 0 3 0 3 0 9 9 3 3 9 3 6 1 1 1 2 2 2 0 1 3 2 4 3 2 0 2 2 0 6 9 0 2 2 0 • Percentage Paris green, not copper. Soda Bordeaux, a mixture consisting of one part copper sulphate plus one part sodium hydroxide, killed the spores at 0.0159 per cent, but did not kill them at 0.0079 per cent. Lati-: Bi.Kiirr of Poiatoiis 37 Dauphiny mixture, which consists of one part copper sulphate and 1.84 parts of crystallized sodium carbonate, killed the spores of Phylophthora at ().(H)79 per cent, but one culture out of four showed germination at 0.00395 per cent. This last mixture is more toxic than soda Bordeaux. The current idea is that Bordeaux mixtures vary in efliciency with the amount of lime present. In order to shed light on this subject from the standpoint of toxic effect on spore germina- tion, the series of mixtures numbered 2, 3, 5, 6, and 8, in Table X varying in amount of lime, were prepared and tested. These Bordeaux mixtures were of approximately the follow- ing composition. The exact proportions of copper are given in Table X. Mixture Parts CUSO4 .5H2O Parts CaO 3 1 0.2 8 1 0.6 5 1 1.0 6 1 2.0 A comparison of the data recorded in Table X shows that increasing the lime does not increase the toxicity to spore germination, and that the various mixtures are quite equal in their action. Plasmopara spores are here again more easily killed by copper mixtures than are the spores of Phytophthora. Commercial Paris green did not prevent slight germination in 0.5 per cent solutions of its total dry weight. No analysis was made to determine the amount of copper present. It is doubtful, judging from the results obtained with commercial Paris green, whether it is sufficiently toxic to prevent ger- mination, and it therefore has little fungicidal value. This is especially interesting in view of the fact that it has been thought by various writers to have quite marked fungicidal value. Fungicidal Vaue of Certain Calcium and Sodium Salts When calcium hydroxide and sulphur are brought to- gether under the influence of heat and moisture, there results a mixture of variable composition, containing among 38 Wisconsin Research Bulletin 37 other compounds one or more polysulphides of calcium. Under ordinary conditions of exposure, these decompose, giving rise to various compounds, among which are free sulphur, sulphite, and thiosulphate. The thiosulphate appears to break down still further into sulphur and sulphite, both of these being finally oxidized more or less completely to sulphate. Table XI. — Toxicity of Somk of the Compounds of Sodium Polvsulphide Compound Crystallized sulpiuir Powdered sulphur Sodium sulphide Sodium sulphide Sodium thiosulphate Sodium thiosulphate Sodium sulphite Sodium sulphite Sodium hydroxide Sodium hydroxide Sodium hydroxide Sodium hydroxide Sodium sulphate Sodium sulphate Sodium hydrogen sulphide Sodium hydrogen sulphide. Hydrogen sulphide Hydrogen sulphide Hydrogen sulphide Dry weight Per cent 5 5 1 .5 1 l' With P. infestans No. expts. Cults, germ. Cults, not germ. With P. viticola No. expts. Cults, germ. Cults, not germ. ' Saturated solution. - Saturated solution diluted with equal volume wa: 3 Diluted with three parts water. The first substance tested was sulphur, ])olh crystallized and powdered, because it is advocated as having toxic properties, h'our experiments of three cultures each were made. Germination took place in every case as easily as in pure water. This is what one would naturally expect, in view of the fact that sulphur is not soluble in water. Six compounds of sodium were tested in various strengths, ranging from 0.125 to 2 per cent. As shown in Table XT, a one-per cent solution was necessary to prevent germination. Of these, sodium hydroxide was the most toxic compound. Normal germination took place in all the sodium compounds at one per cent, except as stated above, in tlie sodium hy- droxide. Sodium liydrogen sulphide was slightly toxic at at one per cent. i)ut at one-half per cent germination was Laip: Hlk.iit of Potat()i:.s 39 normal. Hydrogen sulphide (Mercks), saLuraLed solution, prevented germination, and so did a solution obtained by. mixing equal volumes of water and hydrogen sulphide. A solution of one part in three of water was not toxic. From the data at hand it again seems that Phytophthora is the more resistant of the two fungi under observation. It is clearly evident from the data in Table XI that under the conditions of this experiment, sulphur in water is not toxic to germination. Also that none of the constituents of sodium polysulphide are toxic at one per cent, except sodium hydroxide and hydrogen sulphide. The former of these is toxic at one per cent and the latter at one-half per cent. No attempt has been made to determine the exact limits of their toxicity, the object being to learn which com- pound was toxic and at what strength. Thp: F'ungicidal Action of Several Polysulphides Attention was now turned to a study of several poly- sulphides, such as calcium, sodium, and potassium poly- sulphides and the Sherwin-Williams commercial poly- sulphide (lime sulphur). Table XII. — Toxicity of Four Polysulphides Dry weight Dilution Results With P. infestana With P. viticola Fungicide and i)ropor- tion of ingredients No. expts. Cults, germ. Cults. not germ. No. expts. Cults. germ. Cults. not germ. Calcium polysuli)hide n.5S:5CaO Per cent 2. l. .5 .25 1. .5 .25 1. .5 .25 4. .5 .25 1:21.7 1:43.4 1:86.8 1:173.6 1:43.4 1-86.8 1:173.6 1:43.4 1:86.8 1:174.8 1:10.8 1:21.7 1:43.4 1:86.8 1:173.6 2 2 2 2 2 2 2 1 2 2 2 .3 3 3 1 0 2 4 0 1 6 4 4 0 0 1 8 3 4 2 0 4 5 0 0 0 0 4 6 2 0 1 4 3 3 4 3 2 .3 3 4 3 3 0 9 5 4 6 0 S 7 7 0 7 6 6 6 2 11.5S:5CaO 1 11.5S:5CaO 9 U.5S-5CaO Sodium polvsulphide 12.5S:8NaOH 0 12.5S:8NaOH .> 12.5S:8NaOH 0 Potassium polysulphide 3 0 0 Sherwin-Williams lime sulphur 5 4 .i 0 40 Wisconsin Research Bulletin 37 The compounds named in Table XII were evaporated to dryness, and the dry weight determined. The strengths are stated in the table in terms of percentages of dry weight and ratio of polysulphide to water. The proportions of lime and sulphur are also given for the calcium and sodium polysul- phides. A 2-pcr cent solution of potassium polysulphide killed the spores, while a 1-per cent solution killed only part of them. A solution of the same strength of sodium polysul- phides prevented the spores of Phytophthora from germinat- ing, but not those of Plasmopara. The strength of this com- pound required to prevent germination is undoubtedly close to 1 per cent. Potassium polysulphide was non-toxic at 1 per cent. Sodium and calcium polysulphides seem from the data at hand to be like in their effect on spore germination, in both cases 1-per cent solution being required. With the Sherwin-Williams commercial polysulphide, a 2-per cent dry weight solution was non-toxic to the spores of Plasmopara, though quite toxic to those of Phytophthora. A 1-per cent solution permitted germination of some of the spores of both I^hytophthora and of Plasmopara, but the number was less in the case of the latter fungus. This behavior of Plasmopara is of special interest in view of the fact that it has been heretofore more easily killed than Phytophthora. In conclusion it can be stated that the polysulphides here studied are toxic, but not markedly so. Sherwin-Williams calcium polysulphide at 2 per cent dry weight, or 1 part in 21.7 parts w^atcr, is less toxic to Plasmopara than to Phy- tophthora. Discussion and Conclusions Copper sdlis and Bordeaux mixtures. — It has been found, as has been shown by Millardet (1886 and 1887) and others, that the toxic constituent of Bordeaux mixture is the copper sulphate. A 0.00003-per cent solution of this copper sail in water was found by Millardet to be sufTiciont to i)revcnt zoospore germination of tlie conidia of Plasmopara vilicola. More recently Wiithrich (1892), using the same fungus, found that a solulion four limes as strong as that said to be toxic by Millardet permitted germination, while it required a 0.00124-per cent solution to prevent germination. Both of these investigators placed the spores directly in the solulion L.vn-: Bi.i(;irr of Potatoks 41 of copper sail, a nielhod nol comparable with the conditions taking place when a fungicide is api)iied as a spray mixture. In practice it is sought to deposit a thin film of the fungicide over the leaves to be protected. The object is, of course, to prevent new infection and not to check infections already under way. It is well established that when a fungus has obtained a foothold in the tissues before spraying, it con- tinues to fruit even though the surface of the leaves is covered with a layer of Bordeaux mixture. The fungicide on the foliage becomes efTective only when water and spores come in contact with it. The point I wish to make clear is that the conditions under which fungicides act in the open are much different from those secured by plunging the spores directly into a solution, as was done by Millardet, Wiithrich, and others. This fact has also been emphasized by Wallace (1911), who elaborated the method followed in this paper. It requires a 0.0315-per cent solution of copper sulphate to prevent the conidia of Plasmopara viticola from forming zoospores when the salt is sprayed on glass slides as pre- viously described, a method comparable with that generally in practice. By comparing these figures with Wiithrich's, it is evident that according to the method I have used a solu- tion over 25 times as strong as his is required to prevent germination. The contrast is similar though less extreme when we compare the figures given for Phijtophihora in- festans. Wuthrich found that a 0.0124-per cent solution of copper sulphate in water was sufTicient to prevent its ger- mination, whereas in my studies a 0.0628-per cent solution was required, i. e., nearly five times the strength. It is not surprising that a stronger solution should be required by my method since the fungus spore comes in contact with only a fraction of the whole spray mixture applied thus to the slide. The action of the lime sulphur in the liquid and dry state is nicely shown by Wallace (1911). It required a much stronger solution when the mixture was sprayed on slides and allowed to dry. This is easily understood when we remember that the spores are subjected only to the action of the film of spray mixture upon which the droplet of spore suspension rests, a comparatively small area. When this fact is considered, the discrepancy between my results given above and those of earlier workers is not so marked. Indeed 42 Wisconsin Research Bulletin 37 it may well be that they are wholly alike. Be this as it may, the significant point in my studies, not clearly brought out heretofore, is the fact that when spray mixtures are applied, as in practice, a stronger solution is required than shown by previous studies on the toxic action of copper salts. The question of the influence of temperature on the toxic action of fungicides and on the reaction of the spores is an interesting one. Prevost (1807), in his study of the effect of dissolved copper on the spores of Tilletia tritici, found that germination was materially retarded in a 1:1,000,000 solu- tion, at a temperature of from 6| to 7|° C, while it was not retarded in a 1:200,000 solution at higher tempera- tures. This condition in the smuts is just the reverse of that existing in the case of indirect germination of Phytophthora. The most efficient 2-per cent Bordeaux mixtures contain nearly 200 times as much copper sulphate as is needed to prevent indirect germination of the spores of Phytophthora infestans. This suggests that a reduction in the amount of copper sulphate can be made in Bordeaux mixture used for the control of Phytophthora if these laboratory tests are a true representation of conditions existing in the open. It has already been shown by Istvanffi (1903) and others that a one- and even a one-half-per cent Bordeaux mixture control Plasmopara on the cultivated grape. A one-fourth per cent mixture has been used for the control of apple diseases by Clinton (1912) with promising results. Hawkins (1912) has shown that a Bordeaux mixture (3-2-50), with 2 pounds of resin fish-oil soap as a sticker, is very efficient in controlling the black rot of grape. Should it be found that a reduction in the amount of copper sulphate, say from 1 to 0.5 per cent, can be made in the spraying of potatoes, it would mean a considerable saving financially, even though this salt is not very expensive. It is quite generally agreed that the fungicidal value of any copper spray mixture is due to the action of the contained copper, as such. The data tabulated in Table X bring in- teresting evidence to bear up this hypothesis. It will be seen that toxicity is governed by the amount of copper present in the compounds. It makes no diflcrence whether it is com- bined as nitrate, sulphate, or acetate; practically the same percentage of copper (about 0.0156 per cent) is required to Late Bi.ic.jit or- Potatoi;s 43 prevent germination in each case. When copper is com- bined as a chloride a slightly stronger solution is necessary, 0.0233. Cuprammonium sulphate cannot be directly com- pared with the other salts because it ionizes in a different way from the other copper salts studied. A solution only about one-eighth as strong as that required when using the copper salts named above was necessary to prevent germina- tion. This fact supports the contention made by certain writers that it is more eflicient than Bordeaux mixture. Its use as a spray mixture is not practical because of its toxic action on the foliage. The toxic limits of the Bordeaux mixtures studied are about the same from the standpoint of copper present, as already brought out for the copper compounds. Here again is evidence tending to support the theory that the copper ions, as such, are the fungicidal constituents. This, however, raises the question as to the action of the copper on the fungus spores. It is generally agreed by students of this ques- tion that it is the soluble copper that is active, but there is wide disagreement as to the way the insoluble basic copper salts are brought into solution. It is held by jMillardet and Gayon (1887), Crandall (1909), Pickering (1910), and Butler (1914) that the insoluble basic copper salt is brought into solution by some purely chemical means, e. g., by the action of carbon dioxide or free ammonia, present in meteoric water and the air. On the other hand, it is held by some that the exudate liberated from the host tissues through wounds or other abrasions may dissolve enough of the insoluble copper to prevent germination. My experiments, in which the mixture was allowed to dry on slides and later droplets of water charged with spores were placed on the mixture, show that the fungicidal action takes place even though the solvent action of the leaf is excluded. Frank and Kriiger (1894), Aderhold (1899), Clark (1902), Schander (1904), Barker and Gimingham (1911) are quite agreed that the fungus spore has the power to dissolve the insoluble salt in sufficient quantity to destroy its germmating capacity. It has also been shown by Aderhold and by Barker and Gimingham that the filtrate of Bordeaux mixture after standing a time is non-toxic, while the gelatinous precipitate is toxic. This is ascribed to the action of the spore on the 44 Wisconsin Research BuLLPrnN 37 insoluble copper salt, and Ihey believe thai the proximity of the spores to the copper particles is a matter of primary importance. It does not seem improbable to assume that the action of the spore in its immediate vicinity is one that should be considered where a germ tube develops, i. e., that it may have solvent powers due to its chemical actions on the medium in which it lies. On the other hand, it is not probable, although possible, that the conidia of Phytoph- thora when producing zoospores, excrete substances having solvent action. As already pointed out, little, if any, growth takes place in this process, it being largely ab- soiption of water and rearrangement of the cytoplasm within the conidium, followed by cleavage. These processes are not comparable with the growth of a germ tube such as occurs, for example, in the rusts. It is thus evident that the action of the copper on the spores presents some interesting problems not solved by these experiments and is a matter that deserves further careful study. Pulysulphides. — Lime sulphur has come into prominence during the last two decades and has in a measure replaced Bordeaux mixture as an orchard spray. Only one proprietary mixture of this type has been studied — the Sherwin-Williams lime sulphur. This may, without any particular unfairness, be taken as typical of many other commercial mixtures. It has been found that it is not decidedly toxic to the germination of the spores of Phytophlhora and Plasmopara. The latter fungus is ap- parently the more resistant. Since I have used Wallace's (1910) method, direct com- parison of our results, within certain limits, is })ossible. Such comparison makes it evident that the spores of Sphae- ropsis and Venturia, as tested by Wallace, are more resistant than those of Phytophthora and Plasmopara. This is in accordance with what we should expect, considering the difference in the type of spores. Sclerotinia is more like these mildews, i)olh in type of spore and in its susceptibility to poisons as shown by Wallace's tiials. The relative resist- ance of Venturia as compared with Plasmopara to copper sulphate has also been pointed out by Crandall (1909). Lativ I^i.ic.nr ok Potatoes 45 "Solutions perfeclly effective against grape mildew do not have even a retarding action on the growtli of spores of the scab fungus." The behavior of Phytophthora spores in my laboratory tests is interesting in view of recent field trials of lime sulphur as a potato fungicide. vStewarl (1912), in New York, found that it injured the potato plant, although in the absence of the disease he could not judge of its fungicidal value. Pethy- bridge (1912), in Ireland, pronounced it worthless for the control of Phytophthora infestans. This can easily be under- stood in the light of my results, which show that a 1:21 solution is necessary to prevent the germination of the spores, a solution stronger than that commonly used in practice. Sherwin-Williams' commercial calcium polysul- phide is about equally toxic with the other polysulphides tested. Potassium polysulphide, another commercial mixture used for spraying, has been studied from the standpoint of toxicity to spore germination by Forman (1911). He found it toxic to Botrytis spores when it carried 0.25 per cent of either potassium or sulphur. My own tests have shown it to be only slightly more toxic, if any, than commercial lime sulphur (calcium polysulphide). A comparison of Forman's results with those in this paper is not readily possible because he placed the spores directly in the solution. He also studied some of the decomposition products of the mixture which are of interest in the light of my tests on the same or similar compounds. He found that sodium hydroxide was the most toxic of the compounds tested. My results con- firm his findings, except that a slightly stronger solution was required. Forman showed that a 0.33-per cent solution was sufficient to prevent the germination of the spores of Botrytis, while a solution one-half as strong was not toxic. Potassium and calcium hydroxide of the same strength permitted germination, although a 0.66-per cent solution checked it. Sodium sulphide, sodium thiosulphate, sodium sulphite, and sodium hydrogen sulphide were all found to be non-toxic at 1 per cent. Two facts have been learned from the study of the compounds in sodium and calcium polysulphides: first, that only two of the more common compounds, namely, sodium hydroxide and hydrogen sulphide, are toxic, and, 46 Wisconsin Research Bulletin 37 second, they are not toxic at tlie strengths at which they can occur in the polysulphides. A Study of Infection of Potato Foliage WITH PhYTOPHTHORA Earher in this bulletin it has been shown that tem- perature markedly influences spore germination. Because of the close relation that exists between germination and infection, from the pathological standpoint, it was thought advisable to determine whether there is a similar relation of termperature to infection. Other closely related problems were also given consideration, such as susceptibility of the upper and lower surfaces of leaves and relation of method of germination to infection. Method The method used is in general that descri])ed in an article published in Phytopathology, (Melhus, 1912). Potato plants were grown in pots and were infected when from four to twelve inches tall. The spores of Phytophthora were placed in water and sprayed on the potato plants with a De Vilbiss atomizer. When the fresh conidia were used, the plants were chilled for from 6 to 24 hours in order to allow ample time for the conidia to germinate. When the plants were exposed to infection by using zoospores instead of conidia, they were held constantly at the greenhouse tem- perature. Infection usually became visible in from two to eight days after the plants were subjected to infection, de- pending upon the environmental conditions and the spore material used. Experimental Studies Following the above method, plants have been infected and treated in a variety of ways in order to give us a clearer understanding of how infection takes i)lace and what ex- ternal factors favor or retard it. The first experiments were |)lanned to determine the influence of temperature on infection. Lati-: ]5i.i(;ii'|- oi- Potatohs 47 Influence of Temperature on Infection Plants have been exposed to infection and held for a short time at different temperatures ranging from 10 to 32° C, as shown in Table XIII. It will be apparent by referring to fig. 5.— relatiox of temperature to infection by phytophthora Six potato plants were exposed to infection by spraying them with a suspension of conidia of Phylophthora infeslans on August 21, 1911. The three plants at the left were chilled, 1.3° C, for eight hours following the application of spores, whereas the three plants at the right were held continuously at the higher temperature, 25-27° C. The chilled plants showed 98 per cent of the leaves infected, while those not chilled, at the right showed less than 10-per cent infection. this table that in general at the low temperatures, 10-13° C, from 95 to 100 per cent of the leaves became infected in from 4 to 6 days. At 17° only 85 per cent showed infection and as 48 Wisconsin Research Bulletin 37 the temperature increased above this the infection per- centage decreased until at 25-30° C. there were few or no infections. Figures 5 and 6 show photographs of some of these plants. In Figure 6 the plant held at 28° C. shows no infection, the one at 13° was totally killed. It is plain, there- fore, that temperature influences the behavior of the spores of the leaves in the same way that it does when they are FIG. 6.— RliLAI'lON OF TLMPFRATHHF TO INFECTION in' PIIYTOPIITIIORA These two pluiits were Irealed as descrilx-d for Figure .'>, except that tlie plant at the right was held at 28° C It did not become infected, while t lie control at the left which was chilled at 1.3° C. was totally killed. placed on glass slides. The presence or absence of infection is determined primarily by the behavior of the spores. At high temperatures these fail to germinate and therefore the possibility of infection is precluded. These results are in general accord with those previously published (Melhtis, 1911) for Cyslopus. Late Hi.khit of I'otatoics 49 Tabi-k XI II. -Kffkct np Temperature on thk Amount of Infection Obtained by Kxposino Potato Plants to Phytophthora (Jonidia At Low Temperatures At High Temperatures Plants Refrigeration" Incuba- Leaf infec- Plants In moist chamber Incuba- Leaf in- tion2 tion tion2 fection Hrs. Degrees Davs Per cent Hrs. Desrees Days Per cent 4 21.5 11 4 95 4 21.5 19-20 4 50 1 i) 12 4 95 2 9 19-25 4 5 1 9 12 4 95 2 9 26-32 4 0 2 10.5 12 4 95 2 12.5 26-29 6 0 3 IS 11 5 95 3 18 17-18 5 85 2 22 13 6 95 22 18-19 6 57 2 20 14 5 95 20 26-32 5 0 2 30 10 5 98 30 15-30 5 50 2 18 13 4 100 18 19-25 5 33 1 18 13 4 80 18 19-25 5 10 1 18 13 4 74 1 18 19-25 5 4 1 18 13 4 34 1 18 19-25 5 17 5 8 13 5 98 3 8 25-27 5 10 ' Refrigeration period in this table, indicates the period the plants were held at the termperatures given. 2 Incubation period indicates the time from the date the plants were exposed to infection to the develop- ment of visible evidence of the same. Does Chilling Increase the Sltsgeptibility of the Potato Plant It is also of interest to learn whether the potato plant be- comes more susceptible when chilled a few hours. In other words, does a comparatively low temperature influence the reaction of the host to the fungus? In order to secure evi- dence on this point, it was necessary to devise some method of inoculating potato plants at both high (23-27° C.) and low (12-15° C.) temperatures. As shown in the discussion of Table XIII, infection does not take place readily unless the plant is held at 12-15° C. This was due primarily to the less vigorous germination of the spores. In order to avoid this difTiculty, the spores were germinated in water in a beaker and later the zoospores were applied to the plants. It has been shown in an earlier part of this paper that when the zoospores are liberated from the conidium they grow- more rapidly at about 24 than at 13° C. Experimental data bear- ing u.pon this point are recorded in Table XIV. Two lots of plants (var. Irish Cobbler) were sprayed with zoospores and subjected to like conditions in all respects except as to temperature. In one lot the plants were chilled at 12-14° C. for from 8 to 24 hours. An equal number constituting the second lot were not chilled, but held constantly in a greenhouse where the tem- 50 Wisconsin Rkskarch Billetin 37 perature varied from 23 to 27° C. Table XIV shows that approximately the same amount of infection developed on both the chilled and unchilled plants. It is worthy of note that it reciuired nearly a day longer for the plants held at Table XIV. — Comparative Effect of Low Temperature on the Susceptibility of the Potato Plant Height of plants Exposed to infection at 12-15° C. Exposed to Infection at 23-27° C. Infected Plants ChiUed Infection .Plants Infection Time Temper- ature Visible Percent- age Visible Percent- age Jan. 24 Jan. 25 Jan. 29 Feb. 5 Feb. 6 Feb. 12 Feb. 14 Feb. 24 Feb. 25 In. 7 8 6 6 6 12 10 12 8 3 3 3 3 3 4 4 5 2 Hrs. 8 8 19 24 18 24 24 24 24 Degrees 12-13 14-15 12-13 13-14 12-13 12-15 10-13 12-14 12-13 Jan. 28 Jan. 28 .,. Jan. 31 Feb. 8 Feb. 9 Feb. 16 Feb. 18 Feb. 28 Mar. 2 95 80 90 90 85 95 85 95 95 3 3 3 3 3 4 4 5 2 Jan. 28 Jan. 28 Jan. 31 Feb. 7 Feb. 8 Feb. 15 Feb. 17 Feb. 27 Mar. 1 80 8U 90 95 90 95 95 85 85 the lower temperature to show visible signs of infection than for those held at the higher temperature. The fact that the chilled plants were slower in showing infection was probably due (1) to the longer period of motility of the zoospores on the plants chilled, and (2) to the fact that their subsequent rate of growth was retarded by the lower temperature. In an earlier part of this paper it has been shown that low tem- perature retards both loss of motility and germination of the liberated zoospores. This fact considered, the time reciuired for infection to become visible on the chilled and unchilled plants was not materially different. Likewise the amount of infection in each case was ven>^ nearly the same. It is plain, too, that chilling for from 8 to 24 hours does not make the host more susceptible. Tiiii Rate of Development of Piiytopiithora Infection Various statements exist in the literature in regard to the time required for infections of Phytophthora to become visible. The usual time given is 6 days, although many in- vestigators allow even 10 days. It has been suggested in the above paragraph that temperature is a factor. In the series of experiments recorded in Table XV, the plants were L.\Ti-: Bi.K.iir oi- 1^)IA'I()i:s 51 sprayed with a water suspension of zoospores obtained in the same way as described in the discussion of Table XIV and held at dilTcrent temperatures, namely, r2-l(i° and 25-27° C. Infection readily took place on both lots of plants. The amount was equal, but the rate of development or spread of the funiA riVl I^l'l'l-Urr OI" C()MII)L\L AND ZOOSPOH1-: INFl'.Cl ION 'rhc plant at the lofl was sprayed with a siispesion of zocipsores in water and held in a saturated atmosphere in the greenhouse for 18 hours at a temperature of "if) 27° C rhe other r)lant was sprayed witli a suspension of conidia and chilled for 24 hours at 12 14° C 'l"he ehillinR for 21 hours did not increase its suscepti- bility. Practically the same amount of iiirection developed on the two plants ex- posed to zoospores and conidia. to infection through the upper surface Ihiiu Ihe lower. (Figure 7.) The difference in susceptibilily of the upi)er and h)wer surface of the leaf is attributed lo the difference in Llic rela- tive number of stomata. I. ATI. Hi.K.irr OK P()TAr()i:s 61 LITERATURE CITED Only those titles to which reference has been made in the text are included in this bibliography. Papers not examined by the author but accepted on the authority of other writers are marked by an asterisk (*). Aderhold, R. F. T. Ueber die Wirkungsweise der Bordeauxbriihe (Kupferkalkbruhe). Centbl. Bakt. [etc.] Abt. 2. 5:521. 1899. *Babo, a. von, and Mach, E. Handbuch des Weinbaucs und der Kellerwirtschaft. 1:1076. 1910. [Aufl. 3.] Bedford, Duke of, and Pickering, S. U. Copper fungicides. Woburn Expt. Fruit Farm Rpt. 11:1-191. 1910. Bred.\ de Haan', J. VA.N. De bibitziekte in de Deli-tabak veroorzaakt door Phytophthora nicotiarme. pp. 107, pi. 1. 1896. (Meded. Lands Plantentuin, 15). BuRRiLL, T. J. Bitter rot of apples. 111. Agr. Expt. Sta. Bui. 118:553- 608. pis. 10. 1907. Butler, O. Bordeaux Mixture. Phytopath. 4:125. 191 1. Clark, J. F. On the toxic properties of some copper compounds with special reference to Bordeaux mixture. Bot. Gaz. 33:26-48. figs. 7. 1902. Clinton, G. P., and Britton, W. E. Tests of summer sprays on apples, peaches, etc. Conn. Agr. Expt. Sta. Ann. Rpt. 33 ([1910] ll):347-406. pis. 17-24. 1912. Coleman, L. C. Diseases of the Areca palm. I. Koleroga. Dept. Agr. Mysore, Mycol. ser. Bui. 2. pp. 92, figs. 6, pi. 18. 1910. (First pub- lished in abbreviated form in Ann. Mycol. 8:591-626. figs. 4, pis. 7-9. Cr.\ndall, C. S. Bordeaux mixture. 111. Agr. Expt. Sta. Bui. 135:199- 296. figs. 8. 1909. De B.\rv, Anton de. Die gegenwartig herrschende KartotTclkrankheit, ihre Ursachc und ihre Vcrhiitung. pp. 75, pi. 1. 1861. Ueber Schwarmsporen-Bildung bei enigen Pilzen. Ber. Verhandl. Naturf. Gesell. Freiburg i B.. 2:314-329. 1861. -Recherches sur le developement de quelquest champignons parasites. Ann. Sci. Nat. Bot. IV. 20:5-148. pis. 1-13. 1863. (For translation see The potato disease. Jour. Quekett Micros. Club. 22:1.39-145. 1873. Dodge, B. O. Methods of culture and the morphology of the archicarp in certain species of the .Vscobolaceae. Bui. Torrey Bot. Club. 39:139- 197. figs. 2. 1912. *Dufour, Jean. Le traitement du mildiou par le sulfate de cuivre. Vignc Franc. 2:21. 1889. -Note sur Taction du sulfare de cuivre sur la germination de quelques champignons. Landw. Jahrb. Schweiz. 4:97-104. 1890. Eriksson, Jakob, and Henxing, Ernst. Die Getreideroste. 73. 1896. Faes, H. — Sur quelques recherches concernant le developpement ct le traitement du mildiou. Rev. Vit. 39:161-165. 1913. 62 Wisconsin Hiisi:AR(;ii Bulletin 37 Farlow, W. G. The potato rot. Bui. Bussey Inst. 1:319-338. figs. 7. 1875. On the grapevine mildew. Bui. Bussey Inst. 1:419. 187fi. Foreman, F. W. The fungieidal properties of liver of sulphur. Jour. Agr. Sci. 3:400-416 December, 1911. Frank B., and Kruger, F. Uber den Reiz, welchcn die Behandlung mit Kupfer auf die Kartoffelpflanze hervorbringt. Ber. Deut. Bot. Gesell. 12:8-11. February 22. 1894. Gerneck, R. Einfluss der Witterung auf das Auftreten der Peron- osporakrankheit der Reben. Weinbau u. Weinhandel .'iO: 199-200. May 4, 1912. GiMiNGiiAM, C. T. The action of carbon dioxide on Bordeaux mixture. Jour. Agr. Sci. 4:69-75. 1911. Gregory, C. T. Spore germination and infection with Plasino/xira viticola. Phytopath. 2:235-249. figs. 7. 1912. IIallier, Ernst. Neue Untersuchung der durch Peronospora infeslans Gasp, hervorgerufenen Krankheit der KartolTeln. Zlschr. Parasitenk. 4:263-286. pis. 5-6. 1875. Halstead, B. D. Peronosporeae and rainfall. Jour. Mycol. 5:6-11. March, 1889. Hawkins, L. A. Some factors influencing the efficiency of Bordeaux mixture. U. S. Dept. Agr. Bur. Plant Indus. Bui. 265. pp. 29, figs. 4. 1912. Haywood, J. K. The lime-sulphur-salt wash and its substitutes. U. S. Dept. Agr. Bur. Chem. Bui. 101. pp. 29. February 28, 1907. Hecke, Ludwig. Untersuchungeniiber Phytophthora infesians De By.als Ursachc der Kartoffclkrankhcit. Jour. Landw. 46:71-74, 97-142. pis. 2. 1898. Der Einfluss von Sorte und Temperatur auf den Steinbrandbefall. Ztschr. Landw. Versuchsw. Oesterr. 12:49-66. pi. 3. February, 1909. Humphrey, J. E. The Saprolegniaceae of the United States, with notes on other species. Trans. Amer. Phil. See. n. s. 17:63-148. pis. 14- 20. 1893. Istvanffi, Guyla de. Mikrobiologische LJntersuchungen uber einigo Krankheiten der Obstbaumc und der ^^'ei^I■el)e. Ztschr. i^llanzcn- krank. 13:241-242. 1903. and PalinkAs, Gy. Infektionsversuche mit Peronospora. Centbl. Bakt. [etc.] Abt. 2. 32:551-564. February 6, 1911. Etudes sur le mildiou de la vigne. Ann. Inst. Gentr. Anipelol. Roy. Hongrois. 4:1-122. pi. 9. 1913. Jaczewski, Arthur von. Studien iibcr das Verliallcn des Schwarzrostcs des Getreides in Hussland. Ztschr. Pflaiizcnkrank. 20:321 -.359. figs. 8. 1910. Jensen, J. L. Moycns de combattre et des detruire le Peronospora de la pomme de terre. Mem. Sov.. Nat. Agr. France. 131:31-156. 1887. Johnson, E. G. Cardinal temperatures for the germination of uredo- spores of cereal rusts. Phytopath. 2:47-48. 1912. (Abstract. j Lati-j Blicht of P()TA'rf)i:s 63 JoNiiS, L. H. Disease resistance of potatoes. U. S. Dept. Agr. Bur. Plant. Indus. 13ul. 87. pp. 39. 1905. GiDDiNGS, N. J., and Lutman, B. F. Investigations of the po- tato fungus Phijtophthora infcstans. U. S. Dept. Agr. Bur. Plant. Indus. Bui. 245, pp. 100, pis. 10. 1912. Klebahn, IIeinrich. Krankncitcn des Flieders. pp. 75, figs. 45. 1909. KusANo, L. On the life-history and cytology of a new Olpidiuni with special reference to the copulation of motile isogametes. Jour. Col. Agr. Iniper. Univ. Tokyo. 1:141-199. fig. 1, pis. 15-17. 1912. Livingston, B. E. On tne nature of the stimulus wnich causes the change of form in polymorphic green algae. Bot. Gaz. 30:289-316. pis. 17-18. 1900. (Contril). Hull Bot. Lab. 22.) Lutman, B. F. Potato diseases and the weather. Vt. Agr. Expt. Sta. Bui. 159:248-296. 1911. MgAlpine, Daniel. Some points of practical importance in connection with the life-history stages of Phijtophthora infcstans (Mont.) De Bary. Ann. Mycol. 8:156-166. pi. 2. 1910."^ Melhus, I. E. Experiments on spore germination and infection in certain species of Oomycetes. Wis. Agr. Expt. Sta. Research. Bui. 15:25-91. figs. 10. 1911. Culturing of parasitic fungi on the living host. Phytopath. 2:197-203. figs. 2. pis. 20. 1912. Millardet, Alexis. Observations nouvelles sur le developpement et le traitement du mildiou. Jour. Agr. Prat. 50(2):664-665. 1886. and G.woN, Ulysse. Researches nouvelles surl'action que les preparations cuivreuses execent sur le Pernospora de la vinge. Jour. Agr. Prat. 51(1):123-129. 1887. Morse, W. J. Two recent epidemics of late blight and rot of potatoes in Aroostook county. Maine Agr. Expt. Sta. Bui. 169:165-184. figs. 15-16. 1909. MuLLER, Hermann (Thurgau). Ifektion der Weinrebe durch Plasmopar viticola. Centbl. Bakt. [etc.] Abt. 2. 29:683-695. fig. 1. 1911. Orton, C. R. Disease resistance in varieties of potatoes. Proc. Ind. Acad. Sci. 1910:219-221. 1911. PfiE-LABY, E. Sur quelques efTets de parasitisme de certains cham- pignons. Rev. Mycol. 21:77-78. July, 1899. Pethybridge, G. H. Investigations on potato diseases. Dept. Agr. and Tech. Instr. Ireland Jour. 12:334-360. figs. 5. 1912. *Prevost, I. B. Memoire sur la cause immediate de la carie ou charbon des bles et de plusieurs autres maladies des plantes. pp. 80. pis. 3. 1807. Prillieux, Edouard. Maladies des plantes agricoles. 1:78. 1895. Ravaz, L., and Verge, G. Sur le mode de contamination des feuilles de vignc par le Pzasmopara viticola. Compt. Rend. Acad. Sci. [Paris]. 153:1502-1504. 1911. Infiuence de la temperature sur la germination des conidies du mildiou. Prog. Agr. et Vit. 29:170-177. figs. 2. 1912. Reddick, Donald, and Wallace, Errett. On a laboratory method of determining the fungicidal value of a spray mixture or solution. Science n.s. 31:798. 1910. 64 Wisconsin Research Bulletin 37 Reed, H. S. Does Phytophthora infestans cause tomato blight? Phyto- path. 2:250-252. 1912. RuHLAND, W., and Faber, F. C. von. Zur Biolgoie der Plasmopara viticola. Mitt. K. Biol. Anst. Land. u. Forstw. 8:19-21. April, 1909. RuMM, Christian. Ueber die Wirkung der Kupferpraparate bei Bekampf- ung der sogenannten Blattfallkrankheit der Weinrebe. Ber. Deut. Bot. Gesell. 11:79-93. 1893. Ruth, W. E. Chemical studies of the lime-sulphur-lead-arsenate spray mixture. Iowa. Agr. Expt. Sta. Res. Bui. 12:109-419. June, 1913.' Salisbury, R. D. Physiography. 356. 1908. *ScHAGHT, Herman Bericht an das Konigliche Landes-Oekonomie- Collegium iiber die Kartoffelpflanze und deren Krankheiten. pp. 29, pis. 10. 1856. ScHAFFNiT, Ernst. Biologische Beobachtungen iiber die Keiirifahigkeit und Keimung der Uredo- und Aecidiensporen der Getreideroste. Ann. Mycol. 7:.509-523. fig. 1. 1909. ScH.\NDER, Richard. Uber die physiologische Wirkung der Kupfer- vitriolkalbruhe. Landw. Jahrb. 33:517-584. 1904. Scribner, F. L. — -Report on the fungous diseases of the grape vine. U. S. Dept. Agr. Bot. Div. Bui. 2:10. 1886. Selby, a. D. On the occurrence of Phgiophthora infeslans Mont, and Plasmopara ciihensis (B. & C.) Humph, in Ohio. Ohio Nat. 7:79-85. 1907. SoRAUER, Paul. Handbuch der Pflanzenkrankheiten. 2:154. 1908. [Aull. 3.] Stewart, F. C. and French, G. T. A comparative test of lime sulphur, lead benzoate, and Bordeaux mixture for spraying potatoes. N. Y. State Agr. Expt. Sta. Bui. 347:75-84. 1912. Swingle, W. T. Bordeaux mixture: its chemistry, physical properties, and toxic effects on fungi and algae. U. S. Dept. Agr. Div. Veg. Path. and Phys. Bull. 9. pp. 37. 1896. VOCIITING, Hermann. Ueber die Keimung dor Kartoffelknollcn. Bot. Ztg. Abt. 1. 60:87-114. pis. 3-4. 1902. Viala, Pierre. Les maladies de la vigne. pp.93. 1893. [Ed. 3.] Wallace, Errett, Blodgett, F. M., and Hesler, L. R. Studies ol Ihe fungicidal value of lime-sulphur preparations. N. Y. Cornell Agr. Expt. Sta. Bui. 290:163-307. figs. 79-80, pi. 1. 1911. Ward, II. M. Illustrations of the structure and life history of Phifloph- thora infestans, the fungus causing the potato disease. Quart. .lour. Micros." Sci. [London], n. s. 27:413-425. pis. 31-32. January, 1887. Disease in Plants. 150. 1901. Whetzel, H. H. Onion blight. N. Y. Cornell Agr. Expt. Sta. Bui. 218:139-161. figs. 17. 1904. WUTHRICH, E. Ueber die Einwirkung von Metallsalzcn und Sauren auf die Keimfiihigkeit der Sporen einiger der verbreitctsten parasitischen Pilze unserer Kullurpllanzen. Zlschr. Pllanzenkrank. 2:16-31, 81-94. 1892. Research Bulletin 38 December, 1915 The Control of Cabbage Yellows Through Disease Resistance L. R. JONES AND J. C. OILMAN AGRICULTURAL EXPERIMENT STATION OF THE UNIVERSITY OF WISCONSIN MADISON. WISCONSIN CONTENTS Page The cabbage industry in Wisconsin 1 The various cabbage diseases 4 Black rot; soft rot; club root; black leg; cabbage yellows. 5 Yellows distinguished from black rot 9 Season of attack 10 Climatic conditions 10 Symptoms and progress of the disease 10 The organisms 11 Cabbage yellows in Wisconsin 11 Occurrence 11 Destructiveness 12 Introduction and spread 12 Persistence 14 Control measures previously recommended for yellows 15 Trials of certain control measures in Wisconsin 16 Seed and seed bed 16 Seed disinfection 16 Selection of seed bed 17 Crop rotation 17 FertiUzation 19 Soil Analysis 24 Soil disinfection 26 The possibilities of control through disease resistance 31 A comparison of commercial varieties as to resistance 31 The development of disease resistant strains by selection 37 The plan outlined 37 Initial head selections of 1910 and seed-growing in 1911 38 Trials in 1912 and 1913 of these first generation, selected head strains 41 Seed-growing in 1913 from the second generation heads selected in 1912 47 Trials in 1914 of the second generation, selected head strains.... 57 Some other questions of practical or fundamental interest 59 The nature of disease resistance 59 Disease resistance in relation to seed production 59 Comparison of first and second generations as to disease resist- ance 60 Will disease resistance remain constant in different localities.... 61 Locality of seed-growing in relation to disease resistance 62 General conclusions 62 Summary 64 Literature 69 The Control of Cabbage Yellows Through Disease Resistance L. R. JONES and J. C. OILMAN The Cabbage Industry in Wisconsin Cabbage-growing as a specialized industry has assumed considerable proportions in Wisconsin. In Racine county it began on a commercial scale some thirty years ago with a small acreage by one man, according to Moyle (1913), II has since continued to develop in southeastern Wisconsin, notably in Racine and Kenosha counties, until in certain sections it has become the dominant cash crop. An even earlier beginning in commercial cabbage culture was made in the vicinity of Green Bay^ Here also the local success stimu- lated growers in nearby towns to take up the crop. This has proved highly profitable, especially westward of Green Bay in Outagamie county where Shiocton and neighboring places are large shipping points. Most Wisconsin growers produce winter cabbage chiefiy for storage and for the southern mar- kets. With the development of kraut factories a few centers have given increased attention to growing the summer or kraut types for local manufacture, e. g., in Grant, La Crosse, Racine, Rock, Sheboygan, and Waupaca counties. Cabbage growing in a somewhat experimental way is now being under- taken in various other sections of the state and it seems that with the reclamation of marsh lands attention to cabbage culture is certain to increase. There are no full or reliable statistics available as to the yearly production of cabbage in Wisconsin. The Annual Reports of the Wisconsin State Board of Agriculture give yields by counties for the six years 1907-1912. These are, however, based upon the returns sent in by the county clerks ' George B. Smith of Green Bay has kindly reported on conditions in Brown and Outagamie counties. He writes that his father began shipping cabbage from Brown county in 1867, and that the industry has had a steady growth ever since; also that in the Shiocton region, Outagamie county, cabbage culture on a com- mercial scale began over twenty years ago and the acreage there is increasing each year. 2 Wisconsin Research Bulletin 38 and are quite irregular and evidently inaccurate. No figures are given for cabbage before 1907 or since 1912. The yields, in tons, of the five leading counties in 1911, the last year when complete lists are given in these reports, are as follows: Outagamie, 20,630; Milwaukee, 5,041; Racine, 3,461; Eau Claire, 3,397; St. .Croix, 3,188; Biown, 2,789. Kenosha doubtless should be included in this list, probably about equalling Racine, but foi some reason no report was sent from that county. The acreage and yield from Milwaukee county has been generally uniform during this six-year period, but a com- parison of Outagamie and Racine counties shows a significant change in both actual and relative production. 1907 1908 1909 1910 1911 1912^ Tons Tons Tons Tons Tons Tons Outagamie county 220 2,324 6,120 12,037 20,630 37,000 Racine county 12.155 5,144 15,555 2,829 3,461 3,300 From field observations in Racine county we are confident that the above figures considerably understate the produc- tion there^ and it is quite possible that they do for Outagamie county also. It is probably safe to assume, however, that they indicate for each county approximately the relative pro- duction year by year. These figures show two peculiarities of the cabbage crop, (1) its intensive development in certain localities, and (2) its fluctuation, with the possibility of rapid decline after a few years. The reasons for local specialization in cabbage production are to be found in part in the fact that the cul- tural and marketing methods are somewhat pecuhar, and in part in the further fact that cabbage is best suited to certain soil types (Fig, 1). Although it will grow on any good soil, it thrives best on reclaimed swamp land or similar deep rich land with abundant humus, and, in addition, it requires high manuring. Since, under such favorable conditions, it frequently brings an unusually high return in proportion to the labor investment, the temptation is strong, upon ground ' For 1912, only acreages were given in the board's report and from these we calculated the yields, assuming that they would be approximately the same per acre as in 1911 when both acreage and yield were given. ' We also sought figures from the railroad freight departments showing annual car shipments. These were not available for 1912 or earlier but for 1913 they indicate a shipment probably totaling 1,.')00 cars from stations in Racine county, a figure so much in excess of the report of the Board of Agriculture for the pre- ceding years as to indicate that the figures of the board are probably much below the actual production Control of Cabbage Yellows liir. 1. — A I'UWIM 1 Aiil.h i.AliHAGE FIELD Cabbage is a highly profitable crop when grown on rich, deep black, soil. This field was recently drained and plowed for the first time. suited for cabbage, to grow it on the same soil in successive years. Not infrequently in the earlier days, Racine cabbage growers continued to grow cabbage on a field for from five 4 Wisconsin Research Bulletin 38 to ten years and at Green Bay even longer cropping is prac- ticed without rotation^. In such cases, the fertihty is usually kept up by frequent dressings with stable manure. Neverthe- less the outcome in the Racine sections has been that sooner or later these fields have tended to become "cabbage sick" so that instead of the uniformly profitable crops, partial or complete failures occurred (Fig. 2); hence the extreme iluctuations already noted and the tendency to abandon cabbage in Racine county. The first complaints of such failures came in 1895-1896 from the Racine growers where Dr. Russell (1898), upon investigation, found the cause to be the bacterial disease, black rot. This malady has con- tinued to plague the cabbage fields of that vicinity and now is found over the state generally. Meanwhile, other serious parasitic diseases have been introduced and are rapidly spread- ing. In short, growers are finding that the cabbage crop is no exception to the rule that intensive culture and continuous cropping bring an accumulation of diseases in their train, and that ability to cope with these diseases becomes the deciding factor in continued success with the crop. The Various Cabbage Diseases There are a number of serious diseases which may attack the cabbage in Wisconsin. All are due to parasites which when once introduced persist in favorable soil. Although the present publication will deal with only one of these, the so-called yellows disease, it has been necessary to give con- sideration to all of them in connection with these investiga- tions, and it will be helpful to describe briefly some of the others in order to forestall possible confusion. These include, as of major importance, black rot, soft rot, club root, and black leg, any one of which under certain conditions may ruin a crop. Besides these, there are some minor diseases not requiring mention here. Black roi. This is due to a bacterial parasite, Pseude- monas campcstris (Pam.) Smith, (Bacterium campesire), * We arc again indebted to (ieorge B. Smith of Green Hay for the following statement. "One of my neighbors grew cal)bage continually for seventeen years and then changed to another crop, not because he had poor cabbage, for the cab- bage was fine. This land was, however, located in a valley where the wash from neighboring hills brought some new soil to it each year. I have gradually grown into the practice of rotation of all crops myself so as to prevent sickness. I think four or five years is as long as I have ever planted cabbage continuously on level land and the last year it showed some yellows." Control of Cabbage Yellows FIG. 2— THE BEGINNING OF YELLOWS This was the second successive crop of cabbage on new If^d Note the ol^d cabbage stumps on the ground. The plant in the foreground is healthy but the next etght in tliis row all showed -yellows" Note dwarhng =*"d shedding ot lower leaves. Probably a single plant in this p ace. Possibjy one of these stunijj.w^^ diseased the previous year. Spots like this scattered throughout the heia. mean the end of profitable cabba.ge culture on this land. 6 Wisconsin Research Bulletin 38 which finds its way into the plant at the leaf margins or through the vascular system of the plant and causes a black rot which starts with the veins. This seems to be carried on the seed according to Harding, Stewart, and Prucha (1904) and also on seedlings and is widespread in the state. Russell, as already noted, found it common and highly destructive in southeastern Wisconsin in 1895-1896, when it was the chief factor in the cabbage losses then prevalent in the Racine district. It has also caused similar severe losses elsewhere, but the degree of its destructiveness varies widely with climatic and other conditions. It is more likely to be con- fused with yellows than is any other malady. The yellows and black rot may be distinguished by field characters through the differences in both season and mode of attack. These will be pointed out following the discussion of yellows. The primary measure looking to the control of black rot as shown by Harding and his associates is seed disinfection, a practice quite inapplicable to yellows, as will be shown later. Soft rot. Black rot kills the tissues but does not rot them rapidly. Their destruction is usually hastened and completed by another bacterial disease, soft rot, caused by a distinct organism. Bacillus carotovorus Jones. This ma^^ also follow other maladies or work independently, getting started through wounds, insect punctures, etc. It works rapidly, reducing stem and leaves to a soft vile-smelling mass. This occurs more or less in practically all cabbage fields and is often termed "stump rot" by the growers. Club rooi. Club root is caused by the parasitic slime mold, Plasmodiophora brassicae Wor. It is an introduced pest, but is already widely scattered through Wisconsin, .especially in the vicinity of Green Bay. It becomes evident by the stunting of the plants and their tendency to wilt, but is recognized with certainty by the enormously swollen and deformed roots. The use of lime is the specific remedy, a treatment which our later discussion will show has no value for yellows. Black Leg. Black leg is caused by a fungous parasite, Phoma lingam (Tode) Desm. (Phoma olcracca Sacc.)^ It develops on the leaves, stems, and roots. It spots the leaves, » M. P. Henderson, while working recently in this department, made a special study of black leg and the above statements represent his conclusions relative to the organism and its nomenclature. Control of Cabbage Yellows 7 but this is not a serious matter. The chief harm results when the stems and tap root are attacked near the ground or below, whereupon they blacken and die, hence the name "black leg." With the rotting of the stem, the leaves wilt and the plant slowly perishes. This disease occurs quite generally over the state, but has proved most destructive in La Crosse county. Seed disinfection and sanitation have proved im- portant measures in its control. Cabbage yellows. This is the disease with which the present publication is primarily concerned. It is widely scattered in Wisconsin, at least as far north as Green Bay and west to La Crosse. It is caused by a soil fungus, Fusarium conylutinans Wollenw. Cabbage yellows has been known to American phytopathologists for nearly two decades, but has not been recorded from any other country. It was first ob- served by E. F. Smith as occurring seriously in the eastern United States in 1895 although he did not publish his records until some years later (1899, a, b). Later Orton and Harter (1909, 1912) continued work upon it in the southeastern states and Manns (1911) published upon it from Ohio. Work was begun in Wisconsin in 1910. Comparative study of the organism and its description under the name Fusarium conglutinans was made by Wollenweber (1913)'^. Fusarium attacks the roots under favorable conditions, either in the seed bed or within a short time after trans- planting. The attacked plants are stunted and the foliage assumes a pale, lifeless yellow. Sometimes the plant is uniformly attacked, more often the symptoms appear earlier and continue worse on one side, and this one-sided check results in a lateral warping or curving of stem and leaves (Fig. 3). The trouble apparently begins, however, with the invasion of the fibrous roots, and from these it passes to the stem tissues showing first in the vascular bun- dles. These appear simply water soaked to begin with, but soon darken and finally become brownish-black in the later stages, while the cortical tissues overlying them gradually die and collapse. These invasions of root and stem result in diminution of water and food supplies from the soil and hence 8 Since the genus Fusarium is a most complex one and Wollenweber's descrip- tion is meager, it may be worth recording that in a letter, July 2, 1913, he states "I based the Fusarium conglutinans on your (Wisconsin) material because it gave me the first normal conidia and was identical with a strain from Dr. E. F. Smith and one from Mr. Harter." Wisconsin Research Bulletin 38 FIG 3— CABBAGE SEEDLING ATTACKED BY FUSARIUM large. Control of C'.AiiBAcii-: Yellows 9 the foliai,'e symptoms noted above. Meanwhile, the funi^us passes into the vascuhir system of the upper parts, stem and leaves. The invaded plants begin early to shed their lower leaves while making a weak attempt to continue growth above. Death may result in the worst cases within a week or so after transplanting. The majority of the diseased plants continue a sickly existence for a month or more, then suc- cumb; a few less severely attacked, live through the summer, but rarely forms heads (Fig. 4). FIG. 4.— CABBAGE YELLOWS, LATER STAGES Where the attack is not too severe or the plant is somewhat resistant, the plants may continue a sickly existence through the season. Such plants are yellowish and the lower leaves keep dying and falling. The attack is often worse on one side, warping or curling the stems. Yellows Distinguished from Black Rot As already noted, black rot is the only disease liable to confusion with yellows. The two may occur in the same field. More often, however, according to our observations and the testimony of experienced growers, they appear at their worst in separate seasons or situations. While this is in part a result of the accident of introduction, it is largely because the conditions favorable to one are repressive to the other. The diseases differ in these respects as follows: 10 Wisconsin Research Bulletin 38 Season of attack. Both may appear in the seed bed, but the yellows more conspicuously. Following transplanting the yellows appears promptly, some evidence of the disease being seen within a week and it is at its height within three weeks, providing climatic conditions favor. Thereafter, there are few new infections. The black rot does not appear seri- ously until the plants begin to head up; thereafter it spreads rapidly, and continues its development into the autumn. FIG, 5.— BLACK ROT STARTING ON THE LEAF Diseased areas (B) unshaded except blackened mesh of veinlcts. (A) I'olc eaten by insects. Disease introduced at this point, and spreading l>ackward to main rib. (C) Blackened veinlcts affected l)y disease. (D) Water pores through which disease gcntis gain a foothold, producing marginal infection, (.\ftcr Rus- sell.) Climaiic <'on While warning against reliance upon seed disinfection as a chief safeguard from Fusarium, especially in the older cabbage centers, we wish at the same time to emphasize our belief in its importance as a general preventive measure against ciibhiige diseases, especially black rot and black leg. We, therefore, recommend the disinfection of cabbage seed as follows: soak the seed for 20 minutes in a solu- tion of 1 part of standard formaldehyde (40 per cent solution) in 250 parts of water (1 oz. in 2 gals), then wash in water, and dry. This practice was first recommended by Harding, Stewart, and Prucha (1904) for black rot control and is in general accord with the advice of later writers, Manns (1911) and Harter (1912). Its merits were recently proved by the results of trials by M. P. Hen- derson in the investigations of black leg already referred to. Control of Cabbage Yellows 17 Selection of seed bed. The importance of the selection of new soil each year for the seed bed deserves to be empha- sized most strongly. Examples have frequently come under our observation of even experienced cabbage-growers con- tinuing their seed bed on the same soil year after year. The inevitable result is that one or more of the cabbage parasites is soon established. This applies especially to black leg, club root, and yellows. The arguments that follow for rotation in the field apply even more emphatically to the seed bed. The safe rule is to make the seed bed each year on soil which has never grown cabbage before. Here again, however, we stop short of the fmal solution of the problem since sound plants from a healthy seed bed contract the disease promptly when transplanted to an infected field. The control of the disease after transplanting is the final essential to success and it is to this that most attention has been given in these studies. CROP ROTATION Since the Fusarium is a soil parasite, the possibilities of crop rotation naturally received early consideration. At the outset a careful canvas was made among experienced cabbage-growers working under a variety of soil conditions with reference to their experience upon this point. The results showed with practical unanimity that when once a soil be- comes thoroughly infested, i. e., cabbage sick, no reasonable rotation will eliminate the fungus. These conclusions have been further confirmed in connection with our experiments. We secured abundant evidence from farmers that the disease will appear seriously upon such sick soil which has been used for crops other than cabbage for from three to eight years. For example, Benjamin Bones, an expert truck- grower who had the longest experience in cabbage-growing of any man in the Racine district, stated that after his soil was once infested rotation did not as a rule rid it of the fungus. In one case, he seeded cabbage-sick soil to clover and timoth}^ let it lie thus for five years, broke it up, re- planted to cabbage, and did not get a sound head from the entire five acres; indeed, he not only sold none, but was obliged to get cabbage from his neighbors for the use of his own family. At this time, however, Mr. Bones did not de- 18 Wisconsin Research Bulletin 38 termine whether the disease was black, rot or yellows. From his description, and from subsequent observations on his farm, we believe it was yellows. In 1910, in order further to test out this matter, we planted experimentally on this same "farm a piece of cabbage-sick land on which no cabbage had been grown for at least six years. During this time it had been in grass for three years. The outcome was that more than one-half of the stand died of yellows and there was considerable disease in the balance of the crop. In only one case in Mr. Bones' experience had he succeeded through long rotation. This was where cabbage-sick soil was put into strawberries for two years, then seeded to grass for five years. At the end of the seven years he broke it up, decided to again try cabbage and "to his surprise" got a good crop with no serious amount of yellows. The fact that there was a little of the disease in this case, however, shows that only one crop of cabbage could be secured even then, following which a return must be made to other crops for another long period. A. L. Curtis, another Racine grower, said that when forced by cabbage-sickness to discontinue cabbage culture on a piece of land, about twelve years previously, he raised good truck crops of other kinds, onions, beets, and potatoes, but cabbages tested twice during the twelve years had shown the disease in the soil without evident abatement. In still another case where cabbage failed from disease, the field was seeded down for fourteen years, then plowed, in 1914, and planted to cabbage. The result was a large per- centage of loss from yellows (Fig. 7). While none pf these cases carries the conviction of experi- mental evidence, as there is always the possibility of reinfec- tion of such soil in the interim, nevertheless the cumulative evidence has convinced us that the fungus persists indefinitely in the soil. Smith (1899 a) reports that the disease reap- peared when cabbages were planted on soil from a sick field which had been kept in dry storage for over three years. While, therefore, we were forced to abandon the idea of crop rotation of any reasonable length as insuring complete recovery of the cabbage-sick soils, we were convinced that when beginning with healthy soil in a region liable to yellows, attention to proper rotation will prolong the usefulness of CONTHOL OF CaBBAGK YeI.LOWS 19 the soil for cabbage culture. As already noted, it is not un- common in cabbage-growing sections to find cases where cabbage is grown continuously on a field for from five to ten years and even longer. While this may sometimes be done with success, disaster is almost inevitable as soon as the Fusarium or, indeed, any other soil parasite is introduced. The more rational method for long-time practice is to grow cabbage not oftener than once in three years in rotation with other crops. FERTILIZATION While little evidence favorable to crop rotation as a spe- cific remedy for cabbage-sick soil was received from the growers frequent testimony favored one or another mode of fertilization. This seems natural, since it is known that the cabbage is one of the rankest feeders among cultivated crops, profiting from enormous applications of stable ma- nures and responding well on some soils to commercial fer- tilizers, especially to potash salts. The use of lime, gypsum, ashes, and common salt was urged as helpful by one or an- another grower and one man testified to the benefits of a complete commercial fertilizer. Harter (1909) has shown that cabbage is sensitive to malnutrition following improper fertilization. It seemed reasonable to suppose that even with the liberal use of stable manures, an unbalanced con- dition of soil fertility might obtain follow'ing prolonged cab- bage culture which might in turn predispose toward disease. There was also the specific suggestion from the work of Hart and Peterson (1911) that there might be a deficiency in sulfur. Cabbage tissue is especially rich in this element and they show that an average crop of cabbage will remove about 100 pounds of sulfur trioxide from the soil. Such con- siderations led us to plan a comprehensive series of fertihzer trials designed to cover fully the possibilities suggested. FERTILIZER TRIALS, 1910'2 A careful survey of the older cabbage-growing section where the yellow disease prevails was made in the spring of 1910 and areas of cabbage-sick land were selected on four farms for the fertilizer trials. •2 Before this work was undertaken, A. J. Rogers, of the horticultural depart- ment, had become interested in the question and lie cooperated in these trials the first year. 20 Wisconsin Research Bulletin 38 Field I. On the farm of Witcheber Brothers, Kenosha. Soil — black alluvial loam, tile drained, received 15 tons stable manure per acre previous autumn. Cabbages on land the year before showed some disease. Applied fertilizer on one- fourth acre plots at the following rates, with controls on either side untreated. Applications of fertilizer made on surface of cultivated field May 27, field subsequently disked, and cabbage planted June 25. Plot 1. Potassium chloride 300 pounds per acre Plot 2. Acidulated bone meal 450 pounds per acre Plot 3. Potassium sulfate 150 pounds per acre Plot 4. " " 300 pounds per acre Plot 5. Calcium sulfate 500 pounds per acre Field II. On farm of Matt. Broesch, Kenosha. Soil — strong clay loam, tile drained in high state of fertility. Re- cent history of field as follows: For preceding 15 years cabbage on field about every second year, alternating with either corn or potatoes; during first few years cabbages healthy and yields large; since about 1900 disease trouble- some; some eight years before disease (yellows?) so severe that every head was lost on ten acres; last cabbage grown on here two years before (1908), badly diseased, loss 75 per cent; last year (1909) potatoes. Light applications of stable manure nearly every year; about 15 loads per acre in 1908; none in 1909; about 15 loads per acre again in winter 1909- 1910 and plowed under in early spring. For our experiments fertilizers at the following rates were applied on the surface and disked in on May 27; cabbages planted June 25. Plots 2, 6, 8, control, no fertilizer added, balance treated as follows: Plot 1. Potassium chloride 250 pounds per acre Plot 3. Potassium sulfate 250 jiounds per acre Plot 4. Acidulated bone meal 250 pounds per acre Plot 5. Potassium sulfate 250 pounds and bone meal 250 pounds per acre Plot 7. Air-slaked lime 120 bushels per acre Plot 9. Calcium sulfate (gypsum) 600 pounds per acre Field III. On farm of A. L. Curtis, Berryville. Light sandy soil, good surface drainage; no stable manure for 5 years previously; 125 pounds per acre of commercial fer- tilizer" with sugar beet crop, 1909. 1' Mr. Curtis states that he used Homestead Complete Fertilizer M. & I. Special carrying fertilizing elements in the proportions N. 3; P. 8; K. 6. C.ONTHOi. OF Cabbagi-; Yu.i.ovvs 21 In this experiment, the soil was fitted and fertiUzer ap- phed at the following rales, May 27; cabbage planted June 18. Plot 1. Potassium chloride 250 pounds per acre Plot 2. Potassium sulfate 250 pounds per acre Plot 3. Potassium sulfate 250 pounds and bone meal 250 pounds per acre Plot 4. Control (no fertilizer) Plot 5. Acidulated bone meal 250 pounds per acre , Plot 6. Calcium sulfate (gypsum) 600 pounds per acre Plot 7. Air-slaked lime 120 bushels per acre Field IV. On farm of Benjamin Bones. Strong clay loam, surface drainage, fair state of fertility, cabbage last grown on it eight years before. Experimental application of fertilizer at following rates made about June 1; cabbage planted June 20. Plot 1. Potassium sulfate 400 pounds per acre Plot 2. Potassium chloride 400 pounds per acre Plot 3. Control, no fertilizer or manure Plot 4. Stable manure, liberal application Outcome of fertilizer experiments of 1910. The sea- son was favorable for the disease and the yellows began to show in all of these fields, without evident relation to fer- tilization, within two weeks after transplanting. On July 30, when careful notes and counts were made, the condi- tions were as follows: I. Witcheber field: 10 per cent of plants dead and 10 per cent of balance diseased; no evi- dence of difTerence attributable to fertilizer. II. Broesch field: over 90 per cent of plants dead and no evidence of difference in favor of any fertilizer. III. Curtis field, the same. IV. Bones field not visited at this date. In October, 1910, when the final notes were taken upon these plots, there was absolutely no evidence of gain from any type of fertilizer in any field. In the Witcheber field, about one-third of the crop was dead or diseased, the dis- ease appearing worst in irregular areas or spots which stood in no relation whatever to the fertilization. In the Broesch field of one-half acre, only about twelve plants headed up and these were scattered irregularly over the field. Most of these showed symptoms of yellows. In the Curtis field, only one plant headed in the entire field. In the Bones field, the loss was over 50 per cent from yellows, scattered over the field regardless of fertiUzation lines. No encouragement 22 Wisconsin Research Bulletin 38 could be found in these results for the hope that fertiliza- tion would control the disease. FERTILIZER EXPERIMENTS OF 1911 Plan. Although the outcome of the fertilizer experiments of 1910 was purely negative and seemed to show that no method of fertilization under trial had any effect upon the disease, it seemed wise to make similar trials in 1911 upon one field, using heavier applications. Since the Broesch land had proved most seriously diseased and was by uniformity and other conditions admirably suited for experimental work, it was decided to repeat the same applications on the identical plots. This was done May 5, 1910, except that the gypsum on plot 9 was added at the rate of 1200 pounds per acre, twice as much as in 1910. The fertilizers were harrowed in and the ground lay until planting time. On June 20, before planting, the field was disked and on one- half of each fertilized plot, Nos. 1, 3, 4, and 6, the original application was repeated. The result was that in the case of each commercial fertilizer under trial there was the single application made in 1910, repeated in 1911 on one-half of the same plot and doubled on the other half. With the slower-acting gypsum and lime it was considered that the single heavy application each year would do all of which these chemicals are capable. The soil was uniform and owing to the general occurrence of the disease in the crop of the preceding year, it was known to be uniformly infected unless the chemicals were exerting an influence. Results of fertilizer trials, 1911. The cabbages were planted June 20. A period of dry, hot weather such as most favors the disease, followed, thus giving ideal conditions for a severe, but fair, test of the efficacy of the chemicals. On July 8, examination showed yellows abundant over the entire field with no evidence whatever of any difference either in the diseased condition of the plants or in their general vigor except that in the portion of the potassium chloride plot which had received a double application the plants were very slightly larger. It was doubtful, however, if there was really less disease even here. There seemed, however, to be a liltle less disease on the limed plant. (Fig. 8 shows the appearance of the field on July 8.) During Control of Cabbage Yellows 23 the next week the disease worked rapidly on all plots alike and by July 16, 90 per cent of the plants were dead, or practically so, and most of the remainder were diseased. The slight advantage which the potassium chloride plots showed at first was lost, showing that although it gave a little stimulus to growth it had no appreciable influence upon the disease resistance of the plants. The same was true of the lime plot. m^gta^^^^^^^^ H >^SH l^^^^^^HS^^B!^^^^^^^^^^^^H ' M^I^^^^^^H HH^^S Py^- ., . j^^^^^^^^^^^B, FIG. 8.— EXPERIMENTAL FIELD, 1911, BROESCH FARM This soU was thoroughly "cahbaae sick" through Fusarium infestation. Practi- cally every plant in it was attacked by yellows in spite of applications on different sections of lime, gypsum, various commercial fertilizers, sulfur, and soil disin- fectants as described in the text (see summary Table 3). That this soil is fertile and in fine tilth was shown by the fine stand of potatoes at the left and of corn and eats at the right. A photograph of the latter crops is shown in Figure 9. Figure 1 shows a neighboring healthy field of the same day. At the end of the season, in October, the outcome was fully convincing. Only 8 plants were alive on the entire series of plots where over 3,000 were started, and only one of these formed a head. This was upon the potassium sulfate plot, but its location was evidently purely accidental, its survival being attributable to its individual disease-resistant qualities rather than to the fertilizer. possible INFLUENCE OF SULFUR As already explained, especial consideration was given to the idea that sulfur-depletion of the soil might be a factor in 24 Wisconsin Research Bulletin 38 predisposing the crop to yellows. It would seem that the sulfur compounds used in the preceding trials, potassium and calcium sulfates, should have had a beneficial elTect if sulfur were lacking. No such benefits followed their use, however. The possible influence of sulfur was further tested by applications, on two fields, of flowers of sulfur, 500 pounds per acre. This was made in connection with the trials of soil disinfectants, to be discussed later; but this again was absolutely without effect. SOIL ANALYSES Further evidence bearing upon the relation of fertilization to the disease was furnished by chemical analyses of the soil from two of these experimental fields, made by the depart- ment of agricultural chemistry. ^^ For this purpose, in 1910, samples of the soil were collected from the middle of the cabbage-sick fields on the farms of A. L. Curtis, Berryville, and M. Broesch, Kenosha. These were the fields selected for the 1910 fertilizer trials already described and of which the history has been given. We also secured similar samples of virgin soil from the permanent fence rows near by, which the owners in each case said had never been cultivated or cropped. In securing these, the first two inches were rejected and the underlying soil taken to a depth approximating six inches. As before explained the cultivated soil in both fields had received liberal appli- cations of stable manure in previous years and on the Curtis field commercial fertilizer had been used each of the last three years. The percentages of sulfur trioxide found were as follows: (Airtis field, virgin soil 0.108 per cent, cultivated soil 0.115 per cent. Broesch field, virgin soil 0.119 per cent, cultivated soil 0.140 per cent. This chemical analysis justifies the conclusion that the sulfur content of these soils "was maintained and even slightly increased" by the liberal application of farm manures and fertilizers. Sulfur depletion could not, therefore, be the cause of the cabbage sickness. '* We are indebted to E. B. Hart of the department of asricullural chemistry for advice in the collection of these samples as well as for the supervision of the soil analysis recorded above. These results arc (iiiotcd from the [jublication already referred to. Wis. Exp. Sla. Res. Bui. 14. p l.i. C.ON'iHoi. oi" (".ABnAf.i: \'i:i.i.()\vs 25 CONCLUSIONS SUMMARIZlin These results seemed fully conclusive. The trials were made upon four representative types of cabbage-sick soil; they extended over two seasons, involving in some cases repeated applications, and included the following chemicals: potassium chloride, potassium sulfate, phosphates from both rock and acidulated bone, calcium sulfate (gypsum), calcium hydrate (lime), and sulfur. The yellows disease was abundant on all the fields both seasons, and its occurrence FIG. 9— CABBAGE SICK SOIL IS NOT LACKING IN FERTlLll V Excellent com and oats grew on the land bordering the experimental field of 1911. The picture was taken .luly 8. Note that the cabbage plants are all dying from yellows. (Compare Figure 8.) was not influenced in any way by any fertilizer. Moreover, chemical analyses of sick and neighboring healthy soils showed no evidence of the depletion of the former in sulfur. The conclusion seems clear that the trouble was simply due to the occurrence of the parasitic Fusarium in the sick soils and thai the api)lication of fertilizers did not influence the inter-relation between the host plant and the fungus parasite. 26 Wisconsin Research Bulletin 38 SOIL disinfection In view of these facts the question next pertinent relates to soil disinfection. One who has had experience with soil inhabiting parasites knows how difficult it is to dislodge them by any mode of soil treatment which is practicable under field conditions. Nevertheless, from the outset of these in- vestigations efforts were made to do this. The nature of this work and the conclusions reached will be briefly summarized. :--^, -^ fig. 10.— steam STERILIZATION PREVENTS YELLOWS 'rhe soil in pot 2 was as taken from an infected cal)bage field. Cabbage seed was planted in it, under greenhouse conditions, and all the seedlings promptly developed Fusarium yellows and most of them died. Pot 1 contained like soil but was autoclaved U hours at 11 jiounds [jressure. Steam sterilization. F. D. Bailey (1912) while working in our laboratories, made trial of steam sterilization of cabbage-sick soil and found that it was efficacious in pre- venting yellows. This method has been employed frequently in our greenhouse trials, (see Fig. 10). Here would be a sure remedy if practicable. It is not feasible, however, to practice steam sterilization of soils for cabbage culture in Wisconsin even for the seed-beds. It might seem at first that Control of Cabbage Yellows 27 the methods of steaming seed beds recently shown by Johnson (1914) to result so advantageously for tobacco culture in Wisconsin would be applicable to cabbage. There are two difTiculties in the way of this, either of which would be fatal in practice. (1) Wisconsin cabbage-growers plant their seed beds in rows wide enough apart to permit clean culture of the seedlings in order to get the strong plants desired for machine setting. This requires seed fields alto- gether too large for steam-sterilization. (2) The chief prob- lems in lighting the yellows are not concerned with seed bed conditions. It is true that a contaminated seed bed is very serious where it occurs, but we have found the best growers so alert to this fact that upon realizing the danger they have generally been able to avoid it by making the cabbage seed bed on uncontaminated soil, a practice at once simple and effective. If, therefore, soil treatment is to contribute to the control of cabbage yellows it must be applicable to one or the other of two things. (1) To disinfecting the soil in the immediate vicinity of an infected plant. If this could be done inexpensively one might at least lessen the rate of spread of soil infection in the early stages of the disease by pulling the few "yellowed" plants as soon as seen and applying the soil disinfectant. (2) In checking the invasion by the Fusarium of the healthy seedling following its transplantation into infected soil. The evidence, discussed elsewhere, leads us to conclude that this invasion occurs soon after transplanting and is dependent upon certain soil conditions favorable to the fun- gus. It is conceivable that the presence at this stage of some fungicidal element in the soil, especially if close to the cab- bage roots, might so inhibit the fungus as to prevent in- fection. With these possibilities in mind, trial was made of certain fungicidal compounds, first in the greenhouse with the coop- eration of F. D. Bailey, and later in the field. These involved the use of formaldehyde solution, potassium sulfide solu- tion, flowers of sulfur, and a proprietary soil fungicide sup- plied by the Sherwin-Williams Company. Without going over the experimental details, the results may be summarized as follows: 28 Wisconsin Research Bulliitin 38 Formaldehyde. Formaldehyde was applied to cabbage- sick soil in greenhouse Hats 4 inches deep, until the soil was thoroughly wetted. One per cent and 2 per cent solutions were tried. The Hats were covered for twenty-four hours, then left uncovered to air out for three days, and planted to cabbage seed. The effect of the formaUn was to retard slightly the germination of the cabbage and to reduce some- what the percentage of disease in the early stages of its development. Nevertheless, considerable disease developed later, showing that even this drastic treatment was not fully effective. Although these results gave little encouragement, trial was also made in the field as follows: Upon setting healthy cabbage plants in a soil known to contain the Fusarium, one- half-pint of formaldehyde solution was poured into the hole with each plant. In one row a 1 per cent solution was used, in another a | per cent solution. Control row's were planted with water. The result was that the treated plants either failed to start, evidently because of root injury by the chemical, or were less vigorous than the untreated plants. Within three weeks the surviving plants in all rows, both treated and un- treated were attacked by yellows. At the end of the season, out of 150 plants set with formalin solution, only 6 were alive and none formed heads, whereas the corresponding two con- trol rows averaged 12 alive with 1 headed in one row and 2 in the other. (See details in the table at the close of this chapter). It is evident, therefore, that formaldehyde solu- tion is ineffective in checking the Fusarium, whether applied to the entire soil in amount sufficient to inhibit germination of the seed, or about the roots when transplanting, in amount sufficient to be decidedly injurious to the cabbage plants. Potassium sulfide. Much the same methods were used in testing the effect of potassium sulfide. In the field trials with one row, each plant received one-half pint of 2 per cent potassium sulfide about the roots in the planting hole, and another, row^ received a like amount of 1 per cent solution. The injury to the plants was even greater than from the corresponding formalin treatments so that most of the plants failed to establish themselves and renew growth. Those which ('ONTHOI. OK C.ABBAfii: YlCLF.OWS 29 did, however, were all killed by yellows, showing that there is no hope from this treatment. Sulfur. As already stated there was some hope that sulfur might have value as a fertilizing element with the cabbage crop. It is also understood that sulfur has value as a soil fungicide, having been recommended by various investigators in America and Europe as a preventive of potato scab'^ and other soil inhabiting diseases. A heavy- application, at Ihe rate of 500 pounds to the acre, was, there- fore, made on cabbage-sick soil in each of two experimental fields in 1911. This was applied the day before planting. The soil, which was in fine tilth, was first dragged. The sulfur was then broadcasted, and gone over with a plank which covered it well into the surface layer of soil. Within three weeks yellows was showing up generally on both fields with no appreciable difference where sulfur was used. By the end of the first month every plant, control and sulfured alike, on one field was either dead or badly diseased with yellows. On the other field, the disease was abundant throughout, with no difference as to treatment. At this time the sulfur was not only visible when the surface soil was stirred but one could smell it when in the field. At the end of the season very few plants were alive on either field, and there was no appreciable difference attributable to sulfur treatment. In one field, not one plant out of the 400 planted on the sulfur plot lived through the season. In the other, as shown in the later tabular summary, of the 300 planted, only 10 plants lived and none headed, whereas in the control alongside, 14 lived. Clearly, therefore, sulfur had no retarding efTect upon the development of the Fusarium. Soil fungicide. The Sherwin-Williams Company sent a soil fungicide for trial in connection with these experiments. They did not state the composition, but it was a white, finely granular substance having the appearance and odor of naphthalene. They advised its use at 200 or 300 pounds per acre. Knowing that we were dealing with a resistant soil fungus and wishing to be sure, that if this fungicide had any efficacy we might learn it, we applied about twice the I' Halsled (1897) recommended sulphur as the best soil treatment tested for potato scab. Bernhard (1910> states that 350 pounds of flowers of sulfur per acre was successful against potato scab. 30 Wisconsin Research Bulletin 38 amount recommended — 500 pounds per acre. It was used on a plot next the one treated with flowers of sulfur, applied broadcast at the same time and in the same way, on the two cabbage-sick fields. The results were again entirely negative. The application, even at this strength, did not seem to harm the plants in any way. When the yellows appeared, however, it was equally severe on treated and control plots. In the plot treated to the soil fungicide, on one field where 400 plants were set not a plant survived the summer, all being killed by the yellows. On the other the disease was not quite so bad, but the results, as shown in the following tabular summary, were no more favorable for the soil fungicide. The untreated controls on either side averaged practically the same results as did the treated plot. It was evident, therefore, that this "soil fungicide" like the other chemicals tested, was without any influence whatever upon the cabbage Fusarium. In order to place the essential results of the work with these fungicides in brief form for comparison, the following tabular summary is given of the yields on the less diseased of the two fields. As already stated the destruction on the other field was practically complete on all plots. Table I. — Results from Fungicidal Soil Treatments Treatment No. of cabbage plants set out Condition at end of season Plot Living Headed No. Per cent No. Per cent Applied broadcast A Control .•M)() 300 300 300 1500 14 10 14 21 103 5 3 5 7 7 0 0 2 2 12 0 B Sulfur 0 C Control 0.7 D Soil fungicide 0.7 E Control 0.8 Applied in planting hole F Control 150 150 150 150 - 150 22 6 13 4 4 15 4 <) 3 3 1 0 0 1 0.7 G Formaldehyde 0 H Control 1 .3 I Potassium sulfide 0 J Control 0.7 Conclusion. The conclusion seemed clear, therefore, that these experiments furnish no basis for hope that any Control of Cabbage Yellows 31 fungicidal soil treatment will help even in a minor way in the control of this disease. The Possibilities of Control Through Disease Resistance As soon as the disease was well developed in 1910, the first year it was under observation, it was noted that individual plants, even in the most severely infested fields, remained comparatively unharmed by the disease. The possible importance of this was at once apparent since it has been shown by Orton (1900-1909) and BoUey (1901, 1903) that certain Fusarium diseases of other plants can be controlled by the use of disease resistant strains or varieties.'*' With the hope of similar results with cabbage'^, plans were developed to work upon this idea along two different lines involving (1) the comparison of commercial varieties as to relative resistance, and (2) the development of disease- resistant strains by selection. Although the work on these has gone on simultaneously, it will conduce to clearness to discuss the results of the first two years separately. THE comparison OF COMMERCIAL VARIETIES .\S TO RELATIVE resistance In the southeastern Wisconsin district where the yellows occurs only two types of cabbage are grown at all extensively. '• It is a matter of common experience that closely related varieties of culti- vated plants often differ widely in their relative susceptibility or resistance to some fungus attack. Tluis, the Fameuse apple is especially susceptible to apple scab (Venluria iridegualis), the Russets arc resistant; the Transcendent crab and Yellow Transparent are highly susceptible to fire blight (/?. amylouorus). the Mcintosh relatively resistant, 'rhe Wealthy apple may be ruined i)y the rust (Gymnospo- rangium Juniperi-virginianae), while the Oldenburg standing beside it is unharmed. It is only comparatively recently, however, that the importance of disease resist- ance as a factor in the control of plant diseases has received full recognition. Con- siderable progress has been made in Europe and America in securing potato varie- ties resistant to late blight and wheat varieties resistant to rust. The most encour- aging results to date nave, however, been secured with the Fusarium diseases where disease resistance seems to be the hopeful method of attack. 1' With reference to the possibilities of disease resistance in cabbage, the follow- ing reports have come to our attention. Edwards (1907 and 1908) reported that trials in Ontario had shown the variety Houser to be especially resistant to black rot. Dr. F. Kolpin Ravn' stated in correspondence (1912) that he has perfected a strain of turnip resistant to club root. Dr. Ravn also sent us a sample of this seed, but we have not had a favorable opportunity to test its resistance to this disease. Manns (1911) in his publication from the Ohio Experiment Station expressed, hope in the possibility of securing disease-resistant strains; and E. G. Arbzerger, writing us from the same station in 1911. stated that he had observed the variety All .Season to show especial resistance to yellows in Ohio. He kindly sent us samples of the commercial seed from Ohio which were used in our 191 1 trials Close and White (1909) reported from their trials at the Maryland Experiment Station that there was a difference in the susceptibility of cabbage varieties to black rot. and in 1913 Professor White sent us a sample of seed selected for resist- ance to this malady which was included in our trials of 1914 and proved highly resistant to yellows. 32 Wisconsin Research Bulletin 38 These, as already explained, are the winter or storage cabbage and the kraut cabbage. Of each of these types more than one variety or race is used. Over 90 per cent of the acreage is given to winter cabbage and less than 10 per cent to the earlier or kraut types. While both types suffer badly from yellows, the winter cabbages generally suffer more severely, largely because they are planted later and so are likely to pass through a period of trying weather immediately there- after. Inquiry among experienced growers showed that in practice no reliance was placed upon any variety or strain then in use as peculiarly resistant to disease. The belief was expressed by some, however, that the increasing loss of winter cabbage from disease in recent years over former years was because the seed formerly secured from Europe was of a hardier type. Accordingly, the cooperation was secured of Dr. F. K(/)lpin Ravn, Plant Pathologist of Denmark, in arranging for the selection and importation of the most promising Danish-grown strains. Along with these, trial was made in 1911 of the two standard commercial strains of Hollander or Danish Ball Head in general use in this district and also of the summer or kraut varieties of most promise. TRIALS OF 1911 Ferry's strain of Hollander Ilansche's strain of Hollander, grown in Pnget Sound linporlcd Danish No. I (Amager long stem) Iniported Danish No. H (Amager long stem) Imported Danish No. HI (Amager short stem) Imported Danish No. IV (Amager short stem) Ini|K)rted Danish No. V (An^iigc short stem) 1 louser All Season Trials of these varieties and strains were made in two fields, l)oth having soil thoroughly cabbage sick with Fusa- rium. The climatic conditions favored a serious attack of the disease, making liie trial a severe one. The results follow. Field A. — llansche farm. Within three weeks after transplanting, the disease was prevalent throughout the experimental field so that it was estimated that from 75 to 90 per cent of the plants were either diseased with yellows or already dead. At this time the Puget Sound strain was showing up better than any of the imported strains yet it was Control of Cabbaok Yf;i.i.o\vs 33 eslimaled that 75 per ccnl of these plants were diseased and many were already dead. The liouser was judged slightly more vigorous. The evidence in favor of the Plouser increased from this time on and at the end of the season it gave decidedly the best returns, with the Puget Sound the best of the winter cabbage strains. Table II. — Tabular Summary of Variety Trials, Hansche Plot: 1911 Variety or strain Number of plants set out Condition at end of season Row Living Headed Num- ber Per cent Num- ber Per cent 1 2 Fuget Sound Hollander Imported Danish I 1.50 150 150 1.50 150 1.50 150 150 150 150 150 20 0 5 2 14 1 0 58 S 22 13 0 3 2 T 9 0.7 0 39 5 15 1 0 0 0 0 0 0 0 8 3 1 0.7 0 3 0 4 Imported Danish III ... 0 f) 6 7 Imported Danish II Imported Danish II Imported Danish IV 0 0 (' 8 Imported Danish V Ho user 0 10 •> 11 Puget Sound Hollander 0.7 Field B. — Broesoh farm. Fewer varieties were included here and the trial was, therefore, on a larger scale. The dis- ease was even more severe than on the Hansche field. The loss recorded in the following tabular summary was entirely due to the Fusarium. Table III. — Tabular Su.mmary of Variety Trials. Broescii Field: 1911 Variety or strain Number of plants set out Condition at end of season Row Living Headed Num- ber Per cent Num- ber Per cent 1, 5, 13, 17 2, fi. 14, 18 3, 7, 15, 19 4, 8-12, 16. 20-22. Imported Danish I Imported Danish II Imported Danish III .. Ferry's Hollander 1000 1000 1000 2500 2 0 11 5 0.2 0 1 .1 0.2 0 0 1 1 0 0 0.1 0.04 These figures show clearly the extreme destructiveness of the parasite under favorable conditions, since except for this disease the above fields would have shown practically a perfect stand with 95 to 99 per cent heading. 34 Wisconsin Research Bulletin 38 As it was, none of these varieties showed an^^ special degree of resistance, two of the imported, Danish I and II, being inferior to the standard American-grown (Ferry) strain and the other, Danish III, being only slightly superior to it. The two heads which formed were saved in the hope of growing seed from them, but they were not strong enough to survive the winter. Trials of these Danish strains alongside of the American commercial seed were made also in two commercial cabbage fields in the same neighborhood. In both cases there was enough of the yellows to give some data as to relative sus- ceptibility and yet fair commercial yields were secured. None of these Danish strains showed any peculiar excellence as to either disease resistance or yield. TRIALS OF 1911 Similar trials were made in 1912 on the same cabbage-sick soil in which were again included the same strains of im- ported Danish seed, along with a standard commercial "Hollander" strain grown in Puget Sound and the Houser. Trial was also made of the best kraut variety being grown in this neighborhood, the Brunswick. i** The yellows was not as destructive as in 1911 but the comparative outcome was similar. The varieties are grouped in Table IV in the order of resistance. Table IV. — -Outcome of Commercial Variety Trials: 1912 Variety name Source of seed Condition at end of season No. Living Per cent Headed Per cent VI Imported, Denmark Imported, Denmark Imi>orted, Germany 14 36 47 64 24 69 0 II 13 XI Brunswick 21 XII 18 XV Hollander Hansche's PuRct Sound.. Imported, Denmark 24 III 27 TRIALS OF 1913 The same varieties used in 1912 were tried again in 1913 along with several additional ones, including some of the '" For selecting this variety and supplying seed for these trials as well as those of subsequent years we are indebted to F. \V. Gunther, kraut grower and manu- facturer of Racine. This seed was imported by him from Germany. Control of Cabbage Yellows 35 standard early and kraut types. They were planted on the same cabbage-sick soil. While the disease was not as bad as in 1910 and 1911, it was rather worse than in 1912. The full list of varieties and the outcome was as follows: Table V. — Outcome of Commercial Variety Trials: 1013 Variety name Source of seed Condition at end of season No. Living Per cent Headed Percent XVI Hollander Ferry's Commercial Imported, Denmark 30 7 23 12 38 25 32 59 46 29 48 45 41 83 0 VI Large Winter 2 XXI Copenhaaen Market 3 II Amager, long stem Imported, Denmark 4 XVII All season 5 XVIII Succession 11 XV Hollander Hansche's Puget Sound 12 XXII Early Summer 15 XII Houser 16 XX Early Jersey Wakefield Charleston Wakefield Commercial 18 XXIII 18 XI Brunswick Imported, Germany Imported, Denmark 19 III Amager, short stem 22 XIX Volga 48 Conclusions. The results of these three years' trials served to convince us that there are well-marked differences as to disease-resistant qualities between the commercial varities or strains of cabbage. It was evident from the first year, 1911, that the Houser was quite highly resistant and the trials of Volga during 1912 and 1913 have shown that to be even more so^^ (Fig. 11). If it were merely a ques- tion of producing a yield of cabbage regardless of season or quality, either of these would constitute a promising variety from which to breed or select. Unfortunately both of them, at least under Wisconsin conditions, are of the summer type, of no value for either winter storage or kraut manufacture. Our chief problem has been to get a resistant strain of winter or storage cabbage of the Hollander or Danish Ball Head type. Second to this has been the desire for a resistant strain of the kraut type. The results have shown that none of the commercial varieties of the winter type have marked preeminence as to disease resistance, although it is evident that there are »9 The variety Volga, which has shown such a promising degree of disease re- sistance, was first brought to our attention by Dr. J. T. Barrett, then at the Uni- versity of Illinois. He reported (December, 1912) that an Illinois cabbage-grower was succeeding with this variety, growing his own seed, in a district where other varieties of cabbage failed, presumably because of the yellows. 36 Wisconsin Research Bulletin 38 differences worthy of note in the relative siisceptibihty of these. Thus, of the imported Danish seed, the strain desig- nated III Amager, short stem, has proved consistently superior to the others. On the other hand, various field trials, of which we have not here included the details, made of this alongside the standard local types of Hollander, have shown even this one to be on the whole commercially in- ferior to them. With the valuable kraut types, and the early market garden varieties the outcome has shown similarly that there FIG. 11.— THE VOLGA IS THE MOST RESISTANT COMMERCIAL VARIETY TESTED There are two rows of Volga in the middle of this trial field, 1913. The first two rows at the left of the Volga are Puget Sound Hollander; the first two at the right arc Ferry's Hollander, In the trials of 1914 Volga made an even better showing; but unfortunately it is not suited for commercial culture. Figure 20 shows the results of a similar trial on this same field in 1911 when Houser proved the best of the varieties there tested, Volga not being in that trial series. are differences, but none is sufficiently resistant to make its culture safe on cabbage-sick soil. Advantage has been taken of these differences, however, in making selections of resistant heads from the most promising of these varieties. From these we have secured seed for further trial of Volga and the German kraut variety, Brunswick. Trials of these Control of (Cabbage Yellows 37 will be continued and lhroui«h further selections it is hoped that Fusarium-resistant strains may be secured'-". In addition selections from the two standard commercial types of winter cabbage were made in 1910 and trials and further selections made in the succeeding years. The re- sults of these are discussed in the following pages. The Development of Disease-Resistant Strains by Selection THE PLAN outlined Very soon after the inception of the work in 1910 it be- came evident that this was a promising line of attack. As fig. 12. — resistant plants on uniformly diseased soil It is characteristic of a Fusarium infected cabbage field that the individual plants vary widely in susceptibility, some appearing quite immune and maturing perfect heads. (Hansche field. Racine, Sept. 1910, from which selections were made. See Figure 13.) already explained it is characteristic of the yellows that even in badly diseased fields of winter cabbage it rarely makes a clean sweep of any considerable area. Instead the plants are affected in varying degrees and individuals 20 Through the cooperation of Professors A. D. Selby and S. N. Green of the Ohio Agricultural Experiment Station, we have also received seed of a strain of the favorite kraut varietv. All Season, which has been selected at the Ohio Experi- ment Station for disease resistance. This will be tested in our 1915 series. Through the further cooperation of Professors J. W. Britton of the Ontario Agri- cultural College, Guelph, Canada, and Professor C. E. Myers, of State College, Pennsylvania, we have received samples of seed of the Houser variety, which has been selected in Canada by Edwards (1907. 1908) for resistance to black rot. This will also be included in our further field trials. 38 Wisconsin Research Bulletin 38 here and there appear normal and form sound, fully-devel- oped heads. (Fig. 12.) Evidently there are two possible explanations for this; it may be due to difference in oppor- tunity for infection, or, if infection is uniform, it may be due to different degrees of susceptibility or resistance of the individual plants. Even if the latter be the true explanation it remains a question of the greatest practical moment whether such individual differences are transmitted with constancy through the seed from generation to generation. It was also obvious that in case they are, the practical solution of the problem was further conditioned upon find- ing such fixed and inheritable disease-resistant quality in com- bination with those other qualities which characterize the best commercial type of winter cabbage: keeping-quality, yield, texture, size, and shape of head. Steps were at once in- augurated to determine these points. INITIAL HEAD SELECTIONS OF 1910 The disease was unusually severe in 1910. In the autumn selections were made in three different fields, all badly cab- bage sick. In each case the aim was to select plants that were both sound and of the best commercial winter cabbage type. A. J. Rogers, of the department of horticulture, and F. D. Bailey, of the department of plant pathology, cooperated in this, and much reliance was placed on the judgment of W. J. Hansche, of Racine, an experienced cabbage grower and dealer. (Fig. 13.) Proceeding in this way, about 100 heads were chosen and placed in storage, November, 1910, from the three fields, and of these 50 heads were finally selected for replanting in May, 1911, as follows: From the field of Matt. Broesch, Kenosha, three heads; variety bought from D. M. Ferry Company as "Hollander or Danish Ball Head." This selection was from the worst- diseased field under observation. Upon the one-half acre of this only about twelve plants survived to form heads at the end of the season. Six of these were evidently dis- eased, and of the other six placed in storage, only three sur- vived the winter for seed growing in 1911. These are the heads later designated as VHIa, VII lb, and 1X3. Control of Cabbage Yellows 39 From the field of Witcheber Bros., Kenosha, 50 heads; variety the same as the last. This field was not nearly as badly diseased as Broesch's, and the plants had not, there- fore, been subjected to as severe a process of natural selection. Of these, 30 heads passed through the winter in good con- dition, were replanted and produced seed in 1911. These are the heads later designated as Vlla-y and IX 5, 6, 24, 26, 32. From the field of W. J. Hansche, Racine, 44 heads (Figs. 12 and 13). This was from seed of the Hollander type. FIG. 13.— SELECTING DISEASE RESISTANT GARBAGES The selections were made in 1910 in the fields where the yellows was as bad as could be found. 21 The above field would have shown a full stand except for Fusarium. grown for Mr. Hansche in the Puget Sound region. Of these 16 were selected in the spring, 1911, replanted and produced seed. These are the heads later designated as IX 105, 116 and X 101-143. It will thus be seen that we grew seed in the summer of 1911 from 49 seed-mother-heads, all selected for disease 2' Mr. W. J. Hansche. an experienced cabbage grower and Professor A. J. Rogers of the horticultural department cooperated in selecting heads of the best commer- cial type as well as free from disease. 40 Wisconsin Research Bulletin 38 resistance, secured from three fields, and repre- senting two distinct com- mercial types. An important question at the outset concerned the possibility or neces- sity of cross-pollination as between these heads. Cabbage seed-plants have a prolonged period of blossoming and are freely visited by bees and other insects which are instrumental in cross- pollination. Tracy (1906) advises that such cross- pollination seems essen- tial, that a single isola- ted cabbage plant pro- duces little or no seed and that for the best re- sults seed-plants repre- senting different types or strains should be set to- gether. With these sug- g e s t i-o n s i n mind the plantings were planned so as to be sure to get seed in one or all of three ways: (1) by self -pollination, (2) by cross- pollination between heads selected from the same field, and (3) by cross- pollination between heads selected from different fields. To this end the 49 cabbage heads were handled in four lots, each of which was planted by itself and with no other cabbage seed plants in its neighborhood. The three lots collected from the three different fields were replanted, each by itself on the farm where grown, as follows: FIG. 14— CABBAGE SEED PLANT For seed growing in this climate the plant is pulled up by the roots in late autumn, stored in a cool cellar, or trench, and set oiit again the following spring. Such a plant will in general produce from an ounce to a quarter of a pound of seed. Control of Cabbac.I': Yi;i.i.ows 41 (1) At \Vilchel)er's, 25 heads, dcsignaled as Nos. Vlla-y. (2) At Broesch's, 2 heads, Nos. Villa, \lllb. (3) At Hansche's, 14 heads, designated by numbers ranging between X 101 and X 143. (4) In addition to the above, 8 heads representing selections from each of the three fields were planted, inter- mingled in close proximity, at Madison. These included the following. No. IX 3 (Broesch held,) IX 5, 6, 24, 26, 32 (Witcheber field), IX 105, 116 (Hansche field). With lot 4 some of the branches were bagged, but no seed was secured from these. The uncovered branches produced seed abundantly, and it is assumed that this, in general, resulted from cross-pollination between the various heads representing the different fields. Lots 1 to 3 were grown without bagging. Each of these fields was isolated with no other seed-cabbage in the vicinity. In each case, it is presumed much of the seed resulted from cross-pollination, but this was, of course, restricted in each lot to its own type, i.e., it was in each lot betw^een plants selected from the same field. Since these had in each lot originated from the one commercial strain of seed and had, moreover, in all cases alike been selected for disease resistance and for uniformity to one type, such crossing commended itself to us as desirable rather than objectionable. ^^ The seed secured in 1913 from each of these 49 heads was collected and saved by itself and represents the beginning of one of the head strains. Each of these has since been maintained distinct in our later trials and bears the number by which it has just been designated, e.g.. Villa, X 143, etc. The behavior of these in the trials of the succeeding years will now be considered. TRIALS IN 1912 AND 1913 OF THESE FIRST-GENERATION SELECTED-HEAD STRAINS Trials w^ere made of these selected head strains in 1912 and 1913. The trials of 1912 included all of the head strains except certain of the VII series, a total of 38 strains. The " Since the outcome from this latter method has proved highly satisfactory and better than that from lot 4, as will appear later, we have continued to follow this method in our subsequent work. Our associates, R. F. Howard and G. F. Potter, of the horticultural department, and L. J. Cole of the department of experimental breeding, have, however, undertaken to supplement this work by further breeding experiments, including close-pollination, and cross-pollination with these selected strains and with certain other commercial types. 42 Wisconsin Research Bulletin 38 1913 trials included a repetition of the trials of all tested in 1912 of which enough seed remained and in addition all of the strains VII omitted in 1912. These trials were made on the same cabbage-sick soil (Hansche field) where the disease was prevalent in 1910 (Figs. 12, 13) and which has been re- planted to cabbage, experimentally, every year since. The soil has, therefore, become more sick from Fusarium infesta- tion each year, if this be possible. Comparison of the losses in the control plots in these respective years shows that the FIG. 15.— TRIALS OF SELECTED HEAD STRAINS, 1912 The three row.s at the left of the center (IX .3) and at the extreme right (Villa) are both resistant Hollander .strains. The three rows in right center (VI) iniported Danish, non-resistant, have been nearly exterminated by the yellows. For fur- ther details see Table VH. seasonal conditions were not, however, quite as favorable for the fungus in either 1912 or 1913, as in 1910 and 1911. There was, nevertheless, enough yellows both seasons to insure a thorough comparative trial. The trials were conducted as follows: The selected seed was planted along with controls in disease-free soil, where all developed with like vigor. A uniformly strong, healthy lot of seedlings from each strain was then selected and all transplanted on the same day, in the last week of June, into the infected trial field, in parallel rows, 45 plants of each strain being used in 1912 and 81 of each in 1913. Control of Cabbage Yellows 43 In 1912 a corresponding series of seedlings was at the same lime transplanted to virgin soil, free from Fusarium, on another farm. On this latter field all alike produced a fine stand of healthy cabbage which headed up uniformly well. On the sick soil of the trial field the yellows appeared each season within two weeks of the time of transplanting and rapidly increased on the commercial varieties planted as controls. From the outset, and in both seasons alike, a marked difference was shown in the amount of disease in the selected head strains as compared with these commercial FIG. 16.— TRIAL OF A SELECTED HE.VD STRAIN. 1913 At the left is Leuker's Short Stem Hollander (non-resistant), at the right the resistant head strain Villa. This trial was on the Broesch farm, and was made to supplement the main field. (Compare Figure 15.) controls. (Figs, 15 and 16). The results are brought to- gether for convenient reference in the following table. They are expressed in percentages of the original stand except that, since there was always an_occasional loss from insects 44 Wisconsin Research Bulletin 38 and in cultivation, where this occurred it was noted and proper deduction was made. There was no disease other than yellows responsible for any of the loss so that the results as tabulated represent correctly the loss due to the Fusarium attack, or yellows. In each season, 1912 and 1913, further selections of heads were made from the most promising head strains and saved for seed-growing. The less promising were ruled out from further trial as indicated. Inasmuch as the commercial varieties of winter cabbage of the Hollander or Danish Ball Head type reported upon in the last chapter (Tables IV, V), were grown in this same field and constitute the controls in these trials, the results ob- tained with them are repeated at the beginning of the follow- ing table. (Table VII.) SUMMARY OF RESULTS OF TRIALS OF 1912-1913 The significance of these results may perhaps be most clearly grasped by condensing and tabulating the averages for the two seasons as follows: Table VI. — Summary of Significant Results from Table \TI Commercial varieties Average 5 varieties, commercial seed Puget Sound Hollander Ferry Hollander Selected head strains Ferry's Hollander from Witcheber field; mixed stand, average 5 strains pure stand, average 25 strains head strain, Vlly..; head strain, VIIv Ferry's Hollander from Broesch field; mixed stand, average 1 strain pure stand, average 2 strains head strain Villa head strain Vlllb Puget Sound Hollander from Hansche field; mixed stand, average 2 strains. pure stand, average 14 strains head strain X 135 head strain X 143 Per cent •Per cent lived headed 29 12 28 18 30 0 69 35 59 40 86 60 ' •^82 64 85 64 92 67 , 94 ' 64 89 70 90 50 80 57 98 78 98 93 *The above figures for "per cent headed", although correct for 1013. fail to do full justice to the Ferry Hollander since the trials of 1911 and 1914 show that it is at least equal to other commercial varieties, if not superior, in disease resisting qualities. Control of (^abbaoi-: Yi.li.ows 45 Tablk^VII. — Trials in 1912 and 1913 of Sekd i-hom the Head Selections of 1910 Condition crop Condition crop Source and name Number end season end season Hemarks 1912 1913 Per Per Per Per CommercUl not selected cent cent cent cent i (Inserted for comparison) living headed living headed Imp. Danish Large Winter VI 14 0 7 2 Under trial 1911 also with poor poor Imp. Danish Amager long stem II 36 13 12 4 record Under trial 1911 also with record Imp. Dani.sh Amager short stem.. III 69 27 41 22 Best of imported varieties Hansche's Puget Sd. Hollander... XV 24 24 32 12 Our selections X, 1910, were of this variety. Our selections VII and VIII. Ferry's Hollander XVI Not tested 30 0 1910, were of this variety Head selections of 1910, seed grown in pure plantations, 1911 Ferry's Hollander selected from Witcheber field. 1910 Vila 63 35 54 38 Discontinued Vllb 17 17 60 48 Discontinued VIIc 68 52 56 44 Discontinued Vlld 56 40 40 30 Discontinued VUe 71 52 50 36 Discontinued Vllf 81 63 42 22 Saved 10 heads, 1912 Discontinued, 1913 Vllg 75 58 60 46 Discontinued Vllh 73 57 58 32 Discontinued Vlli 80 68 66 50 Saved 10 heads, 1912 Vlli 68 41 43 33 Discontinued Vllk 67 53 46 28 Discontinued VIII 44 21 35 25 Discontinued Vllm 24 24 75 59 Discontinued Vlln 65 43 60 44 Discontinued VIIo not tested 70 46 Di.scontinued VIIp not tested 50 30 Discontinued Vllq not tested 53 18 Discontinued Vllr not tested 43 12 Discontinued VI Is not tested 64 32 Discontinued Vllt not tested 62 28 Discontinued VIIu not tested 60 34 Discontinued VIIv not tested 82 64 25 heads selected, 1913 VIIw not tested 78 57 Discontinued VIIx not tested 62 38 Discontinued Vlly not tested 86 60 25 heads selected, 1913 Ferry's Hollander selected from Broesch field, 1910 Villa 96 80 92 48 Saved 35 heads, 1912 Vlllb 87 67 91 73 Saved 29 heads, 1912 Puget Sound Hollander selected from Hansche field, 1910 XlOl 71 58 not tested Discontinued X104 70 37 not tested Discontinued X108 62 44 not tested Discontinued X112 71 51 not tested Discontinued X113 58 33 not tested Discontinued X117 69 53 not tested Discontinued XI 18 78 60 not tested Discontinued X120 93 59 not tested Discontinued X122 91 66 not tested Discontinued X123 90 45 not tested Discontinued X124 80 67 not tested Discontinued X128 89 57 not tested Discontinued X135 98 78 not tested Saved 32 heads, 1912 X143 98 93 not tested Saved 38 heads, 1912 Head'seleclions of 1910 grown in miied plantation, 1911, Ferry's Hollander selected from Broesch field,. 1910 IX 3 88 5S 82 70 Discontinued Ferry's Hollander selected from Witcheber field, 1910 IX 5 78 42 not tested Discontinued IX 6 63 35 not tested Discontinued 1X24 60 28 70 46 Discontinued 1X26 65 23 48 12 Discontinued 1X32 75 41 78 53 Discontinued Puget Sound Hollander selected from Hansche field, 1910 1X105 87 42 not tested Discontinued 1X116 91 51 92 57 Discontinued 46 Wisconsin Research Bulletin 38 DISCUSSION OF RESULTS OF THE TRIALS OF 1912 AND l'.)13 Table VII shows that even the poorest among the selected head strains averaged better than the best of the parent commercial strains from which selections were made. Table VI brings out still more strikingly the relative superiority of the selected head strains over the parent commercial strains. These averages prove conclusively the correctness of the principle of selection here practiced. On the other hand, the greatest stimulus to continued selection is found in the significant differences in the behavior of the various selected head strains, some of which show up far better than others. It has already been explained that in most cases the seed- mother-heads of each kind or type have been planted sepa- rately but that certain of them each year have been grown in mixed plantation which would permit of cross-fertiliza- tion. Comparison of the results from the strains where the seed plants from the three fields grew in mixed stand showed that the outcome as to disease resistance and yield was somewhat poorer than where each was grown in pure stand. Since the ptire stand also yielded as much seed as the mixed stands it was decided to discontinue further trials with any of those mixed strains, which included all of the IX series. It is also to be noted that the Ferry Hollander selections from the Broesch field (Series VIII) averaged considerably better than those from the Witcheber field (Series VII). This superior disease resistance of the VIII (Broesch) strains over the VII (Witcheber) strains, is, indeed, what was to be expected considering their history. The two fields were planted with the same commercial variety, Ferry Hollander, in 1910, but the disease was not nearly so severe on the Witcheber field, hence the opportunity for selection was less favorable. It should be recalled that on the Broesch field of one-half acre only three mother-heads survived to produce seed and heads Villa and VI I lb were the best of these. Control of Cabbage Yellows 47 It was also easy to select X135 and X113 as the best two head strains of the Puget Sound Hollanders, although it was unfortunate that no seed of this X series was available for further trials in 1913. This prepared the way for focusing attention in the trials of 1914 upon the relative merits of these most promising head strains from each of the two types, viz., Villa and Vlllb, of the Ferry Hollander, and X135 and X143 of the Puget Sound Hollander. Seed of these w^as grown in 1913 from the heads selected in 1912. SEED GROWING IN 1913 FROM THE SECOND-GENERATION HEADS SELECTED IN 1912 Selections were made in 1912 as already explained in Table VII of heads for the continuation of the trials. It will be noted that the chief attention in these second-generation head selections was given to the four head strains already shown to be most promising, viz.. Villa and Vlllb, X 135, and X 143. These heads were in all cases selected as apparently healthy on badly diseased ground. A few heads were lost in winter storage. In 1913 seed was grown from the balance. The same general plan was followed as in seed-growing in 1911, i.e., part was grown in isolated planta- tions at Madison, and the balance in mixed plantation at Racine. The seed from each head was again kept separate. Following is a list of these together with the yields of seed in grams per head. Table VIII.— Seed Grown in Pure Plantations, Madison: 1913 Strain Weight Head strain Weight Head strain Weight Head strain Weight grams number grams number grams number grams Vllf 1 82 Vlllaie 58 VIIIbl6 9 X 13.5-23 56 Vllf 4 38 VIIIal7 30 VIIIbl7 2 X 135-30 38 Vllf 5 103 VIIIa20 24 VlllblS 32 X 135-33 57 Vllf 7 100 VIIIa2I 13 VlllblO 28 X 143- 1 22 Vllf 9 67 VIIIa22 39 VIIIb20 2 X 143- 2 40 Vllf 10 6 VIIla23 23 X135- 2 54 X 143- 3 75 Vlli 1 70 VIIIa25 38 X135- 3 65 X 143- 4 23 Vlli 2 127 Vlllb 1 27 XI 35- 4 13 X 143- 7 35 Villi 3 33 Vlllb 2 21 X135- 5 48 X 143- 8 26 Vlli 4 3 Vlllb 3 38 X135- 6 31 X 143- 9 43 Vlli 6 63 Vlllb 4 30 X135- 8 85 X 143-10 69 Villa 2 5 Vlllb 5 5 X135- 9 28 X 143-12 16 Villa 3 22 Vlllb 6 6 X135-11 58 X 143-13 24 Villa 5 15 Vlllb 7 27 X135-12 30 X 143-14 24 Villa 6 1 Vlllb 8 10 X135-13 41 X 143-15 20 Villa 7 32 Vlllb 9 39 X135-15 10 X 143-16 11 Villa 8 16 VlllblO 4 X135-17 76 X 143-18 52 Villa 9 6 Mllbll 8 X135-18 63 X 143-20 31 VIIIal2 15 VIIIbl2 1 X135-19 122 X 143-21 72 VlllalS 54 VlllblS 50 X135-20 59 X 143-24 15 VlllaU 19 VIIIbH 53 X135-21 64 X 143-25 65 VlllalS 35 VIIIbl5 3 X135-26 6 X 143-28 40 48 Wisconsin Research Bulletin 38 Table IX. — Seed Grown in Mixed Plantation, Racine: 1913 Head strain Weight Head strain Weight Head strain Weight Head strain Wright number grams number grams number grams number grams Vllf 2 5 Villa 33 15 Vlllb 29 9 X 135-32 20 Vllf 3 49 Villa 34 33 X 135-24 29 X 143-29 47 Vllf 8 22 Villa 35 33 X 135-25 26 X 143-30 16 Vlli 9 42 Vlllb 22 28 X 135-26 28 X 143-31 9 VIIIa27 30 Vlllb 23 12 X 135-27 22 X 143-33 2 VIIIa28 8 Vlllb 24 24 X 135-28 25 X 143-34 18 Vllla29 14 Vlllb 25 15 X 135-30 38 X 143-35 5 VIIIa32 6 Vlllb 27 3 X 135-31 11 X 143-38 68 TRIALS OF SELECTED-HEAD STRAINS 1914 The number of head strains available for trial in 1914 had become so large that it was feasible to make trial of only a select list of the more promising of each series. Therefore, in order to eliminate any especially weak plants only those were included which produced one ounce or more of seed in 1913. This gave 44 head strains as listed in the Table X. Along with these selected strains commercial controls were introduced. The procedure was as heretofore, to make the seed bed in disease-free soil, select uniformly strong seedHngs, and trans- plant to Fusarium-infested soil. In order to make these 1914 trials as convincing as possible, trials were made in three scries as follows: A. The larger numbers were planted on the W. J. Hansche farm on the same cabbage-sick soil which had grown experi- mental cabbage each year since 1910. Owing to the large number of plots only 81 heads could be planted in each plot in this field. B. A smaller number of the most promising selections were planted on the farm of Matt. Brocsch, Kenosha. Mr. Broesch selected the area as representing the worst cabbage- sick soil in a field where he had abandoned cabbage culture because of successive losses of his crop through yellows. This being a larger field with fewer strains, about 240 plants were set of each head strain under trial and twice as many of the commercial controls. Control or (Cabbage Ykllows 49 C. A sample of seed of a single head strain was given to each of several farmers in the neighborhood with instructions to test it alongside of the best commercial seed on cabbage- sick soil. The climatic conditions of 1914 were favorable for a severe attack of yellows. Although there were copious rains im- FIG. 17.— TRIAL OF SELECTED HEAD STRAINS, 1914, HANSCHE FIELD At the left of the center are 3 rows of commercial Puget Sound Hollander (Nos. 67-69), non-resistant. At the end of the season 20 per cent headed. At the right of the center are .S rows (Nos. 64-66) of selected head strain. X 135-33, re- sistant, but not one of the best; at the end of the season 78 per cent headed. Other selected head strains are shown in the background at the extreme right and left. For further data from this field see Table X. mediately following the transplanting which gave the plants a satisfactory start, these were followed by a period of dry, hot weather through the middle of July which so stimulated the Fusarium development that before August 1 the disease was much in evidence on all the non-resistant commercial varieties in the trials fields. The outcome of these trials is shown in Table X. 50 Wisconsin Research Bulletin 38 In order to enable a better analysis of the results to be made fuller data are included in this table than in those of the FIG. 18.- -A FARMER'S TRIAL OF RESISTANT STRAINS VS. COMMERCIAL VARIETIES Wm. Rraid of Bcrryville made a trial of resistant strains in 1914. He planted for his general field two varieties of commercial seed, both of which were ruined by yellows. Through the middle of the field one row was set with plants of the selected head strain X 143-18 as shown in the figure. This row matured 174 heads; the row at the right. Imported Danish, matured 17 heads; the row at the left, I^uget Sound Hollander, matured 8 heads. Control of Cabbage Yellows 51 Table X.— Trials in 1914 of Selected Cabbage Strains; Field A, IIansciie Farm (Figure 17) Description of seed Condition at end Marketable heads pro- season duced Per Yield Source and kind Rows Year Yel- Liv- Head- cent Aver- per number grown Number lows ing ed* of total age weight acre (Rows 1-69. Paget Sound Per Per Per Per Hollander) cent cent cent cent Lbs. Tons Commercial control 1-3 4-6 1913 XV X 143- 2 82 12 35 93 13 83 6 67 3.8 4.3 0.9 Second generation selection 11.4 7-9 X 143- 3 10 91 83 68 3.9 10.4 10-12 X 143- 9 35 91 63 44 3.5 6.1 13-15 X 143-10 15 91 81 67 4. 10.4 16-18 X 143-18 5 100 97 89 3.8 13. 19-21 X 143-21 19 86 80 68 3.8 10.5 Commercial control 22-24 XV 86 34.6 19 IS 3.9 2.3 Second generation selection 25-27 1913 X 143-25 40 86 79 73 3.5 10.1 28-30 X 143-28 17 90 84 84 3.8 12.4 31-33 X 143-29 4 100 100 91 3.7 13. 34-36 X 143-38 48 56 53 47 4.3 8. First generation selection 37-39 1911 19 91 86 77 4.2 12.7 40-42 X 135 •20 91 81 78 3.8 12.3 Second generation selection 43-45 1913 X 118- 4 28 83 75 75 4. 11.8 Commercial control 46-48 XV 90 26 15 11 3.2 1.8 Second generation selection. 49-51 1913 X 124- 2 25 85 79 72 3.5 11.1 52-54 X 135- 2 7 96 93 86 3.9 13.2 55-57 X 135- 3 1 100 99 97 3.5 13.6 58-60 X 135-20 21 90 79 80 4. 12.9 61-63 X 135-23 24 S3 72 78 4. 12.2 64-66 X 135-33 27 84 73 78 3.6 11.2 Commercial control 67-69 XV 84 37 26 20 2.6 2. (Rows 70-12G Ferry's Hol- lander) Second generation selection 70-72 1913 Villa 7 3 100 85 81 3.5 11.4 73-75 Villa 13 0 100 94 83 3.5 11.6 76-78 Villa 15 0 100 96 88 3.5 12.2 79-81 Villa 16 1 100 97 86 3.8 13.1 82-84 Villa 17 7 100 99 88 3.4 11.8 Commercial control 85-87 88-90 1913 XV Villa 22 74 0 62 100 30 99 27 94 1.8 4.6 1.8 Second generation selection 17.1 91-93 Villa 25 0 100 100 94 4.9 18.3 94-96 Villa 27 0 100 97 94 3.8 14.3 97-99 Villa 34 1 99 94 75 3.1 9.3 100-102 Villa 35 0 100 94 64 3.1 8. Commercial control 103-105 106-108 1913 XVI Vlllb 3 75 0 51 100 33 99 23 92 2. 3.4 1.9 Second generation selection 12.2 109-111 Vlllb 4 9 99 91 91 3.7 12.2 112-114 Vlllb 9 7 100 95 91 3.6 11.9 115-117 Vlllb 13 4 99 90 85 3.9 12.6 Commercial control 124-126 XVI 71 55 34 34 2.1 2. *The data for this column were taken October 1 and are based on a count of all plants which had then formed a hard head, regardless of size of the head. The next column represents the precentage of heads of marketable size at time of cutting, in November. It wih be noted that the figures in this first column in most cases are a little higher than those in the second. They are included partly because they represent rather more exactly than the next column the growth conditions but chiefly because they correspond to the data taken in the years 1911- 1913 on this same field and furnish the best basis for a comparison of developments during these four years. preceding years, including percentages of disease, average weights, and yields. FARMER S TRIALS In addition to the trials in these two experimental fields planned and planted under our immediate supervision, a 52 Wisconsin Research Bulletin 38 number of farmers' trials were made in 1914. In each such case a 10-gram sample of one of the selected strains was given to a farmer who had experienced loss from cabbage yellows Table XI. — Trials in 1914 of Selected Cabbage Strains; Field B, Broesch Farm (Figure 19) Description of seed Condition at end season Average weight Yield per Source and kind Row Number Yellows Living Headed acre (Rows 1-17, Puget Sound Hollander) Per cent Per cent Per cent Pounds Tons Second generation selection 1 X135- 8 46 76 53 3.7 7.8 2 X135 17 38 88 63 3.2 8.1 3 X 135-18 69 53 38 3.7 5.6 4 X 135-19 31 88 71 3.5 10.1 5 X 135-21 63 70 48 4.7 8.9 Commercial control 6 7 XV XV 95 95 15 8 8 5 4.1 4.8 1.3 0.9 Second generation selection 8 X 143- 2 63 69 48 4.5 8.6 9 X 143- 3 51 81 62 3.7 9. 10 X 143- 9 87 43 25 3.9 3.8 11 X 143-10 78 61 39 3.4 5.2 12 X 143-18 48 82 63 4.4 10.9 13 X 143-21 55 70 56 3.8 8.3 14 X 143-25 66 59 45 4.2 7.4 15 X 14.3-28 63 67 52 3.9 8.1 16 X 143-29 20 94 80 4. 12.7 17 X 14.3-38 89 35 18 3.6 2.5 Leuker Hollander, short stem* 18 XIII 11 97 29 8 2.8 .9 Leuker Hollander, long stem* . 19 XIV 18 96 23 8 2.2 .6 (Rows 20-40, Ferry's Hollander) Second generation selection 20 Vllf 5 47 83 62 4.5 11.1 21 Vllf 7 64 75 46 3.3 5.9 22 Vllf 2 55 71 54 4.4 9.3 23 Vllf 6 57 72 52 4.7 9.8 24 Villa 7 9 99 88 4. 10.6 25 Villa 13 9 98 91 3.6 8.7 2C Villa 15 8 99 93 4. 9. 27 Villa 16 11 99.6 94 4. 11.1 28 Villa 17 23 93 83 3.6 9.8 29 Villa 22 2 99.6 99 5.4 18.2 30 Villa 25 3 100 96 6. 19.2 31 Villa 27 U6 99 91 4.5 11.8 32 VIII.1 34 8 98 89 3.8 10.1 33 Villa 35 5 100 96 4.1 12.8 Commercial control Ferry 34 XVI 89 37 16 3.6 2.3 35 XVI 89 35 18 3.2 2.2 Second generation selection 36 Vlllb 3 4 99.6 93 5. 18.2 37 Vlllb 9 6 99 95 4. 15.1 38 Vlllb 13 11 97.4 86 3.8 13. 39 Vlllb 14 17 94 83 3.3 9. 40 Vlllb 18 22 95 78 3.7 10.9 *Rows 18 and 19 were Hollander cabbage of two strains which have been grown for eight years by Theo- dore Leuker, of Racine. They are excellent commercial cabbages but these and other trials have shown that they are fully as susceptible to yellows as the ordinary commercial Hollander strains. with the request that he put the plants for trial on sick old cabbage land alongside of the standard commercial variety of cabbage he considered best. Owing to the small amount of seed available each sample was necessarily taken from a different strain. In all cases samples from the same strain were also under trial in one or both of our experimental fields as reported above. Control of Cabbage Yellows 53 Naturally some of these were not so managed as to have experimental value, but in a number of fields where the disease was bad the outcome was quite as convincing as on our experimental fields. Some of these results are worth citing. FIG. 19.— TRIALS OF SELECTED HEAD STRAINS, 1914, BROESCH FIELD The cabbage at the right is resistant head strain VIIIb-3, yielding 219 heads per row, or an equivalent of 18.2 tons per acre. At the left are two rows of com mercial (non-resistant) Ferry's Hollander, averaging 40 heads per row, or only 2.3 tons per acre. At the extreme left are resistant head strains. Villa Series, which yielded 19.2 tons per acre. For further data see table XL Graber field. N. Graber planted in 4 parallel rows on cabbage-sick soil, 138 plants each, 2 rows of commercial Puget Sound Hollander and 2 rows of our selected head strain X 135-34. . The condition as to yellows was recorded August 21 and counts made of the numbers living and headed at the end of the season. The results were as follows: Yellows Per Plants cent Commercial Selected head strain X 135-34 110 80 40 3 Living Plants 210 268 Per cent 76 97 Headed Per cent Plants 189 68 262 95 54 Wisconsin Research Bulletin 38 Drummond field. . W. S. Drummond planted his com- mercial cabbage field with the commercial Piiget Sound Hollander. Through the middle of the field he set one row with 328 plants of the selected-head strain X 143-29. The soil was supposed to be not badly infested, but as the season progressed considerable yellows developed. The following figures show the relative amount of disease as well as the conditions at harvest time of the selected row as compared with the commercial rows on either side. Yellows Living Headed Per Per Per Plants cent Plants cent Plants cent Commercial (average 2 rows) 142 43 322 98 202 62 Selected head strain X 143-29 10 3 326 99 311 95 Mr. Drummond was so favorably impressed with this outcome that he saved the heads of the selected strain for seed-growing. A. and S. Hansche field. A. and S. Hansche planted a small field with commercial cabbage seed of the Puget Sound Hollander and planted at one side of it 5 rows of the selected head strain X 135-19. Examination on August 22 showed 30 per cent of yellows among the commercial plants and 3 per cent in the rows of the selected strain. At the end of the season 97 per cent of the selected strain developed heads as compared with 75 per cent in the com- mercial rows. Piper field. A. J. Piper planted a small area for purposes of this trial on old cabbage ground which he knew to be decidedly cabbage sick. He planted 6 rows through the middle of the field with the selected head strain X 143-10, with 5 rows on one side and 3 rows on the other of commercial Puget Sound Hollander. The yellows developed quite badly in this field and many plants in the commercial rows died early with very little disease in the selected strain. No counts were made of the original stand and no exact percentage records of yellows were made during the summer. The yields in the autumn in the same length of row averaged. Control of Cabbage Yellows 55 commercial strain, 156 heads, selected head strain, 303 heads. Moreover, the difference in size of heads was fully as striking as the difference in number, the heads in the selected row being apparently twice as heavy. Mr. Piper was so well pleased with the outcome that he saved these heads of the resistant strain for seed growing. Braid field. William Braid planted a large field with cabbage, supposing his soil to be not sufficiently infected to prevent his growing a profitable crop. He used two commer- cial varieties, Puget Sound Hollander and an imported Danish seed from Copenhagen. Through the middle of his field between these two varieties he planted one row of the selected head strain X 143-18 (Fig. 18). The yellows ap- peared in very severe form on this field so that when counts were made on August 21 it was found that 75 per cent of the commercial plants showed the disease and in the row approx- imately GOO feet long only 44 plants were still living. There was considerable disease, 29 per cent, in the selected-head row, but 219 plants were then living. At the end of the season the counts were as follows on the resistant row and the two commercial rows immediately alongside. Living Headed Commercial Pugel Sound Hollander 25 plants 8 plants Selected head strain X 143-18 219 " 174 '| Commercial Imported Danish 29 " 17 " Although comparison with our main field plots will show that this w^as not one of the best strains, it looked so good to Mr. Braid that he saved the best heads from this row for seed-growing. Krause field. William Krause made a trial of selected- head strain Vlllb 3 by planting one row of this and one row of commercial cabbage through the middle of a field where he knew the soil to be cabbage sick, the balance of the field being planted to sugar beets. The percentage of yellows on August 21 and the condition at the end of the season were as follows : Yellows Living Headed Commercial variety 84% 55 plants 16 plants Selected-head strain Vlllb 3 8% 293 " 233 " 56 Wisconsin Research Bulletin 38 J. Hansche field. J. Hansche planted a row of selected- head strain VI I lb 9 through the middle of his trial field the balance of which was planted to the commercial cabbage of the Danish Ball Head variety known as Grenadier. Yel- lows was very severe on this field also, so that on August 21 counts showed 88 per cent of these commercial plants dis- eased as compared with 9 per cent on the selected-head row. At the end of the season counts on the resistant row and the commercial rows on either side of it gave the following figures: Living Headed Commercial variety Grenadier 45 plants 8 plants Selected head strain Vlllb 9 277 " 238 " Commercial variety Grenadier 65 " 18 " Broesch field. Henry Broesch was given the selected head strain VHIa 13. For his commercial cabbage crop he used seed of two varieties, Danish Ball Head Grenadier, and Ferry's Hollander, planting several acres of each. Through the middle of this field where these two varieties met he planted two rows of the selected strain. Although he had supposed the field suitable for a successful cabbage crop, considerable yellows developed so that on August 21 counts showed an average of 59 per cent diseased along the commer- cial varieties with only 1 per cent in the selected head rows. At the end of the season Mr. Broesch reported the outcome as follows and our counts subsequently substantiated the fig- ures: Conditions (2 rows each) Ferry's Hollander Number plants at beginning 524 Died from yellows 236 Died from other causes 14 Lived but did not head 30 Heads harvested 244 Percentage that lived 52 Percentage that headed 47 Since this is from the Villa series which included the most promising strains in our own trials, we encouraged Mr. Broesch to save the heads of this strain for seed-growing with the expectation Ilia I he would be able to supply some others as well as himself in this way. Commercial Commercial Selected Strain Danish Villa 13 Ball Head 509 519 1 352 4 7 2 30 502 136 99 32 98.6 26 Control of Cabbagi-: Yellows 57 DISCUSSION OF RESULTS OF I'JM TRIALS The results of the 1914 trials serve fully to establish our confidence that the disease-resistant characteristics of the selected strains are fairly well fixed, constant, and inherit- able. Examination of the above tables shows that in every case the selected head strains stood up much better than the best commercial strains. One of the encouraging things is the evidence that on the whole the second generation selections showed some improvement over the first generation. It will be noted, however, that among these there is a con- siderable variation, indicating probable opportunity for further improvement through continued head selection. The general facts justifying these conclusions may perhaps be brought out more clearly by summarizing the more im- portant results from each of the two trial fields. Table XII. — -Summary Hansche Trial Field: 1914* Description of seed Yellows Living Headed Market- able Average weight of heads Yield per General averages Commercial, averages o( all kinds Selected head strains, averages of all Puget Sound Hollander Commercial averages of all 12 rows First selection head strain X 143 Second selection head strain X 143 averages all X 143-29 the best of tliis series First selection head strain X 135 Second selection head strain X 135 averages all X 135-3, the best of this series Ferry's Hollander Commercial averages of all 9 rows Selected head strains Villa series, average all 30 rows Villa 25, the best of this series... Selected head strains Vlllb series, averages of all 12 rows VUIb 3, the best of this series Per cent 80.1 12.5 85.5 19 21 4 20 16 1 Per cent 42.9 93.5 33.2 91 88.4 100 91 91 100 56 100 100 99 100 Per cent 24.1 87.2 100 81 83 99 32 96 100 95 Per cent 18.9 79.5 13. 77 69.8 91 78 83.8 97 28 84.7 94 Pounds 2.8 3.8 3.4 4.2 3.9 3.7 3.8 3.8 3.5 3.7 4.9 3.7 3.4 Tons 1.8 11.7 1.8 12.7 10.5 13 12.3 12.6 13.6 1.9 12.2 18.3 12.2 12.2 •Averages from Table X. 58 Wisconsin Research Bulletin 38 Table XIII. — Summary Broesch Trial Field: 1914* Description of seed Paget Sound Hollander Commercial, averages of all kinds Selected head strains X 143 series, average all.. X 143-29, the best of this series Selected head strains X 135 series, average all X 135-19, the best of this series Ferry's Hollander Commercial, averages of all kinds Selected head strains Villa series, average all.. Vnia 25, the best of this series Selected head strain VHlb.series, average all.... Vlllb 3, the best of this series Average Yellows Living Heads weight Per cent Per cent Per cent Pounds 95 12 7 4.5 62 66 49 3.9 20 94 80 4 49 75 55 3.8 31 88 71 3.5 89 36 17 3.4 8.4 99 92 4.3 3 100 96 6 12 97 87 4 4 99.6 93 5 Total yield per acre Tons 1.1 7.7 12.7 8.1 10.1 2.3 12.1 19.2 13.2 18.2 'Averages from Table XI. The results in these two trial fields show beyond doubt the superiority of the Villa strains. The evidence, as shown in the field by the uniformity, general thrift, and healthy appearance of these plants, was even more con- vincing than the figures as tabulated. A committee of local cabbage growers and buyers of long experience went over these fields in the autumn and all agreed as to the su- periority of these strains in commercial cjuality as well as in respect to disease resistance. It was their unanimous ver- dict that these strains represent a highly satisfactory com- mercial type of Hollander cabbage. Moreover, from a comparison of these results in the two fields it is evident that the head strain Villa 25 is some- what superior to any other of the head strains. The only one to rival it was Villa 22, but a comparison as they grew side by side in the two fields showed the same slight superiority of Villa 25 evidenced by the figures in the tables. The following summary serves to bring out these points. Averages from the Broesch and Ilansche Fields, Villa strains. Yellows Living Heads Average Total Per cent Per cent Per cent Weight yield per pounds acre Commercial, Ferry 81 46 24.5 2.6,'j 2.1 Villa, all strains .5 99. .5 94 '4.0 12.2 Villa 22 1 99.8 99 .^).0 17.7 Villa 25 1.5 100. 98 5.45 18.8 For further seed-growing and selection looking to the maintenance or possible improvement of the present stand- ards, a series of 25 of the best heads of the Villa 25 strain Control of Cabbage Yellows 59 were selected from the Broesch field, where the disease was worst, and these will be planted by themselves for seed- growing in 1915. The balance of the heads of all the Villa strains grown in 1914 were also saved for seed-growing in 1915. These include 2000 heads from which the seed will be available for commercial cabbage-growing in 1916 under the name Wisconsin Hollander No. 8. Since various experienced cab- bage growers are cooperating in this seed-growing it is be- lieved that there will be no difTiculty in so organizing and directing their efforts that they will attend hereafter to the commercial aspects of producing the seed of the resistant strains, always growing their mother plants on cabbage-sick soil. This will leave us free to focus attention upon the main- tenance and possible improvement of the parent strains by further trial and selection. Other Questions of Practical and Fundamental Interest The naliirc of disease resistance. Naturally in con- nection with this work numerous questions have arisen which deserve further attention. Chief of these is the fundamental one as to the nature or cause of the disease resistance shown in these selected strains. While much thought has been given to this we can as yet offer no satis- factory reply other than to note that it seems to be asso- ciated with a high degree of general vegetative vigor. We do not wish to imply, however, that such vigor, by itself, necessarily carries with it resistance to Fusarium. While further attention will be given to these questions no early or easy solution is to be expected. Disease resistance in relation lo seed production. There is a wide range of variation in the amount of seed produced by different cabbage heads. Tracy (1909) calls attention to the belief frequently expressed by growers that the amount of seed produced stands in inverse ratio to the quality of the cabbage. While it is doubtless true that the wild-cabbage type would be more prolific than the cultivated, head-forming strains, still there seems no good reason for supposing that those minor differences between cultivated varieties which give them varying commercial 60 Wisconsin Research Bulletin 38 value should be correlated with seed production. Inas- much as there was a considerable difference in the amount of seed produced by the various selected-head strains under trial in 1914, it seemed worth while to correlate the weight of seed secured from each of these in 1913 with the per- centage of yellows shown in 1914, This is done in the fol- lowing table. Table XIV. — Relation of Seed Production to Disease Resistance Weight of Seed Amount of yellows in grams in per cent X 135—19 122 35 X 135— 8 85 46 X 135 — 17 76 39 X 135—21 64 63 X 135 — 18 63 69 X 143— 3 75 51 X 143 — 21 72 55 X 143 — 10 69 74 X 143—38 . 68 89 X 143—25 65 66 X 143—18 52 48 X 143—29 47 20 X 143 — 9 43 83 X 143— 2 40 63 X 143—28 40 63 Villa— 16 58 11 Villa— 13 54 9 Villa — 22 39 22 Villa — 25 38 3 Villa —15 35 8 Villa —34 35 8 Villa— 35 33 5 Villa— 7 32 9 Villa —17 30 23 Villa —27 30 6 Vlllb — 14 53 17 Vlllb— 13 50 10 Vlllb — 9 39 6 Vlllb— 3 34 4 Vlllb —18 32 22 Since the test of disease resistance was more severe in the Broesch field and the percentages of yellows are there- fore more significant, the following tabular summary is based primarily upon the results from the strains as tested in the Broesch field. The figures from the Hansche field are, however, added for the X 143 and Villa strains which were tested in both fields. Comparison of first and second generations as to disease resistance. This involves a question of much practical as well as fundamental interest. Is this charac- ter of disease resistance a fixed thing or is it variable? In the latter case it would seem that without continued selec- Control of (lABFiAoi-; ^■l•;^I,o\vs 61 tion it might tend to be lost in succeeding generations and that on the other hand by repeated trials and selection there might be further improvement in these selected head strains. It seems, therefore, worth while to present such evidence bearing upon this as we have to date although it is recognized that this is not adequate for final decision upon the matter. In the trials of 1914 (see summary, Table XII) seed of the 1911 crop of the parent head strains X 135 and X 143, i. e. the first selected generation, was used alongside of the second generation selections of the same strains. Table XV gives the outcome. Table XV. — Comparison of^Disease Resistance of Cabbage OF First and Second Generations* X 143, 1911 seed X 143, 1913 seed, average 10 X 143-29 (best 1913 strain) X 135, 1911 seed X 135, 1913 seed, average 5 X 135-3 (best 1913 strain) Average 1st generations Average all strains '2nd generations... Average best strains 2na generations Yellows Living Heads Per cent per cent per cent 19 91 86 21 88 80 4 100 100 20 91 81 16 91 83 1 100 99 19.5 91 83 . 5 18.5 89.5 81 .5 2.5 100 99.5 Yield per acre in tons 12.7 10.5 13. 12.3 12.6 13.6 12.5 11 .5 13.3 * Seed grown 1st generation 1911, 2d generation 1913, trial 1914. From these figures it will be seen that the average of all the second generation seed, gave results practically equal to those from the first generation. On the other hand the best selected head strains of the second generation in each case show decided improvement over the parent stock. The con- clusion from the above data is in general accord with our experience and justifies the hope that by continued care and selection the present standards may be gradually raised. Will disease resistance remain constant in different localities? Another question of immediate practical im- portance is as to whether the disease resistant quality shown by a strain of cabbage as grown in one locality will be shown in like degree, or in any degree at all, in another locality when environmental factors are quite different. In certain cases with other crops, Orton and Barrus (1911) have found that it may not hold. It will require several years of coopera- 02 Wisconsin Research Bulletin 38 live experiment to determine this with exactness. Such limited evidence bearing upon this as is available is encourag- ing. Thus the Houser and the Volga which with us have made the best showing of all commercial varieties, have a reputation in other sections for disease resistance as already noted. The variety of cabbage selected by Close and White (1909) of the Maryland Experiment Station, as disease re- sistant in that state has proved highly resistant to yellows in the Wisconsin trials of 1914, as has been explained on a preceding page. The most conclusive evidence on this point must come from trials of Wisconsin-grown seed in other sections. Small trial samples of such seed were sent to three states: Iowa, North Carolina, and Delaware, in 1914. The results, while not defmite, are altogether encouraging. These trials are being continued.-^ Locality of seed-growing in relation to disease re- sistance. It is also a question of much practical importance as to whether the environment under which this cabbage seed is grown will affect its disease-resistant quality. Thus it is evident that there might be gain from the commercial standpoint in sending our selected seed to the Puget Sound region as "mother-seed" to be grown for one generation, providing it does not thereby lose in disease-resisting quality. This also is a matter which can be determined only by trial. Such trial has already been started in cooperation with the Washington State I^^xperimcnt Station. Until this matter is decided experimentally we shall encourage seed- growing of the disease-resistant Wisconsin Hollander, in Wisconsin or in the locality where it is to be used. Gi:neral Conclusions The results of the experimental work of these five summers seem to justify the following conclusions: On the one hand, no specific method of soil, seed, or crop treatment offers any hope for the control of the Fusarium or yellows disease of ssThese cooperative trials were continued in lOL'i and we are now fortunately able to furnish further very encouraging evidence. Trials were made this year of Wisconsin grown selected strains in Delaware under the direction of T. F. NIanns. in Ohio under the direction of A. D. Selby and J. G. Humbert, and in Iowa under the direction of (',. L. Fitch. In all cases yellows occurred and the evidence 'was convincing. In all these places the Wisconsin selected strains stood up well, whereas the commercial strains snowed much disease. There seems full justification, there- fore, for the conclusion that the Wisconsin grown selected strains will maintain the disease-resistant character at least in large degree when planted elsewhere. Control of Cabbagic Yellows 63 cabbage. On the other hand, the development of disease- resistant varieties by selection has already given such prom- ising results that full reliance can be placed in it as the feasible method for the practical control of this malady. With the variety to which chief attention has been given, the Hol- lander, it seems assured that we have secured by selection a head strain. Villa 25, which not only represents the best commercial type but has also a high degree of resistance to yellows. This variety will be distributed for commercial use under the name Wisconsin Hollander No. 8. While it should be clearly understood that the trials upon which these conclus- ions are based have been limited in the main to three seasons and a restricted region, nevertheless sufficient additional evidence has been secured to show that this disease-resistant quality is inheritable and that it holds, at least in consider- able degree, when the variety is grown in widely different regions. Some degree of variation from generation to genera- tion and with changed environment must, however, be ex- pected and further trials to determine this are already under way. As to variation from generation to generation, it seems probable that if no attention is given to this there may be a gradual reversion or loss of resistance. On the other hand, by continued trial on Fusarium-infested or cabbage-sick soil and selection repeated annually, it seems probable that we can not only maintain the present degree of resistance but improve upon it. Although the experience with other varieties has been com- paratively limited it has been sufficient to give confidence that strains highly resistant to yellows can be secured without serious difficulty from certain varieties, e. g., Volga and Houser, and that with a little more time and attention disease-resistant strains of the standard kraut and summer cabbages can be selected and fixed. It is, however, quite probable that for the best results local selections will need to be made in the leading cabbage-growing districts of the country to secure the best adaptation to local needs and con- ditions. With this idea in mind cooperative relations for the continuation of this work have been established between those interested in the United States department of agricul- ture and in several of the states. While but little progress has been made upon the fundamental problem, the determi- 64 Wisconsin Research Bulletin 38 nation of the nature of disease resistance, and many of the details of more immediate practical interest also require further attention, it is believed that in the main the con- clusions as above outlined will hold and that the practical success in the control of the cabbage Fusarium or yellows by the use of disease-resistant strains of seed is assured. Summary Cabbage-growing with the winter variety known as Hollander or Danish Ball Head has assumed considerable proportions in various parts of Wisconsin, and with the kraut varieties has become an established local industry in certain sections. With the reclamation of marsh lands this industry is destined to increase. A limiting factor to continued success is the disease known as yellows, caused by the parasitic soil fungus Fusarium conglutinans. This fungus invades the roots, either in the seed bed or soon after transplanting, and by killing them and working up through the stem so weakens the plant that it yellows, drops its lower leaves, stops grow- ing, and gradually dies or fails to head. As a result on badly infected or cabbage-sick soil the loss ordinarily ranges from 50 to 95 per cent. This parasite, probably introduced on seed into south- eastern Wisconsin some fifteen or more years ago, has rapidly spread in that section and is continually invading new territory elsewhere, so that it seems destined to follow intensive cabbage culture wherever in the state the con- ditions favor. The disease is worst when a period of dry hot weather follows soon after transplanting, since a high soil tempera- ture favors its development. Once introduced into the soil, it persists indefinitely, so that ordinary crop rotations arc of little avail for its control. Various methods of treatment of seed, seedlings, and soil, including the trial of possible disinfectants and fertil- izers,were without any practical effect. Steam sterilization of the soil was the only specific remedy used with success and this is, of course, impracticable for field use. Control op- Cabbagk Yellows 65 Clean seed, grown in a clean seed bed, and planted in clean soil will give a sound crop, but any contamination of seed bed or field with diseased cabbage refuse or with infected soil, blown, washed, or otherwise carried from a cabbage sick field, will destroy hope of continued success. The only practical method of control, therefore, seemed to lie in the possibility of securing disease resisting varieties or strains. Trial of various commercial varieties shows that there are marked difTerences in susceptibility among them, the Volga and Ilouser showing the highest degree of resistance against yellows. Neither of these is, however, suited to commercial culture in Wisconsin. The standard winter varieties of the Hollander or Ball Head type are especially susceptible and the practical problem therefore, became that of securing a Fusarium-resistant strain of this type. The method employed has been based on the observation that even in the worst diseased fields in the autumn there are occasional sound heads. These have been selected, seed raised from them, this grown in turn on the sickest soil available and those plants which remain sound again saved as mother-heads for seed-growing, By repeated selection we have thus secured strains of winter cabbage of the Hollander type which have proved in a high degree disease resistant against the Fusarium and have at the same time the best commercial qualities. In 1914, the best selected head strain, VHIa 25, gave a prac- tically perfect stand, the heads averaging about 5| pounds each, with a total yield of over 18 tons per acre, on thoroughly cabbage-sick soil, whereas the best commercial strain immediately alongside it showed over 80 per cent of yellows, the heads averaging about 2| pounds each, and yielding about 2 tons per acre. A series of farmers' trials in 1914 showed gratifying practical results. Where selected head strains were grown alongside of the best commercial strains on badly diseased soil, the yields from the selected strain were in some cases ten times that from the commercial strains. The second generation selections grown in this way have proved even more highly resistant than the first. Thus the first generation of the selected head strain VHIa showed at the end of the season of 1912, 96 per cent living and 80 per Four successive cabbage crops on the same field. 7 sin Hollander. Yellows destroyed the crop on t commercial variet FIG. 20.— EXPERIMENTAL TRIALS, COMMERCIAL VARIETIES, 1911 One row of each variety was planted on "cabbage sick" soil. All succumbed to yellows about alike except Ilouser, shown at the right center. (See Table II). FIG. 21.— EXPERIMENTAL TRIALS, RESISTANT STRAINS. 1912 These are resistant bead strains, Wisconsin Hollander, first selection. )tographs show strikingly the resistance^of the Wiscon- -^ in I 9 I 0, and the disease has been severe on year since. ry FIG. 22.— EXPERIMENTAL TRIALS, COMMERCIAL VARIETIES, 1913 ^'n^l'h^^^''"}^"^^''!^^"'^^^ ''^^'y A.o yellows showing that the soil continued to be cabbage sick". (For more details, see Table V and Figure 11 ) ""^° ^^ "^ Fig.-23.— EXPERIMEXTAL TRIALS, RESISTAXT STRAINS. 1011 Thu'i\'^?h^'fif»h°"^"'*^'^' ^"""^^uu generation selection. Compare with Figure 22. l^nH nnH ^/th succcssive Cabbage crop, without manure, on this "cabba|c sick" land and yet the better resistant strains yielded over 18 tons per acre 68 Wisconsin Research Bulletin 38 cent headed, whereas the second generation seed from the heads selected from this crop in 1914, under a more severe test, averaged about 99 per cent Uving and 94 per cent headed, and the best of this series of head strains showed in turn 100 per cent Uving and 98 per cent headed. It is beheved, therefore, that the degree of disease resistance can, by further selection, be at least maintained and probably somewhat increased. Seed is being grown from some 2000 heads of these selected head strains in 1915, which will permit distribution for planting on a commercial scale in 1916. This will be dis- tributed under the name Wisconsin Hollander No. 8. CJiief attention has been given in this work to the winter or Hollander type of cabbage. Selections have, however, been made from certain standard kraut and early summer varieties and this work is being continued in cooperation with the representatives of certain other state institutions and the United States department of agriculture. It seems reasonable to hope from the experience to date that Fusarium- resistant strains of the various important commercial types may thus be secured. Control of Cabbage Yellows \ and the preparat;on of the manuscript. 2 Research Bulletin 42 culture from widely separated regions in the United States. In this paper he gives an accurate and detailed description of the disease. The fungus was grow^n in culture, but from the brief description it is uncertain whether these were pure. However, inoculations produced the characteristic spots in from 8 to 10 days. Following this the trouble was reported by workers in most of the middle west and eastern states. For some time there was much disagreement concerning the true cause of the disease. Some believed the Macrosporium only a secondary invader and the disease primarily of nonparasitic origin, while others con- sidered the fungus a parasite but not the cause of all the trouble. Jones (1893) writing of the disease reports injuries quite sim- ilar produced by paris green. Here for the first time, appears a drawing of a diseased leaf, affected unquestionably with the dis- ease as we know it to-day. At this time he suggested the names early blight and late blight to separate the two diseases. It was not until some time later when Jones (1895, 1896) published the results of further studies that the relation of the Macrosporium to the various troubles entirely cleared up. His field ai:td lab- oratory studies led him to the conclusion that the fungus was a true parasite and the primary cause of early blight. Here also he clearly differentiates the three other forms of disorder which had been confused up to that time under the name "blight," namely, late blight, arsenical poisoning, and tip-burn. Even after this the troubles were not always separated.* Since the work of Jones, very little has been added to our knowledge of the early blight disease. However, during the past two de- cades much valuable data have accumulated bearing uy)on the control of the trouble bj^ spraying. During these twenty years, early blight has been reported from practically every state in the union. Outside the United States it has been recorded from Canada, Mexico, South America, Europe, Africa. Australia, India. New Zealand, Ncav South Wales, and Java. Thus it probably occurs wherever the potato is an important crop. As *As illustrating- the confusion at this time, reference may oe made to I'le Cornell Agrricultural Experiment Station Bulletin 113. 1896, by E G. Lode ■ man. Accompanying a description (p. 2.54-261) of what is called "eai-l> blight" i.s a colored plate of a potato leaf affected, not with early blight, but with a clear case of tip burn. In the text book. "The Spraying of Plants," by the same author, the illustration on page 3 46, labeled "early blight" represents a typical form of arsenical poisoning. Early Blrsht of Potato and Related Plants 3 to whether the parasite is native to the potato and has spread with it from its original home in South America to the various countries into wliich the potato has been introduced is largely a matter of speculation. However, Jones (1903) reports finding it on specimens of wild potato from Mexico. Economic Importance It is practically impossible to determine the actual loss caused by early blight, owing to the fact that the situation is usually complicated by the presence of tip-burn, arsenical poisonings flea bettle injury, or late blight. Results from spraying experi- ments furnish no accurate l)asis for estimating the loss since bordeaux mixture reduces at the same time the influence of all the other troubles on the vines, and may in itself furnish a stim- ulus to greater vigor. All reports show that the disease is of greater conseciuence in the United States than elsewhere, with the possible exceptions of Australia, Rhodesia and New Zealand. Jones (1903) states that in certain seasons Alternaria solani causes more loss in many parts of New England than does the mildew. Several eases are on record of unusual attacks, but more important, however, is the smaller but yearly toll of the disease. Coons (1914) averages the annual loss in Michigan as about 25 per cent. In Wisconsin Jones (1912) states that it may reduce the yield 10 to 25 per cent. The writer considers these figures a consei'vative estimate. In the southern states, early blight has been reported to at- tack seriously leaves, stems, and fruit of the tomato. Edgerton and Moreland (1913) , in Louisiania, state that it is a close second to "wilt" in destructiveness and in many regions the "all im- portant disease." In one tomato district they estimated a loss of 50 per cent.* Though the disease transfers readily to the tomato and may be found almost every year in the northern states, yet it appears to do little damage. The w^riter has, how- ever, found it in both the Chicago and Madison local markets as the cause of a severe rotting of tomato fruits from the south. The evidence here indicated that the disease had developed dur- ing transit. In the summer of 1916 it w^as isolated along with Gleosporium phomoides Sacc. from decaying tomato fruits at ♦Isolations of the fungus fron fresh material received from Dr. Edgerton. in July 1916 confirmed his diagnosis of the trouble. i Research Bulletin 42 Waupaca, Wis., but which fungus was primarily responsible for the trouble was not determined. Inoculation studies reported later in this bulletin show that A. solani is capable of producing a spotting no wise different in appearance from that on natur- ally infected fruit. The disease has been found the past two seasons on eggplants in Wisconsin; it seems, however, to be of little consequence es- pecially as compared with the leaf spot caused by Phomopsis vexans (Sacc. & Syd.) Harter.* In June, 1916. it was found to be causing a serious blight in seed beds of this host at Eau Claire, Wisconsin. It was learned that the hot beds had re- mained for a number of years in the same place and that it was the practice to sprinkle them frequently with a hose. These fac- tors operating on the crowded, more or less etiolated seedlings may account for the rapid spread and severity of the trouble. * ^ TIG. 1.— EARLY BLIGHT OF POTATO Tlic loaf is soon we.ikened from the en- largi'ineiit of the spots. (Photograph by L. R. J ones 0 Symptoms The appearance of the spots on the leaves of ea3h of the three common hosts is very similar. They are dark brown or black and show Uvsually a series of concentrie ridges which produce a ''torget board" effect. (Fig. 3) There is often a narrow marginal faded zone which spreads out- ward as the spot enlarges. Tlie spots are usually oval in shape but under unfavorable conditions, especially on a vigorous leaf, may remain small and angular conforming to the spaces between several small veins. (Fig. 1.) The spots usually enlarge after the death of the leaf. On the tomato the disease may be * Alternaria solani was first recorded on egg-plant by Chester (1893) in Delaware. Later it was listed by Clinton (190 4) in Connecticut. Early Blkiiit of Potato and Related Plants easily mistaken for the leaf spot (Septoria lycopersici) wliich has been much more common on Wisconsin tomatoes during the past two seasons. Without the aid of a hand lens the spots on the egfj-plant are almost indistinguishable from those caused by PJwmopsis vexatis. Early bliglit on the potato is readily dis- tinguished from arsenical poisoning by the darker color of its spots. With tip-]jurn the leaflet usually shows apical or mar- ginal burning and the concentric rings are absent. There is still less resemblance to the late blight because of the whitish fructi- fication of the venti-al surface of leaves affected with the latter trouble. Potato plants may be at- tacked by early blight at al- most any stage of their ex- istence, but, under ordinary conditions, the disease is not able to gain a foothold until tlie vines have passed their period of greatest vigor and are directing their energy to tuber formation. Before this time, close scru- tiny will generally reveal an occasional spot on the lower, older, and more shaded leaves of the plant. Such leaves have frequently been covered and uncovered (with soil) a time or two during the process of cultivation and are conse- (juently yellowed and weak- ened. Under favorable con- ditions the spots increase rap- idly in number, and the leaves beginning with the low^r ones gradually die until only a few green, spotted leaves remain at the top of the plant. (Fig. 2.) In severe cases spots develop on the petioles and upper stems of the plant. J „^ ^ ^"1^ ■^^^v .J^r sO WLj^mtmmmylmw /^J|' ^H^ - ■ I' > :>;- ":y:--"*^^^';r FIG. 2.— A SINGLE HILL OF POTATO DYING FROM EARLY BLIGHT Early Ohio planted April 28, photo- graphed August 12, 1915. Note the pro- gressive curling and drying of the leaves from the ground upward. 16 Resi:arch Bulletin 42 Studies on the Host Range of Alternarlv Solani The primary object of these studies was to determine whether the leaf spots of potato, tomato, egg-plant, and Jimson weed {Datura stramonium) , which have been ascribed to this fungus, were produced by one and the same species of Alternaria. Jones (1896) proved beyond doubt the parasitic relationship of Alternaria solani to the early blight of potato, but its con- nection with the other plants has never been conclusively shown by inoculation tests. The failure of inoculations on Datura and the comparative studies of Alternaria solani and the Datura fungus show that the latter is a distinct species, bearing no sim- ilarity to A. solani in its host relationship. The results are pub- lished elsewhere (Phytopathology 7: 327-337. 1917)* The see- ondaiy object was to determine within what limits the parasitis^ii of Alternaria solani is confined. During the summer of 1915. pure cultures from single spores ivere obtained for inoculation purposes from potato, tomato, and egg-plant growing at Waupaca, Wis. They were later grown comparatively on fifteen kinds of agar media and in appearance were practically identical. Abundant spores for inoculatiur.:; were obtained from each by :i method later refei'red to. Greenhouse Inoculations The following inoculation methods were used with more or less success in greenhouse experiments made from February to May, 1916 ; temperature 19 to 23° C. (1) Drop of heavy spore suspension placed on flat portion of leaf inclosed by round cover slip. Plant placed in glass moist chamber for 48 to 72 hours. (2) Spores or mycelium introduced into needle punctures. Plant placed under bell jar and atomized frequently witli water for 48 hours. (3) Leaves atomized with spoi-e sus]-)onsion and for 4S hours kept moist by fine spray from nozzle. • This Datura leaf spot which has been widely attributed to Altermiria solani is shown to be due to the fungus named Cercospora crassa by Saccardo in 1877. Examination of type specimens collected by Saccardo and of exsiccati from various parts of the United States show that the fungus was named from immature material and is really an Alternaria. The new combination. Alternai-ia crassa, with technical discription is given in the article in Pytopathology referred to above. Eakfa' P>i,i(;iit ok Pot\to and Related Plants TaULK I.— 'iltKK.VltJL'.SK IXUCULATION'S. M.AUISOX. l'J16 Date Source of inoculum ,„ . , , » , I Method oi I'laiil Inoculated: i„ocula- condilion, etc. tJoQ Results F.-1). ■>.:> Apr. 13 Apr. 13 May 7 I'olato strain. .4. s*)/((»ii;m.vcfl on bits of atjar I'otato strain. .1. .siifaui; spores fioui culture Potato strain, A . .solani: spores fi-om culture Potato strain, A. stAani: spores from culture Potato-2 plants S-lOiii.liiifh; vi},' 10 leaflet.-, inoc. No. 2 March 3. 80'^ infection: sp(jts 8-20 mm. diam on both plants Tomato-1 plant vitr. 10 leaflets inoc. No. 2 March 3, 75% Infection: .spots 4-6 mm. diam. Sutanumniorum- 1 plant 4in.higli: No. 1 April 20. 90*0 with spots 1-4 mm. dlam. I'"trsplant-l plant 3 in. hij^h; vi^. No 1 April 20, 100% infection: .spots 1-2 cm. diam. White Burley tobacco-2 plants 6 in. higli: vig. No. 1 April 20, few spots 1 mm. diam. no further enlargement Potato-1 plant 10 in. higrh: vig. No. 3 Tomato-1 plant S in. hitfh: vig. No. 3 Esrgplant-1 plant 4 in. high; vig. No. 3 April 20, April 20, April 20, minute spots on every leaf: wet continuously few spots: not wet con- tinuously many spot son every leaf: wet continuously l'otato-1 plant 14 in. high: visr. Tomato-2 plants 16 in, high: vig. No. 3 Solamim nlgnim -2 larsre plants: fairly vig. No. 3 Jlay 14. many spots 1-3 mm. diam. May 14. few spots on lower leaves, 2-3 mm.' diam. No 3 May 14, few spots on lower leaves, 1-5 mm. diam. Apr. 14 Apr. 14 Eggplant strain: spores from cul- tu re Eggplant-1 plant 5 in. Iiigh: very Potato-1 plant 12 in. high: vig No. 1 April 20, 100% with spots 3-4 mm. diam. No. 1 April 20, 90% with .spots 1-3 mm. diam. Tomato-1 plant 12 in. high: very: vig. No. 1 April 20. Tomato strain; spores from cul- ture Tomato-1 plant 12 in. high; veryi Potato-1 plant 15 in. high: vig. No. 1 No. 1 April 20, April 20, 80% with spots 2-3 mm. diam: enlarge very slowly 100% with spots 2-3 mm. diam. 100% with spots 3-4 mm. diam. Effgplant-l plant Sin. high: vig. [ No. 1 April 20. 100% with spots 6-8 mm. diam. 8 Research Bulletin 42 These experiments are briefly summarized in Table I. In most cases reisolations from the infected plants were successful. The results show that in the majority of cases Alternaria solani from potato crossed readily to tomato and egg-plant, to some ex- tent to nightshade {Solanum nigrum), and to cultivated to- bacco. In the latter case, penetration occurred, but the mycelium seemed to be unable to spread in the tissues of these vigorous seedlings. The strains isolated from tomato and egg-plant reciprocally crossed quite readily and both in turn produced a spotting of potato in no wise different from that of ordinary early blight on potato. Aside from a few explainable exceptions the uninocu- lated needle punctures healed, and in method 3, the plants ex- posed beside the inoculated plants never developed spots. There- fore it seems justifiable to conclude that the early blight of po- tato, tomato, and egg-plant are caused by one and the same or- ganism, viz., Alternaria solani. Owing to the difficulty of working with mature plants in the greenhouse it was decided to continue the tests under field con- ditions. Field Inoculations Field tests were carried out at Waupaca in central Wisconsin during the summer of 1916. In order to determine within what limits the parasitism of this fungus is confined, it seemed de- sirable to obtain a wide range of plants, especially as to genera, of the potato family. The effort was successful only to a limited extent because it was impossible, on a few months notice to get seed, particularly of the wild members of the family.* Eight to ten plants of each species and variety were properly spaced in rows three feet apart, with every third row in po- tatoes to furnish a basis for comparison. The potatoes were planted May 11 and the other plants were transferred from the greenhouse in early June. The severe and prolonged drought dur- ing July and August proved a serious setback, but by artificial watering most of the plants made normal growth. Prior to Sep- *The writer is indebted to Messrs. Peter Bisset, Plant Introducer U. S. Dept. of Afcriculture. Goo. T. Moore, Missouri Botanical Garden, St. Louis, and W. S. Oswald. Minnesota Seed Labratory for seeds or plants furnished for this work. ^^\lvl.^ liij(;in' oi- Potato and Rklatkd Plants 9 tciiibcr 4 t'onditions for natural infection were very unfavorable and spots which appeared earlier on the potato did not spread. On account of the extreme heat, artificial conditions for infection could not be maintained with the means at hand. After August 15, several plants of each species were atomized occasionally with spores in order to have the plants ready for rainy weather when such a favorable condition for infection should arrive. Spores from pure culture of the potato strain were used. Sep- tember 8, several leaves on selected plants of each species were inoculated by the needle prick method, i. e. by placing a drop of heavy spore suspension on each puncture but always leaving an equal number uninoculated for control. The drought was broken on September 4 when a period of moist weather with heavy dews and rains set in, furnishing ideal conditions for in- fection by Alternaria solani. The main results from these field inoculations are presented in Table II which shows: (1) size and condition of the plants on September 19 and (2) the progress of the disease two weeks after and one month after the beginning of the rainy period. In most cases an attempt was made to reisolate the fungus from the smaller spots even where sporulation occurred on large spots of the same plant. In several instances, it will be seen that the fungus was not reisolated though spores are recorded for the larger spots. This was probably due to the presence in the plates of the saprophytic fungus. Alternarui fasciculatn, which is the more rapid grower and is difficult to eliminate. In- oculations on Nos. 3, 4, 5. 9, 13, 14, 24, and 26 were repeated in the pathological garden at Madison, Wis., in September and Oc- tober, 1916. As the results agree in all essentials with those tabulated, for the sake of brevity they are not listed here. The table .shows that the fungus was able to penetrate almost every plant inoculated. Even the leathery, succulent leaves of S. grandiflorum and S. guttata were infected as was the potato. Leaves of the former inoculated September 2 were found to be thickly peppered with tiny infection spots September 9. These spots, which measured less than two millimeters in diameter, had made no enlargement when the leaves were again examined a month later. Yet when cultures were made from such spots, October 14, almost every one developed the fungus. What checks the advance of the fungus in the tissues of these plants is not 10 Research Bulletin 42 Oh 03 O H <1 1-3 O to & M o < o m 1 I 1 i^S 1 1 1 o - = 1 « C 3 1 " ■ o«-. 1 m 1 w ■/) X 1 « 1 tc 9] tr. (S^^ 1 ® 01 o « 1 * 1 ^ c o ^^ 1 "^ >H >^ 1 ^ pn >H >* 1 1 ^^ 1 .^ 1 ■*-* 1 1 t il C e c 1 1 1 1 c« 1 "^ 1 ^ 1 e« 1 '3 1 1 1 1 1 1 2S \ 3 o ■^ 1 3 3 1 o 1 o 1 o 1 *^ IS 1 1 ■c •/I O s IS \i , 1 m « . ■• S s 5 C 1 *- 1 "a 1 ci2 Ot3 05 3 > c o .5 ■« > 2 a a ° 1 o PQ O .2- c g o oo 1 = o u O O 2 o O a s^J. j • 1 O D >• OI 'O 01 — s 0) B s reen -4 mm., w leaves diam. Id leaves . in dlam. S leaves diam. «3 C a c o a* £ « (S eS i • b( C« tX" o • Spots on 0 8-15 mm on youn, 1-4 mm. tu ilr- y N s o < "5 0) o /5 "S 0) o Spots on leaves. on yell 6-8 mm o Noenlar even oi leaves c II CA Cfl 73 te W 0 1 V o c 9 a c: yjS b< a ■J) o> C (1) a § o o .2 -a = 1 .2- 1 1 C.2 3 a afc -P P3 .- O oi CA a -C 0 \i £a c g 2 o 3 C c » is H H ^=S= rt o^ ^2 O^ o 02 o 3 ?; o oo o ^ ss s ss s 0) a -5 a g a a a o EH t) CO OB 1 s > o T3 Mi2 a ■ a in H cc g ?a 0) 0.2 _' c S e /^a li o S's t»!S a 73 _UM ^ 112 < §^ a 1 o a a^ M s Cu in H (Ti [K tn ;o> 0) 03 5^ 2.0 1 ^ Oi '5 C r, — 3 Oi 3 2 o c ^1 ^1 is 1/' ^ 1 a 3 -^ sill .i'.c o 1 "is i;2 if 52 high; SI ;ulent, v leather, edshrui > c < - X is5 oo > oo £.1 e o 2 rt - 3 3 S 00 w 0.2; o en ^8 S 3 0, S2 ao to tM X ■a 1 .„^^ g« u »2 in ttei. sn mderbei K. 3 e « ;j 0> 2cJ: s s § 6 s . 1 if So 15 c «^ o « c » Q O S = c3 -^ ^.5^ ft Jo o-r; ai 'It =« & |S 1. (» VD «i y/ c/: yi 1/3 i/j - ~ m "*' in tt> •^ oo' Early Blight of Potato and Related Plants 11 «-2 to — o-o in gisa la. II — ■/! 5: W i "■a bi a 01 3 fc-2 ! 2^ £E > e4 ra a S2 3 C oB I B ca Ooo ?" . V4 o ce es a-— O) O fa -a ..a C o C 3 iUO a o 5Sa o ..:s 3a OB 3) aco JJ-CO 100% infection; spots 8-10 mm dlam. Sa li E2 c«a a S lo- a, CO Leaves dropped 100% Infection; spots 8-10 rara. diam. 100% infection: spots 3-6 mm. diam. 100% infection: spots 3-4 mm. diam 100% infection: spots 1-4 mm. dlam. . Faint browning al30ut punctures Leaf dropped No inoculation 1. ^a ia 00 Spots .5-10 mm. dlam. Thickly peppered .5-1 mm. dlam. 10-15 spots per leaf 3-5 mm. dium. Many spots, 3-5 mm. diam., coalescing Peppered with spots 1-4 mm. diam. Thicldy spotted 1-4 mm. diam. Few specks, .5-1 mm dlam. Few siX)ts on yel- lowed leaves 2-3 mm. diam. 1 Few small spots a a 0 1. ii eS-a ■^ C CM « C > ^ 2.2 0 £ 3 • .. •- a *:> 0)^ *- C 3 -T e e« CO 18-24 in. high; fairly vigorous 12 in. high: young upper leaves vig- orous 1-3 ft. high: very vigorous 2 *t. /Igorous leathery leaved woody vine 8-10 In. high; very vigorous 12-16 in. high; much weakened b.v drought and red spider 1 ^ '3 si T. b£3 CO > 1 ~ d •a ea Id d i c 0 U.S. niorumuar. j;uiHeri?t«e Linn, garden wonder- l)erry 12. S. rostratum Dunai. buffalo burr 1^ SCO "5.22 CO 2 is 02 s 16. Physalitfran- cheti Mast. Chinese lantern plant g-Co s ^•■3 a. 1>> 1=2 a. i^3 S.Sa 12 Research Bulletin 42 So ^ rd ♦^ ® o 3 > a S3 CO_Q I &g^ I I a x 3 "" /I tS .2 3'3 oB-S la . -j^ OS a i 34JT3 S 1) oSTJ . I ^£ ^ S o-oi Z 00 C en •r 0) 3 > 3 r^ J= c a- 03 cS n -^ 3 r: 1^ p bi o 5 .tJ R -o 03 oa ti 0> 'O t/J c c 03 83 ^ a 0) "" ^ fc.' s • 3 o; ts ^ 3 03 5 o t« 3 t< c aj3-r- Si ^ r- *" a Is a 7) . O 03; ' i--a C/j ^■0 3^ —J 7) ^ 3 . O a. 22 3 a as j; ! i-oa ■ £a 53 — am CL, 03 I a 03 O ■/! . fci 71 03 a 03^ > a a a 03S S l^a ea S3 H >H Z t^ a c« a O o 3 J3 < I 5= o c'O a c (U iR-l.§ I C3 15' •a 5 Sc4 C4 S.2 V — ^ Sr a 0 » s§ ss s-^ a, o tl = ^ z ^ s ^ s /;X! p bi « 0) ^ > !X > JS J3 tit •-■a &t iio; c c o o t- c — je 0) S3 -H O 30 ^' e a _ -o c ^^ ax: I S ca = ■^^■- I IS** s =" e S 3 o ■^ « S— * 14 Research Bulletin 42 known. It is believed that such plants, i. e., those on which the spots do not enlarge, should not be considered as hosts, since on them the fungus does not produce spores and therefore can- FIG. 3.— POTATO EARLY BLIGHT SPOTS ENLARGED X 3 Thfi5-> show the typical black target board appearance. (Photograph by H. H. W'hctzel.) not complete its life cycle. On this basis the hosts of Alternnria solani determined by these studies are listed below.* 1. Solanum aviculare Forst. 2. Solanum varolinensifi Linn. — Horse nettle 3. Solanum giganteum Jacq. 4. Solanum mclongena Linn. — Egg-plant 5. Solanum nigrum Linn. — Nightshade 6. Solanum nigrum guinennse Linn. — Garden wonderberry •Though not included in these studies it is probable that the two followins species are also hosts of .4. solani. Solanum cnmmerscni Dun. l-sted by Nu.s.'lin (1905) and Stuart (19'' 4). Hyocyamus albiis Linn. White Henbane, according to Ferraris (1913). Early Bmoiit of Potato and Rk[.atki) Plants 15 7. SaloniDii rostratiDii Dun. — Buffalo burr 8. Solanum tuberosum Linn. — Potato 9. Solanum ivarsceioiczii Hort. 10. Hyovyavius nigcr Linn.— Black henbane 11. Lycopersicon esculentum Mill.— Tomato 12. Nican(l7-a physahAdrs Gaertn. — Apple of Peru FIG. 4. — DIAGRAM.ALVTIC REPRESENTATION OF A PORTION OF SPOT ENLARGED The invaded tissue shrinlcs to about one half the original thickness of the leaf, and the surface is thrown into concentric ridges. The cells are darkened. Spores are pro- duced on both surfaces as sho^ii above intermingled with the hairs. During i\Iay, 1916, inoculations were made on tomato fruits of various ages freshly picked from greenhouse plants. In two 16 Research Bulletin 42 trials si)orcs atoiiiizod on the surface failed to give infection after 10 days even though the fruits were moistened frequently and kept in a damp chamber. Under the same conditions needle puncture inoculations invariably resulted in infection. After 15 days there was only slight invasion about the points of inocu- lation on the green fruits while with the ripe fruits almost com- plete rotting resulted. McCubbin (lOlG) in Ontai'io. leports similar results from inoculations of tonmto fruits. Needle puncture inoculations were made on mature fruits of egg plant and green pepper during August, 1916. In each case slight inva- sion of the tissues about the punctures occurred but no en- larged spots or decay resulted. Of the nine genera of the potato family tested, four only were found able to perpetuate the fungus, viz., Solanum, Lycop- ersicon, Nieandi-a and Hyocyamus. Of these the Solanums though showing considerable variation appear as a group to be the most susceptible. From these experiments it is evident that A. solani is not restricted within very narrow limits in its host relationship. Pathological Anatomy An explanation of the "target board effect" (Figs. 3 and 4) characteristic of this disease is suggested by Jones (1896). He believes that such a condition is produced by the more complete collapse and rapid contraction of the interior cells or mesophyll as compared with the epidermal cells. A study of microtome sections of spots in various stages of development shows that greatest contraction occurs in the spongy tissue which would tend to throw the upper part of the leaf into concentric folds. In the spot the cells are collapsed, shrunken, and deeply stained (Fig. 5). No evidence has been obtained to show that the failure of small spots to enlarge on vigorous leaves was due to suberized layers or other mechanical hindrance to invasion. On the contrary, all evidence indicates the resistance to be di- rectly related to the vigor of the leaf. Though the fungus has never been actually observed inside the cells of the host, there seems no reason to suppose that it cannot enter them. Penetra- tion of the leaf usually occurs directly through the epidermis, and in pure culture the fungus can utilize cellulose when this is offered as its only source of carbon. Kaki.v HhKiiiT OK Potato and Rklatko Plants J7 The Causal Organism taxonomy 111 tlu' litei'aturc on early blight the fungus is commonly re- ferred to under the following names — Macrosporium solnni El- lis and Martin, AJtcrruiria snUini (E. & M.) Jones and Grout, and Alternarid solani Sorauer. In foreign references the latter is in more general use while the second occurs most frequently in accounts of tlie disease in America. Sorauer (1896) published FIG. 5.— CROSS SECTION OF LEAF SHOWINC INOIIIPINT INFECTION OF ALTERNARI A SOLANI Pc>netration usually occurs directly through the cuticle. Shrinkage follows the death of the cells. X4 0. on the fungus a few months in advance of Jones (1896), but hi* observations and illnstrations of spore chains, as he found them in crude hanging drop cultures, show plainly that his descrip- lion was based on Alternan'a fasciculata. From type material received from Sorauer. Jones separated the two fungi, the one a typical Alternaria and a saprophj'te which he subsequently named Alternaria fasciculata, and the other the true parasite {Macrosporium solani). Jones reports frequent cases where spores in cultures of the Macrosporium were joined in catenu- late pairs after the fashion of the Alternarias. He then writes a 18 Research Bulletin 42 technical description and gives Sorauer the credit for the new combination. Seymour (correspondence), however, later ruled that inasmuch as Sorauer had applied the binominal confusedly, authority for the new combination should rest with Jones and his assistant (Jones and Grout 1897). This is the usage of Far- low (1905) and of most recent American authors. Mc Alpine (1903) and Duggar (1909) have objected to calling the fungus an Alternaria on the ground that the catenulation of spores does not occur in nature. The author has examined many spots and FIG. 6. — MATURE SPORES OF ALTERNARIA SOLANI Hi)or;s tirawn from pure culture where catenulation has been noted. X200. has never seen catenulation on the leaves. It is true, however, that on oat meal agai* cultures, spore pairs frequently occur (Fig. 6), In view of the chaotic condition of the literature deal- ing with Macrosporium and Alternaria and the slight and un- certain distinctions between the genera, the author considers it inadvisable to break away from the well established usage and go back to Mascrosporium. The following is the probable synonomy of the fungus with citations to the literature: Altcriiai ii soUiiii (E. & M.) Jones and Grout. Bull. Torrey Bot. Club 23:353. Sept. 1896. Vt. Agr. Exp. Sta. Rept. 10:45. 1896. Macrosporium solan i E. & M. American Naturalist 1(J:1003. 1882. Macrosporium solani Cooke, (in part) Grevillia 12:32. 1883. Macrosporium cookci Sacc. (in part), (following Cooke) Sacc. Sylloge Fungorum 4:530. 1896. Early Blight of Potato and Related Plants 19 Alternaria solani Sorauer (in part). Zeitschr. fur Pflanzenkrankheiten 6:6. 1896. Sporidesniium solani var. iHirians Vanha. Naturw. Ztschr. Land - u. Forstw. 2:113-127. 1904. MORPHOLOGY The mycelium at the margin of the spot can be seen, using in toto fixations, as slender, radiating, sparsely branched filaments. Later it becomes closely branched, irregular, and deeply stained. Conidiophores have never been found arising nearer than one- H FIG. 7.— SPORE DEVELOPMENT OF ALTERNARIA SOLANI Progressive stages commencing in the upper left hand corner. Note that the spore starts by budding from the tip of the apical cell of the conidophore as shown in the second and third stages. Drawn from culture XIOO. half a millimeter from the boundary of green tissue. Usually one spore is produced on a conidiophore. The conidia arise from the conidiophore, not by the constriction and subsequent enlarge- ment of a terminal cell, but from a bud which forms on that cell. (Fig. 7.) The first indication of the bud is a faint hyaline area on the wall. Soon (often within a few minutes), the wall at this place pushes out and forms a minute projection which has an ex- tremely thin wall and is less than one-fifth the diameter of the conidiophore. This bud grows very rapidly at fii*st and, on this account, the early stages are not easily followed. 20 Research Bulletin 42 A method for obtaining abundant sporulation in pure cul- tures of this fungus is described elsewhere.* Spores thus pro- duced show greater uniformity in size than those from the spots. "iVIeasurements of 100 spores from large, typical early blight spots on potato leaves gave a range in size of 120-296 x 12-20 microns, an average size of 200 x 17 microns. The same number taken from several pure cultures on potato agar gave a range of 104—184 X 14-18 microns, an average of 141 x 16 microns. Nothing to date has indicated the existence of a perfect stage of this fungus. Overwintered material has been examined and the fungus has been grown on many kinds of media of varying degrees of acidity and exposed to various temperatures but no indications of another stage have developed. PHYSIOLOGY Alternaria sohmi is easily isolated from the spots or from spores, and grows well on all the ordinary culture media. Per- haps the most striking physiological characteristic of the fungus is the intense discoloration which it produces in the medium. On potato agar, young colonies cause a clear yellow pigmenta- tion which, as the colony enlarges, spreads in advance of the my- celium and is eventually succeeded beneath the older part by a deep wine color. In media made +20 Fuller's scale, the colora- tion approaches a deep brick red in some cases. On slightly acid media the yellow pigmentation predominates and it is practi- cally absent in alkaline media, where also little growth occurs. There is likewise no discoloration when the fungus is gi-own on + 10 to +15 casein agar, nutrient gelatin co)itaining dextrose, starch-nitrate agar, and cellulose agar. After 7 to 10 genei'ations in pure culture the pigmentation is much diminished and in some cases has been observed to almost disappear. The fungus readily liquefies the above gelatin medium and shows great proteoclastic activity in the utilization of casein as iiulicatcd by the clear zone sui-rounding the colony when lactic acid is added to a casein agai- plate. Nitrates are quickly re- duced to niti-itcs and even to ammonia when tested on starch- nitrate agar. ♦PTiytopatholosry 7: 316-317. 1917. This niethod consists first, in sevrrely wounding' the mycph'um by .■'hreddins a ton day old culture of the fungus on potato agar, and irecGnd, for 24 hours, controlling the moisture relation so that the surface dees not become dry. Early BLKiirr ov Potato and Rklated Plants 21 Temperature relations. — Both spore germination and colony growth of .1. solani arc greatly intluenced by temperature. At 20 °C., ordinarily five to ten germ-tubes arise from the different cells of a single spore, while at 1-3 °C., germination will finally occur, but with no more than 2 or 3 germ-tubes. (Fig. 8.) Spores geiTuinated ;it a low temperature generally produce several FIG. 8. —GERMINATION OF SPORES OF ALTERNARIA SOLANI Temperature is an important factor in determining the number of gsrm tubes and the rate of germination: the two spores to left with the great'T number of germ tubes after 1 V-> hoiu-s at 35° C. ; the spore to the right with fewer gei-ni tubes after 46 "hours at 1-2" C. more germ-tubes when removed to a higher temperature. Ex- tended studies have been made of spore germination in agar under seventeen dift'erent temperatures ranging from 2 to 45 °C., in which at intervals the approximate length and number of germ-tubes were determined. These results are plotted in Fig. 9. At all temperatures from 6 to 34° C, the spores germinated within one and one-half hours. Germination took place most rapidly at 28-30°, requiring at those temperatures but 35 to 45 22 Research Bulletin 42 1 minutes. The germ-tubes formed at 37° were irregular and knotted with bladder-like swellings at the tip. Growth entirely ceased after six hours and subsequent transference to a lower temperature showed that they were dead. At 45° the spores were killed before any indication of germination appeared. ZBO / / -• / / ' / / ' J / ' / / * / / / ' / v,'« - / / / / 1 / / // s / / °^ '/ .<0 i J^ * i / ^''/ ^200 - / / / // •^ 1/ "^ / »^/ y '/ 5 / y / ''/ ■^ / / J • >^ ^160 - / /^ ^ / //'' ^' X y C / / / ' *c> ^ ^^ .^^ y ^ / / A' '^'v' y^ ^y y^ ^ yXy^ s.^120 - y"^ y^ Ci / / / ' y^ '(^ ^y ^y ^ / // ,Y ^ ^^C^ x^ ^ 1 / / '/ y^""^ "y^ y^ -^ 80 /y / y^ J^^^-^^^^ ^::^^ g^lgli--------'''''''^ ''/y S ^^^^^^^^^^^ (^'i-- \/X GROWTH OF ALTERNARIA SOLANI At most temperatures germination commenced during the first hour. The optimum is 26-28" C. At 45° C the spores were killed. Measurements of colonies grown at these different tempera- tures give a graph similar to that obtained for spore germina- tion. However, no growth visible to the naked eye took place at 3° or at 45 °C., while at 37° there was a slight amount of aerial mycelium. The cardinal temperatures of the fungus are there- fore approximately as follows: minimum 1°-2°C., optimum 26°-28°, and maximum 37°-45°. rjAULV liLKillT OF POTATO AND RELATED PLANTS 23 Life History of A. Solani in Relation to Early Blight SEASONAL development OF THE DISEASE Tlie time at which this disease makes its appearance each year seems to depend largely upon the date at which the crop was planted and upon its subsequent development as influenced by soil and climate. It may be safely concluded that as soon as the crop has passed its stage of greatest vigor and tuber formation has begun, early blight may develop. Whether or not the at- tack becomes severe depends almost entirely on influencing fac- tors later enumerated. SPORE PRODUCTION Spore production is usually delayed until after the death of the host tissues. Very rarely are spores found on spots less than four millimeters in diameter. Both upper and lower sur- faces of a spot produce spores, the upper much more abundantly, however. They are very easily dislodged, especially by rainfall. While considerable variation has been noted in the relative abun- dance of spores on spots of different sizes, it was desired to get some idea of the actual numbers which may occur. For this purpose, spots developed under as favorable natural conditions for sporulation as possible were obtained and counts made. Each spot cut carefully from the leaf was rinsed in a given volume of water which, with a small amount of leached agar, was poured into a level petri dish. One-tenth areas were marked off with a bacteriological counting card. After germination had begun the spores in two such areas were counted by means of the low power of the microscope. The following results were ob- tained : Diameter of Distribution as noted on spot Number of spot spores 10mm. Apparently equally abund. on both surfaces 1475 7mm. Few below, abundant above 930 10mm. About one-half as many below as above 785 5mm. Few below, abundant in center above 415 8mm. Sc■^ttered on both surfaces 140 6mm. Few on both surfaces 115 These figures may give some idea of the abundance of spore8 which may be produced on a badly diseased plant with, for in- stance, ten to fifteen spots on every leaflet. The total number 24 Research Bulletin 42 OO tl i-s Z o H Q O o o •< o o B o M B ■< 2; 02 o H H O rajBAV •jBaio 'Map ONj •qisz 15x18 mm. sp. -H-+ ab. in belt8-10mm. from center of spot; +++ bl. C 'S > c o a 3 raJBAi •jBap 'Map 1 ■m d e asoq qjiM .V[!ABaq pa.ia -IBM s^aBjd ;m.i'BM •jEap 'Map o>j maz 14 mm. sp. -|-++ab.; -bl: If. partly dead No change; sp. -|-+ ab. bl. esp. in later growth rings i?puiM 'iSpnop 'Map /absh qiez 1 1 •m -e zi-oi (SO') "'"BJ 'Apnoj.i 'Map iq^ri "m^z 14 mm. sp. o ab: ++ bl. esp. on veins and later growth rings CO 12x18 mm. sp. o; If. dead ui.iBM 'aBa[o 'Map ciqan pjgz ra.reM 'jBap 'Map ?q3!l ^jaA puzz 01 c 0-I3 S X T. Unchang- ed, sp. o; If. dropped 7x9 mm sp. o: If. dying S g « a CM 7J m.iBA\ pii« .iBap Avap AAieaq Sda\ -jsiz 10x15 mm. sp. —scattered ab. bl; Iflt. dead 7x8 mm. sp. ++ ab. bl. esp. along veins: If. vig. 12x19 mm. sp. -I-+ in outer rings ab. -bl; If. dead a •- .■"if c.a cfl E rt u hi ^-t-g S3 g4- •2+ (3i raJBM 'juaja 'Mapiq^ri 'moz 1 (in"!.! a.ioj -aq ap^ra'saiOM) •(21) >«I«i 'Vpnoio •va d ra.iBM '.iBap 'Map iqan qiei 9x15 mm. sp. — over cen- ter ab: scat- tered bl. (not removed) a _. ■^ ..3 S 1 + 11x19 mm. sp. — scat- tered ab; -|-+ in places bl. s go a So 2^ ra.iBM 'jBaio 'A\ap i?ABaqmn]pai\i 'qigl 8x14 mm. sp. +++ ab. bl; If. yellow- ing 3 CO ^5 . o ■■« Si X 5 ill. raJBM Mijap •ra d (•lu •« sjaMoqs) gO^ uiBj 'Apn()|.> 'MapjtA'Ban qiii 8x12 mm.sp. o ab; ++bl. ga^- m.iEAv 'jBai.T 'MapAABan qigi tn.iBM 'jBoio 'm "B 6 iiuin pM SAI 'gz- aiBJi qisi macM 'jBap 'Map SAV^^ 'z ui^H qin raJBM 'jBap 'Map iqan 'qigi 5x8 mm. If. vig; SP. + ab. - bl. •SON ^otls puB juaT c« .a .^ Early Blkjiit of Potato and Related Plants 25 B+M ■ C "^ ira plants of the Green Mountain variety, bearing many spots on most of the leaves, were selected. Plant A was pro- tected from dew at night, and from rains, when imminent, by placing over it a dew-proof cage. Plants B and C Early Blight of Potato and Related Plants 27 were not protected. Tlie spots were examined as in the pre- vious study, but instead of removing the spores with water and camel's hair brush, the brush alone was used. This method was ('((ually effective while being easier of manipulation. The results from this series are shown in Table IV. Fortunately a rather dry period was selected for this study which made it possible to determine the effect of the single factor, dew, on spore forma- tion. Prior to this experiment, the writer had believed that moderately heavy dews were sufficient to induce abundant spor- ulation of the fungus. The observations recorded in Table IV show that even very heavy dews each night were, with few exceptions, insufficient. The period of the experiment was marked, as a whole, by rather cool weather (see Fig. 10) and where heavy dews are recorded it is positive that the plant sur- face was wet from 8 p. m. until 7-8 :30 a. m. Dews alone were not sufficient but tliey, when aided by .9 in. rainfall (Sept. 5), caused abundant sporulation on all the spots exposed. Plant A, protected, showed none or only a few spores on the spots. There- fore, concluding from both experiments, it appears that fre- quent rains aided by heavy dews furnish the essential moisture conditions for optimum spore production of A. solani in nature. VIABILITY and LONGEVITY OF MYCELIUM AND CONIDIA Jones (1896) states that the mycelium in the spot retains its life for a year or more. The writer's results in the main cor- roborate this. Leaves dried between layers of cotton yielded tlie fungus from both small and large spots when isolations were made after 12 and 18 months. Material 29 months old, appar- ently as well preserved, gave no growth of the fungus in several attempts at isolation. There is no evidence of the existence of any differentiated or resistant form of mycelium in the spots. In pure culture, mycelium in prune agar was found viable after seven months. Potato agar plates, tested for viability after 15 and 17 months gave negative results. The recent work of Bartram (1916) shows conclusively the great resistance of the mycelium of this fungus in pure culture to very low temperatures. The condia are also very resistant. Jones (1896) succeeded in germinating conidia one year old but obtained no growth from those two years old. In one instance the writer got 10 per cent germination after 17 months at room temperature. 28 Research Bulletin 42 is ft. (M C S" W i b a, a, as -J 00 v- /: sS G .a □ 3 c 5 o fe5 Z5? ZS -^ to S5S o 4. ro 0)53 1-3 '>^ 2005 -H ^ ij G X +-' y) , Q ra oj ■-i c c IS a Hi *^ ^ -ji o a o teJ5 i»2-2Ss ^l 0 IS i w a> 0 ;i5 a 09 ce OifC :4 ca c4 cd ;- cd og . gjc" "= X a> ag 3 CI in I- en ^ tn as 25 w °a ») 0 0 a w 01 i^ 0 0 z; X 5a O o ^S 2 Hz "O o .30 woo • ••'a 0 . 8. eavy artly .warm ;eo-^ 7. cool Aug. 2 Med. a partly cloudy Aug. 2 Very h dew: p cloudy © S a. & Eahlv Blight of Potato and Related Plants 29 .Spores scat- tered overup- t)er sur- face, none below J3 cS Hi 3 a S P X 2' cd s ■a X O S X 0/ ta i. Very abund- ant s|X)rula- tion both surfaces X og *% X Very abund- ant sporula- tion both surfaces S|)ores al)und- ant in center above, few below C3 — 35 X 9 1 X o is Very abundant sporula- tlon both surfaces No spores Abund- ant sporula- tion both surfaces X Z X Enor- mous spornla- tiOM both surfaces of entire leaflet No spores Spores very abund- ant both surfaces 1 ZS Leaves dying from tip-burn & early bllyht Z ai Few -scat- tered spores X Few scat- tered spores both surfaces X t Z X X 9 Z^ No spores Very few spores scat- tered above No spores X 0) Z X X Z X 0) £ z°| Aug. 31. No dew; clear, windy, cool Sept. 1. Heavy dew; floiidy, cool; mist in A. M. .-SB iil z._-s 1-^ Sopt. 4. No dew; rain (.27) 6 A. M. vines w<>t • until 10 A. M. cloudy, warm Sept. 5. Liirtil (lew; rain (.9) 8 A. iM -5 1*. M. Itlll? 30 Research Bulletin 42 DISSEMINATION OF CONIDIA The suddenness of appearance of a general and severe infec- tion of early blight following a period of favorable weather has been noted by various workers. Observational data accumulated during the summers of 1915 and 1916 seem to indicate that the wind is the chief agent of dissemination in such cases. For in- stance, a field of early potatoes at Waupaca, Wisconsin, was noted to be suffering severely from early blight and tip-burn to the extent that on September 11, 1916, the majority of the vines were dead while an adjacent field of Rurals on the south was green and showed but relatively few spots. However, on a strip of the latter about 80 feet wide, adjacent to the early field, the disease was much more prevalent, but the number of spots was noted to decrease as one proceeded from the boundary line. To determine the relative occurrence of spots, typical leaves were picked from the first two or three rows next to the early field and an equal number 75 feet back. The spots were counted, in- cluding all the leaflets on each compound leaf. 12 leaves of lot 1 each bore 45 to 356 spots, average 175 spots per leaf ; 12 leaves of lot 2 each bore 20 to 141 spots average 71 spots per leaf. Since potato beetles were practically absent from this field and strong north winds with favorable conditions for spore production and infection had occurred the preceding week, all evidence pointed to the wind as responsible for the general dissemination over this adjacent area. There seems to be little doubt that the Colorado beetle is an- other agent of distribution for Alternaria spores. Twice during July, 1916, the examinations of washings from the beetles were made. Fifty adult beetles collected from diseased potato vines were dipped and shaken for a moment in ten cubic centimeters of sterile water from which microscopic examination and poured plates showed abundant spores. Numerous contaminating sap- rophytes i)rcvented the actual number of spores from being de- termined. METHOD OF INFECTION According to Jones (1896) penetration may occur either through the stomates or directly through the cuticle. With proper conditions the young leaves of a plant can be infected as Early Blight oi- Potato and Related Plants 31 readily as the older ones l)ut the i-ate of enlargement of the spot is distinctly slower in the young leaves. Thouf?h infections in nature frequently occur about flea beetle holes, the obsem'ations of several earlier investigators as well as those of the writer indicate no necessary relation between the two. It is not improbable, however, that these little beetles may carry the spores, as is shown for the Colorado beetle and as a result inoculate the wounds they make. PERIOD OF incubation In the greenhouse where the cover slip method was used the incubation period both for potato and tomato, varied from 28 to 50 hours. Under field conditions, relying entirely upon heavy dews for the necessary moisture, incipient spots were usually no- ticeable within 48 to 72 hours after the spores had been atomized upon the plant. Under favorable conditions, within three or four days these spots may enlarge and produce spores which can cause secondary infectron on adjacent leaves or plants. TIME OF natural INFECTION As observed in central Wisconsin, natural infection is gener- ally first visible from June 20 to July 10 on the crop planted April 25 to May 15. On the late crop, spots may be observed from the middle of August on, depending apparently upon three factors: age, vigor of plant, and weather conditions. SOURCE OF NATURAL INFECTION The source of inoculum for the early crop is probably from the overwintered spores and possibly from new conidia produced by overwintered mycelium which has been harbored in the soil in the debris of form.er crops. It is quite likely that an additional source of infection of the late potatoes is from nearby early fields in the form of spores carried by the wind or by potato beetles seeking the younger and more tender plants. OVERWINTERING OF THE FUNGUS The problem of the overwintering of Aliernaria solani is concerned with but two possibilities, i. e., conidia and mycelium. 32 Research Bulletin 42 It lias already been shown that both these structures possess re- markable resistance toward unfavorable conditions. The writer has no evidence to substantiate, and sees no reason for accepting, the hypothesis offered by Massee (1906) and en- dorsed by McAlpine (1911) that the disease is transmitted from one generation to another by latent mycelium in the tubers. To determine definitely under what conditions the fungus can overwinter in Wisconsin, the following experiment was made. On July 22, 1915, some very good material showing abundant sporulation was collected and the leaves dried quickly in the open air. In October, a 6 x 10 foot plot in the plant disease garden at Madison was marked off into four strips and used as follows : In No. 1 — Diseased leaves on the surface In No. 2 — Diseased leaves buried two inches deep In No. 3 — Diseased leaves buried four inches deep In No. 4 — Diseased leaves buried eight inches deep The leaves were protected by being placed between one thick- ness of cheese cloth and this in turn was placed between two lay- ers of galvanized iron wire netting. At intervals throughout the winter material was removed from each strip and attempts were made to isolate the fungus from it. The bulbs of soil thermo- graphs were buried four and eight inches in the plot to furnish a continuous record of the soil temperatures, while an air thermo- graph nearby registered for the air. The records from Novem- ber 12 to April 20 showed a variation in temperature from +13 to -25°C.. for the air, +10 to -(PC. at four inches dcplh. and + 8 to -r/^(\ at eight inches depth. On !):5 out of Ihe total of 160 days for the period the ground was covered with snow. The extremely low temperatures in each case followed periods of snowfall so that it is probable that even the material on the sur- face was not exposed to as low temperatures as were recorded. Before burying the leaves in the fall, viability tests gave over 95 per cent germination of the spores. Sevei'al attempts failed entii'cly to isolate the fungus fi-om the spot tissues where the mycelium api)eared to be dead. This was an unexpected result, which was not fully understood until the following summer. Then it was found that tlio mycoliuiii could frequently be killed by di-ying freshly collected leaves quickly in the sun. Thus un- fortunatelv this test was limited to the conidia alone. Little Early Blight of Potato and Related Plants 33 (liriiculty was I'xpei'ionced in isolating the spores for germination tests during the early winter, but later, as the cheese cloth and leaf tissue disintcgi'ated, the eonidia were more difficult to find. On Docenibor 11, 1915, tests of 40 to 50 spores from each level gave 80 to 90 per cent germination. At no time was there ariv evi- dence of the formation of new spores and cultures from the spot tissue developed only saprophytic invaders as Mucor, Fusarium, Penicillium, and Alternaria fasciculata. On April 17, 1916, the final examinations were made with the following results: (1) Spores overwintered on the surface — 2 3 per cent germination (2) Spores overwintered at 2-inch depth — 40 per cent germination (3) Spores overwintered at 4-inch depth — 50 per cent germination (4) Spores overwintered at 8-inch depth — 65 70 per cent germi- nation The low fi'4U.e for the surface germination would probably have been highei' liad not the location for the i)lot been selected on low ground where excessive water and ice made conditions unusually severe. From this experiment it seems justifiable to conclude that a relatively large proportion of the abundant spores produced during the moist weather of late autumn remain viable through- out the winter. The primary infections of the next year doubt- less come from sucli spores which have overwintered in the soil. It is easy for these to reach the lower leaves which are indeed often in immediate contact with the soil, and it is noteworthy that the primary infections always occur on such low lying leaves. This theory is in further accord with the observed fact that early blight starts earliest and is worst on old garden soils and suggests the conclusion that crop rotation is a factor in its control. THE relation of CLIMATE AND SOIL TO THE DISEASE Climatic factors undoubtedly exert a great influence upon the dissemination and destructiveness of early blight. As to the climatic conditions best favoring an attack of this disease, Jones (1895) finds that liot, dry weather followed by a moist period is best. Rolfs (1898), in Florida, reports that the disease on to- matoes spread with "alarming rapidity" during moist, warm seasons, while dry, cool weather retarded its progress. Lutman 34 Research Bulletin 42 Early Blight of Potato and Related Plants 35 /fOS Vf »JffffOl^ /t/90 J9J 36 Research Bulletin 42 (1911) suniniarizes twenty years' observation (1891-1910) made at the Vermont station mainly on the relation of the weather to late blight but including data on early blight and tip-burn as well. A careful study of the twenty diagrams and notes pre- sented shows so much contradictory evidence on the occurrence of early blight that few conclusions are possible. His statement that it is a disease of the di'ier seasons is fairly well corroborated l)y tiie diagrams. Various writers have called attention to the greater destrue- liveness of this disease on the lighter, sandy soils as compared with the dama'gc it does on the heavy ones. On account of the very generalized nature of our knowledge of this subject, the author has attempted to get evidence which would more clearly show the influence of climatic and soil factors on the severity of the disease. For this purpose continuous meteorological records (air humidity and temperature) were obtained in a standard Weather Bureau shelter at the same height as the potato vines for the seasons of 1915 and 1916. Soil temperatures among the roots and soil moisture determinations were obtained only for the summer of 1916. The light, sandy soil on which the experi- mental plot was located both seasons, proved ideal for such studies on account of the more decisive response of the plants to changes in environmental conditions. Fortunately for this study tlie two summers represented extremes in opposite direc- tions from the normal in regard to the conditions favorable for the disease. The season of 1915 was characterized by much wet, cloudy weather (hiring July and early August and by relatively high temperatures. The remainder of August and the first week of September were dry and clear, and noi-mal in temperature. The disease was first noted July 3 on the early crop planted April 25 to May 10, but did little damage prior to the third week in. July when the plants began to set tubers. From this time on through August it spread with great rapidity and together with tip-burn resulted in an estimated loss of 35 to 50 per cent. A heavy frost on August 27 and a subsequent severe attack of late blight resulted in considerable loss to the late crop, which had shown but little early blight. The season of 1916 began with a very wet. cold June with ex- cessive rainfall making conditions unfavorable for the planting and growth of Ihe crop. The first ten days in July were marked Kaki.v Bi.Kiiir OK Potato and Ri:lati:i) Plants 37 hy mild favorable wcathei- after which a period of di-y weather with extremely hij?h temperature be^an and continued almost un- broken throughout tlie summer until September 5. On twenty days of this period the temperature at the height of the plants i-eached or exceeded 90 ""F., and on fifteen days the thermometer registered 100°F or more. On June 26, spots could be found on occasional lower leaves of most of the early fields examined but the vines were very vigorous and were just beginning to flower. By the time the hot weather began, the second week in July, the disease had made but little headway. The high temperature of air and soil and consequent reduction in the available soil mois- ture now quickly weakened the plants, thus making ideal condi- tions, 80 far as host susceptibility was concerned, for the rapid spread of early blight. By July 30, the vines were mostly dead from tip-burn and but very few early blight spots could be found. The late crop, planted between June 1-15, escaped very large- ly the severe drought. The rainy weather of September and early October, however, enabled early blight to spread so that 30 to 40 per cent of the foliage was badly diseased. Th's injury, in connection with two light frosts, operating on the already much retarded plants, it is believed, was an important factor in reducing the tuber yield. The meteorological records for 1915, being incomplete, are not given. The data summarized in Figure 10* cover, therefore, only the season of 1916. When these are considered in connection Avith the spot history records (Tables III and IV). made during the same period, there is evident a very close correlation between the various environmental factors and the occurrence and de- velopment of early blight. The evidence shows. (1) that in order to have the optimum conditions for an epidemic there must be relatively high temperatures in combination with a more or less weakened condition of the plant so that the fungus can make its greatest spread : (2) that such development will not occur unless the above conditions are prefaced by relatively moist periods of high humidity and abundant dew or rainy weather when spore production and infection can readily take place. The season of 1915 represented just such a correlation of condi- tions for the early crop. In 1916. on the coiitrary. no such opti- mum climatic combination prevailed, so that, although the •The writer is inciebtefi to Prof H V Stewart of the I'^nivevsity of Wis- consin, who determined the moisti're equivalents from which the hygroscopic coefRcients (approximate nonavailable moisture) were calculated. 38 Research Bulletin 42 plants were in a most susceptible condition, there was no general occurrence of early blight until late autumn. These studies suggest a possible explanation of the severity of this disease in some countries and its practical absence in others. While the writer has had no opportunity personally to observe it in other countries it is noteworthy, according to reports in literature, that the organism occurs in practically all important potato growing regions of the w^orld. The difference in destruc- tiveness, therefore, must be due, not to the lack of introduction, but to a difference in climatic conditions. As already noted it is reported more severe in the United States, Australia, New Zea- land, and South Africa than in Europe. Conclusions from stud- ies in Wisconsin seem to indicate the following interpretation : the disease is more destructive in the first countries named be- cause in general the average summer temperatures of these re- gions are not only higher but probably subject to greater fluc- tuations and extremes which, combined with variations in rain- fall, make conditions less favorable for the growth of the plant. In central Europe, on the contrary, where early blight as a serious disease is practically unknown, the moderately low summer temperatures and the uniformally distributed rainfall furnish highly favorable conditions for the host plant, while less favorable for the best development of the parasite. Control Measures resistant varieties Stuart (1914) summarizes the results of five years' observa- tions of the relative resistance to early blight of 153 American and foreign varieties of potatoes. Four of the ten varieties found most resistant to early blight are also found among the ten most resistant to late blight. But one of the ten was of American origin and it was of no commercial importance. The European varieties, though quite resistant, did so poorly under our climatic and soil conditions as to be practically worthless from a commercial standpoint. He concludes: "the value of the disease resistant varieties is problematical rather than ac- tual. The plant breeder, by mating them with the most desir- able commercial types, may develop commercial types of resis- tant varieties." Green and Waid (1906) of the Ohio station. Eaklv Bliciit of Potato and Related Plants 39 howcvoi', believe that niueli can be done in building up resistant varieties by selecting seed from resistant hills. The McCormick variety is said by Norton (1906) to show de- cided resistance to early blight. Prof. T. H. White of the Mary- land station furnished the writer with seed of this variety which was tried out in 1915 and 1916 in Wisconsin. The unusually large coarse vines showed by far the greatest resistance com- pared with the fifteen other varieties grown. However, in late September, 1915, when the stage of greatest vigor had passed, they also showed 20 to 30 per cent of the foliage badly diseased. The poor quality of the tuber will probably prevent it from be- coming of much commercial importance where more desirable varieties can bo profitably grown, SPRAYING The early spraying trials by Jones, aimed particularly at early blight (see Jones and Morse 1905), as well as the long series of potato spraying experiments at the New York and Connecti- cut stations, have shown the practical control of this disease with bordeaux mixture. Lutman (1911), summarizing the twenty years' spraying in Vermont, states that three to four applica- tions of the 5-5-50 bordeaux "efficiently protects the plants from the attacks of the early and of the late blight. ' ' Milward (1909) states that increased yields result from spraying in Wis- consin when not less than four applications are given and the spraying commenced not later than August 15. Stewart (1914) states that in the ten year series at Geneva there was an average increase from spraying for both blights of 97.5 bu. per acre. The 4 1 50 formula is recommended for the first two applications with an increase to 6-4—50 in the late sprayings. On the other hand Clinton (1916) obtained an aver- age increase of 38 bushels per acre in Connecticut with three applications of the 4-4-50. Additional evidence bearing di- rectly on the control of early blight is given by Jack (1913 and 1916) for Rhodesia in South Africa. There early blight appears to be by far the most important disease of the potato. Several years results showed an increase in yield, due to spraying with bordeaux mixture, ranging from 16 to 57 per cent. Wherever this disease causes practical injury on the tomato, spraying Avith bordeaux mixture has also been recommended. iO Research Bulletin 42 Edgerton and Moreland (1913) advise one application in thv eold fi-ame and one every ten days thereafter in the field if the disease is prevalent. The spraying experiments conducted by the writer were de- signed primarily to furnish evidence on control correlated with his life history studies of the fungus, and secondarily to test out under Wisconsin conditions the recommendations of workers in other states. SPRAYING EXPERIMENTS AT WAUPACA The season of 1915 was unusually favorable for the develop- ment of early blight. Spraying was done, however, only on the late crop which, in the experimental plot, was completely killed by frost on August 26 before much differentiation between sprayed and unsprayed was noticed. In 1916 both early and late potatoes were sprayed, but unfor- tunately for the expei'iments on early potatoes, little disease oc- curred this season. In spite of this the results obtained seem worthy of record. On the late crop, planted between June 5 and 15, it operated in weakening the already much retarded vines and was undoubtedly responsible for a large part of the short- age in yield. On early potatoes — Experiments were undertaken in two gardens which had grown several successive crops of potatoes and in which early blight had been noted as severe in 1915. The plots were sprayed by hand with a modified Hudson and Thui'ber compressed air sprayer. This pump proved quite satis- factory for plots of small size and, with high pressure, gave a very fine spray. Gi-eat care was taken to cover all leaves thor- oughly with the mixtui-c. The amount applied eacH time was de- termined by the differences in gross weight of the container be- fore and after spraying. As a rule about 150 gallons per acre were used for each of the first two applications, and 175 to 200 gallons per acre foi* the later sprayings. Since early blight was a negligible factor on account of the extreme drought, the bene- ficial results o])tained are attributable pi-imarily to the lessen- ing of tip-burn and flea beetle injury. However, it is note- wortliy that, whereas a dozen or more spots developed on each control plant in Plots 1 and 3 (Experiment B), only rarely Eakly Blight of Potato and Related Plants 41 could an iiifoction be found on Plot 2, Avhich received weekly applications (9 in all), beginning when the plants were 6 inches high. The results are combined in Table V. TaUF.E v.— SPK.A.YIN(4 E.Xl'KUIMENTS ON EaKLY P()T.\TOKS Rxperlmeat A, Van Patten Garden; Six Weeks Variety Tit'atment \'IELI> Increase [•lot Actual number lbs. Bu. per A. Large Small Total Bu. Per cent 1 Bordeaux 5-.5-l)0 June 16. 24; July 1,8, 15, and 29 42.5 15,0 57.5 87.0 4,5 5.2 2 Control Paris green and lime 38.0 11.0 49.0 82.5 3 Bordeaux 5-5-50 Jul.v 1 and 15 45,5 8.5 54.0 90.8 5.3 5.8 4 Control Paris green and lime 47.5 7.5 55.0 84.4 5 Bordeaux 5-5-50 Julyl, 10, and 29 56.0 11.0 67.0 107.6 19.3 17.9 6 Control Paris green and lime 23.75 3.75 27.5 88.3 7 Bordeaux 5-5-50 July 8. 18, and 28 77.0 12,0 89.0 144.0 38.8 26.9 8 Control Paris green and lime 55.5 10.0 65.5 105.2 9 Bordeaux 5-5-50 .(une 24: July 8 and 22 53.0 12.5 65.5 104.4 17.1 16.3 10 Control Paris green and lime 43.5 12.5 56.0 87.3 11 Bordeaux 2-4-50 June 24: July Sand 22 43.5 17.0 60.5 92.3 -2.3 -2.4 12 Control Paris green and lime 46.5 15.5 62.0 94,6 Experiment B, Taylor Garden: Early Denver Variety 1 Control Paris green and lime 38 27.5 29,5 65,5 70.5 112.2 2 Bordeaux 5-5-50 June 16. 24: July 1, 8, 15, 22. 29: Aug. 5 and 14 41 120.7 17.5 14.5 3 Control Paris green and lime 15,5 12.0 27.5 94.2 On late potatoes — In experiments A and C (Table VI) the spraying was done on selected rows in one tenth acre plots which had been. cropped successively to potatoes for several years. These were sprayed in the same manner as the early potatoes. The other trials were carried out on various farms near Wau- 42 Research Bulletin 42 paca, where the fields had been subjected to a four year rota- tion. Here an upright barrel outfit on a cart was employed. Though much retarded in development the late potatoes es- caped to a large extent the severe drought during July and Au- gust. Revived by the heavy rains in September they made good growth, and, had frost held off until late Octolier, a fair yield could have been obtained. The entire plot in Experiment A was heavily watered with a hose several times during the early part of the season, which fact accounts partly for the greater amount of disease and the consequent greater difference in yield as compared with the other experiments. This plot also received the greater number of sprayings. Prior to September 13, early blight was practically absent in any of the fields except Experi- ment A. The rains and favorable weather following this date permitted rapid spread of the disease on the already weakened plants. Thus during a month of favorable growing weather for the plants a good portion of the leaf area in most cases became badly diseased. Flea beetles and tip-burn were practically ab- sent and no late blight was found. In all these experiments those rows which received two or more applications of the 5-5-50 bordeaux contained in every case larger and more vigorous plants even before any disease occurred. This seemed to be due entirely to the stimulative action~of the spray. The disease was not absolutely controlled in any case, not even plot 1 of Experi- ment A, which received 7 applications. Several light frosts in September complicated the situation in Experiment B where the sprayed plants on this sandy type of soil showed a striking resistance to frost injury. Aside from this, however, the uni- form and consistent increase from the spraying is atti'ibutablc to but two factors, viz., (1) the practical control of early blight and (2) the stimulative action of the bordeaux mixture on the plants. The results are presented in summary form in Tal)le VI. RECOMMENDATIONS FOR SPRAYING While the period during which the foregoing experiments were conducted was not typical of the average year in many re- spects, yet the intensive study made of the disease in connection with them seems to warrant the following deductions: For tlie ear'y crop under Wisconsin conditions the disease can be profitably controlled by four to six applications of the Early Blight of Potato and Related Plants 43 TabiJ'; VI. -r^i'RAYiN'cj Exitciumknts o.v Latk Potatoes Experiment A, Tiirrell field; Rural Ne«v Yorker No. 'i Treatment VlKM. Inckease I'Kil Actual Xurnl)pr ll:)s. 1 1 Hiis. IX- r I A. Large Small Total Bus. Pel- I cent 1 Bordeaux 5-5-50 June 28; July 8, 18.28: Aug. 7. 17; Sept. 6 123.0 10.5 1.33.5 1 331.9 144.2 43.4 1 2 Conirol Paris green and limp 69.5 15.0 75.5 i 187.7 3 Borfleaiix 5-5-50 J-.ine 28: July 12. 26: Aug. 9, 24: Sept. 6 107.0 16.0 123.5 i 301.4 113.7 1 37.7 ! Experiment B, €on!4tance lield; Kiiral New \'orker No. 3 1 Bordeaux 5-5-50 Aug. 12: Sept. 7 188.0 35.0 223.0 111.5 23.0 20.6 2 Control Paris green and lime 149.0 28.0 177.0 88.5 3 Borc'eaux 5 5-50 Aug. l:^ 22: Seut. 7 190 0 31.0 221.0 110.5 22.0 19.9 4 Control Paris green and lime 133.0 27.0 160.0 .80.0 5 B .rdeau x 5-5-50 Aug 12, 22 167 0 35.0 202.0 101.0 21.0 20.8 6 Control Paris green and lime 133.0 26.0 159.0 79.0 7 Bordeaux 5-5-50 Aug. li. 22; Sept. 7 163.0 38.0 201.0 100.5 21.5 21 4 8 Control Paris green and lime 140.0 24.0 164.0 82.0 9 Bordeaux 5-5-50 Aus:. 12: Sr*p[. 7 Bordeaux 5-5-50 Aug. 12, 22; s^ept. 7 149.0 132.0 34,0 183.0 91.5 9.5 10.4 10 31.0 183.0 91.5 9.5 10.4 Experiment C, Taylor field; Rarai Sew Yorker No. 3 1 Bordeaux 5-5-50 83.0 .Jul.v8. 22: Aug. 5, 17: .-^ept. 13 1 18.9 104.9 190.5 74.7 39.2 2 CoQtrol Paris green and lime 46.9 38.5 16.9 18.1 6i.8 58.6 115.8 102.7 18.5 3 Bordeau.v 2-4-50 July 8. 22: Aug. 5, 17: ^^HPt. 13 18.0 4 Control Paris green and lime 28.3 18.1 46.4 84.2 Experiment D, Plnkertou Held; Rural New Yorker No. 3 1 Bordeaux 5-5-50 Aug. 12; Sept. 7 250.5 95.0 345.5 93.9 13.9 13.7 2 C>ntrol Paris green and lime 198.5 98.0 296.5 80.0 3 Bordeaux 5-5-50 Aug. 22: Sept. 7 299.5 81.0 380.5 103.4 27.7 26.7 4 Control Paris green and lime 197.0 66.0 80.5 263.0 •310.0 71.5 1 5 Bordeaux 5-5-50 Aug. 22: Sept. 7 229 5 84.3 8.6 10.2 44 Research Bulletin 42 standard 5-5—5 0 bordeaux mixture. Complete control can: only be attained by weekly sprayings begun when the plants are six to eight inches high and continued through the remain- ing period of growth. For the late crop, the results indicate that the three to four applications ordinarily recommended for the control of late blight will also largely control early blight. Tlioi'oujjhiiess of aj>i>lication cannot be overemphasized in. spraying for early blight. SANITATION From the evidence already presented that primary infection results from spores overwintering in the soil, and from observa- tional data on the persistence of the fungus in dead vines, it is; clear that in certain cases sanitation becomes an important factor to be considered. Crop rotation is of course the rational measure and in those cases where it is desired to crop the land continu- ously to potatoes, all dead vines should be raked together and burned immediately after harvest. Such measures will tend to reduce the number of primary infections but they should be re- garded merely as contributing to the success of the more certain, method of control, viz., spraying. Summary Early blight, Alfcrnaria solani (E. & M.) J. & G., of potato and related plants is a characteristic leaf spot disease distin- guished by the concentric markings or "target-board" appear- ance of the spot. This disease is practically world wide being found wJierevn" the potato is an imi)ortant crop, but it is of economic importance' in but few countries, especially the United States, Australia, New Zealand, and South Africa. The damage from this disease is indirect, i. e., it causes the premature death of the foliage and this results in decreased yields. During some years early blight docs more damage than late blight but it is the annual small loss which makes it a seri- ous obstacle to successful potato culture. On the tomato, where it causes spotting of both leaves and fruit, Edgerton and More- land, 1913, place it next to wilt in importance. Early blight, in Wisconsin, occurs commonly on potato, to- mato and eggplant. The identity of the fungus on these hosts has been established by morphological and cultural studies and by reciprocal cross inoculations from single spore cultures. The^ Early Blight of Potato and Related Plants 45 leaf spot of Jiiiison weed (Datura) which has been widely attri- buted to the same fungus, is shown to be due to a similar but dis- tinct species of Alternaria which was early described by Saccardo as Cercospora crassa. For this the author has given the new com- bination Alternaria crassa. To determine the host range of the fungus, inoculations were made under field conditions on 30 species and varieties of the family Solanaceae. On 29 of these penetration and incipient infection occurred. However, the fungus was able to complete its life cycle on but 12 of the plants, which in addition to two others not included in the tests, make its known host range 14 species and varieties representing the genera Solanum, Lycoper- sicon, Nicandra, and Hyocyamus. The early blight fungus was first described in 18S2 and named Macrosporium solani Ellis and Martin. Jones and Grout, 1896, and Sorauer, 1896, changed the name (the latter on insufficient evidence) to Alfernari solani. Though the writer has never observed conidia in chains in nature and they occur but rarely in culture, the present uncertain taxonomic relation- ship of the two genera, Alternaria and Macrosporium, and the established usage leads him to provisionally retain the latter bi- nonunal, Alteimaria solani (E. & M.) J. and G. The important diagnostic characteristic of the fungus is the long, single or forked, terminal beak of the conidium. On potato and other vegetable and fruit extract agar, the col- ony produces a brilliant yellow pigmentation of the medium, later becoming reddish. After repeated trials to obtain spores in culture, it was found that by stirring or shredding the agar and mycelium in the petri dish and carefully regulating the moisture for 24 hours abundant sporulation could be secui'ed. This served as the source of material for spore germination and inoculation studies. The cardinal tenijioratures for spore germination and mycelial growth on favorable media fall within the folloAving limits : minimum 1-3^, maximum 37-45-, optimum 26-28^C. Five to ten germ tubes emerge at the optimum while at the minimum usually not more than half this number develop. Spore production in nature may begin when the spot has reached a diameter of 3 to 4 millimeters. A given spot may pro- duce 1500 to 3000 spores in two to three successive crops during a season. 46 Research Bulletin 42 The conidia are readily dislodged from their eonidiophores, and local dissemination appears to be chiefly effected by wind and rain. Colorado potato beetles may also spread the disease. Infection may occur via the stomates or directly through the cuticle. The period of incubation varies from 48 to 72 hours. Primary infection may result from overwintered conidia or possibly from new conidia produced by overwintered mycelium. Though conidia were found to overwinter on leaves on the surface of the ground, the proportion surviving the winter was greater on those buried at 2, 4, and 8 inch depths. Early blight ordinarily makes little development until the host has passed its period of greatest vigor and is being weakened by external conditions or by the drain of tuber formation. Opti- mum spore production is dependent upon frequent rains aided by heavy dews. Climate and soil exert a controlling influence upon the development of the disease. In general it becomes most serious when the season begins with abundant moisture which is followed by high temperatures unfavorable to the host plant but with sufficient moisture to insure maximum sporula- tion. Periods of continued drought check its spread completely. The conclusion is, therefore, reached that the optimum condi- tions for an epidemic of early blight require relatively high temperatures alternating with moist periods in combination with a more or less weakened condition of the plant. The unusual resistance of the McCormick potato to early blight, reported by Norton, 1906. has also been observed by the writer, but unfortunately this variety is a poor commercial type. The possibility of securing resistant varieties with the best commercial qualities has been shown by Stuart, 1914, to offer little immediate encouragement, but he is continuing breeding experiments with this in mind. Sanitary measures are recommended based on the evidence as to the ovei-wfntering nnd origin of primary infections. These in- clude crop rotation and the destruction of the dead potato tops in gardens where continuous cropping is practised. Spraying experiments conducted by the writer confirm the results of others and show that timely and thorough spraying with home made bordeaux mixture profitably controls early blight. (See summarized recommendations for spraying, p. 42). Early Bmoht of Potato and Related Plants 4T Literature Cited Bartram. H. E. 1916 Effpf*^ of natural low temperature on certain fungi and bac- teria. U. S. Dept. Agr. Jour. Agr. Res. 3:651-655. Chester, F. D. 1893 Diseases of the round potato and their treatment. Del. Agr. Exp. Sta. Kept. 5(1892) :67-70. Clinton, G. P. 1904 Diseases of plants cultivated in Connecticut. Conn. Agr. Exp. Sta. Kept. 1<)(>3:320, 349, 365. 1916 Potato spraying experiments, third report. Conn. Agr. Exp. Sta. Repf. 1915:470-480. Coons, G. H. 1914 Potato diseases of Michigan. Mich. Agr. Exp. Sta. Special Bui. 66:31. Duggar, B. M. 1909 Fungous diseases of plants, pp. 301-304. Edgerton, C. W. and Moreland. C. C. 1913 Diseases of the tomato in Louisiana. La. Agr. Exp. Sta. Bui. 142:23. Ellis, J. B. and Martin, G. B. 1882 Macrosporitnn solani E. & M. Am. Nat. 16:1003. Farlow, W. G. 1905 Bibliographical index of North American funei. 1, part 1:183-185. Ferraris, T. 1913 I Parassiti Vegetali delle Plante coltivate od utili. pp. 892-8C3 Galloway, B. T. 1891 The new potato disease. Garden and field, Adelaide, Au- stralia 16:158. 1893 The Macrosporium potato disease. Agri. Sci. 7:370-382 and Soc. for Prom. Agr. Sci. Proc. 14:46-58. Green, W. J. and Waid, C. W. 1906 The early and late blight of notatoes and how to control them. Ohio Agr. Exp. Sta. Circ. 58:4. Jack, R. W. 1913 Potato spraying experiments for the control of early blight (AlteDwria solani). Rhodesia Agr. Jour. 10:852-862. 1916 Does it pay to spray potatoes in Rhodesia? Rhodesia Agr. Jour. 13:354-360. Jones. L. R. 1893 The new potato disease or early blight. Vt. Agr. Exp. Sta. Rept. 6(18921:66-70. 1895 Various forms of potato blight. Vt. Agr. Exp. Sta. Bui. 49:91-96. (Distributed 1896.) 1896 Various forms of potato blight and their causes: studies upon Macrosporium solani E. & M. Vt. Agr. Exp. Sta. Rept. 9(1895): 72-88. 1897 Potato rfiseasps and remedies. Vt. Agr. Exp. Sta. Rept. 10: (1896): 44-53. 48 Research Bulletin 42 Jones, L. R. 1903 Diseases of the potato in relation to its development. Mass. Hort. Soc. Trans. 1903:150. 1912 Potato diseases in Wisconsin and their control. Wis. Agr. Exp. Sta. Circ. 36:10. , and Grout, A. J. 1897 Notes on two species of Alternaria. Torr. Bot. Club Bui. 24:254-258. -, and Morse, W. J. 1905 Potato diseases and their remedies. Vt. Agr. Exp. Sta. Kept. 18(1504-05) :272-277. Lutman, B. F. 1911 Twenty years' spraying for potato diseases. Potato diseases and the weather. Vt. Agr. Exp. Sta. Bui. 159:225-296. McAlpine, D. 1903 Early blight of the potato. Dept. Agr. Victoria Jour. 2(1903): 464-467. 1911 Handbook of fungous diseases of the potato in Australia and their treatment. Melbourne Dept. Agr. Victoria, pp. 56-59. McCubbin, W. A. 1916 Tomato black spot or black rot. Canada Exp'l. Farms. Rept. 1915 (vol. 2):987-988. Massee, G. 1906 Perpetuation of potato rot and leaf curl. Roy. Bot. Gard. Kew. Bui. misc. inform. 4:11-112. Milward, J. G. 1909 Directions for spraying potatoes. Wis. Agr. Exp. Sta. Cir, of Information 3. Norton, J. B. S. 1906 Irish potato diseases. Md. Agr. Exp. Sta. Bui. 108:63-72. Niisslin, 0. 1905 Potato leaf curl (Macrosporium solani). Board of Agr. of Great Britain. Jour. 12:476-478. Rands, R. D. 1917 The production of spores of Alternaria solani in pure cul- tures. Phytopath. 7:316-317. 1917 Alternaria on potato and Datura. Phytopath. 7:327-337. Rolfs, P. H. 1898 Diseases of the tomato. Fla. Agr. Exp. Sta. Bui. 47:124-127. Sorauer, P. 1896 Auftreten einer dem amerikanischen "Early blight" ent- sprechenden Kranklieit an den deutschen Kai toffcln. Ztschi. Pflanzenkrank. 6 : 1-9. Stewart, F. C. 1914 Potato spraying experiments at Rush in 1913. N. Y. (Gen. eva) Agr. Exp. Sta. Bui. 379:3-9. Stuart, Wm. 1914 Disease resistance of potatoes. Vt. Agr. Exp. Sta. Bui. 179:147-183. Tubeuf, K. F. von 1904 Die Blattfleckenkrankhelt der Kartoffel (Early blight oder Leaf-spot disease) in Amerika. Naturw. Ztschr. Land- u. Forstw. 2:264-269. Research Bulletin 46 October, 1919 i- Frost Necrosis of Potato Tubers L. R. JONES. M. MILLER aod E. BAILEY AGRICULTURAL EXPERIMENT STATION OF THE UNIVERSITY OF WISCONSIN MADISON CONTENTS Page Introduction 1 Frost necrosis distinguished from other injuries 1 Previous publications 3 Experimental materials and methods 5 Earlier work, 1915-1916 6 Later work, 1918 9 The symptoms of frost necrosis in potato tubers 11 Effect of freezing upon the potato 11 Symptoms of frost necrosis as developed experimentally. . . 11 Types of necrotic lesions , 14 Symptoms of frost necrosis as found in storage 18 Rate of discoloration of frozen tubers 19 Frost necrosis symptoms contrasted with those of other tuber maladies 20 Dry rot 21 Wet rot, soft rot 21 Ring necrosis 21 Brown rot 21 Net necrosis 22 Black heart 22 Internal brown spot 22 The amount and types of frost necrosis which occur at differ- ent temperatures 23 Injury above -3.2 °C 23 Injury at -3° to -4.5°C 27 Injury at -5° to -5.6° C. and at -6° to -8°C 28 Injury at -10.5° to -11.7°C 29 Relation of tuber condition to susceptibility to freezing 30 Relative resistance of mature and immature tubers 30 Influence of relative turgidity of tubers 31 Relation of sugar content 33 Influence of wounds and bruises upon susceptibility 34 Relative susceptibility of sprout and tuber tissues 34 Supercooling and ice crj'stallization associated with frost ne- crosis 36 Relation of time element to supercooling 38 The ultimate freezing point 40 Relative temperatures of air and potato 40 Summary 42 Xiiterature cited 45 Frost Necrosis' of Potato Tubers L. R. JONES. M. MILLER and E. BAILEY The ImIc or niiiiii crop of pot.itot'S as grown and handled in the nortlicrn tier of states is likely to be exposed to freezing temperatures from the last month preceding- digging through all the stages of harvest, transportation, storage, and delivery to the ultimate consumer. The danger of freezing injuries is one of the most serious risks of commercial potato growers and dealers and the problems of the transportation companies are also seriously complicated thereby. In 1917, Avhen freezing temperatures occurred very gener- ally through the northern states before or during potato har- vest, the resultant losses probably constituted a greater toll upon tlie Wisconsin crop than all other disease factors com- bined, .and even in 1918, Avhen the climatic conditions w'ere es- pecially favoi'able, freezing injuries were common and serious. These consisted not only in the innnediate loss of tubers frozen in the field or warehouse, but also in the later appearance in storage of potatoes exhibiting the more obscure freezing injuries. Frost necrosis distinguished from other injuries. It is iin- portant at tlie outset to point out the general characters of frosi necrosis that it may l)e distinguished from other types of injuiy. It is well known that when once frozen solid the po- tato tuber is killed and collapses immediately upon thawing. If, liowever, the exposure to freezing temperatures is moderate or of short duration, it often happens that only a portion of the tubers are thus frozen solid and collapse, the rest remaining un- affected as far as external appearances indicate. If, however, such supei-ficially sound tubers are cut open, various evidences of internal injury will be found in at least some of them. In 1 The term frost necrosis is synonymous with the term freezing necrosis used by Link and Ganlner in an unpublished manuscript. (See footnote 1. page 20.) The writers agree with them tliat frost necrosis is a local or restricted freezing injury which results from exposure to temperature sufficiently low to cause ice formation in the tissues and is thus distinct from chilling injury which results at temperatures not low enough to induce ice formation in the plant tissues. I'tie writers use the term frost necrosis rather than freezing necrosis since frost necrosis has been used in a previous publication (.Tones, L. R. and Bailey, E., Frost necrosis of potato tubers. Phvtopath. 7. 71-72. 1917). 2 Wisconsin Research Bulletin -IG most cases such injuries remain strictly internal and hence, if the potatoes are marketed their defects are not detected until the potatoes reach the ultimate retailer or consumer. The irregular occurence and distribution of tubers which show such internal lesions of frost necrosis makes them difficult to sort out in storage lots. Naturally, i^otatoes frozen during harvest or transportation become mixed with the sound ones, but it is a surprising fact that when storage chambers are sub- jected to the same freezing temperatures and uniform condi- tions of ventilation, certain scattered individual tubers Avill be injured and others not. This individual susceptibility of po- tatoes to freezing injuries, combined with the still more con- fusing fact that frost necrosis is often mistaken for i)athologi- eal conditions arising from other causes, makes it important that there be a further understanding both of the conditions and nature of freezing injury to potato tultei'S. This is especially needed at this time because of two recent coordinated develop- ments involving critical consideration of potato tuber maladies. On the one hand is the movement for the state inspection and certification of potato seed stocks, on the othei' is the develop- ment of the national market inspection service. In l)oth cases it is necessary to differentiate frost necrosis fi-om otlier types of tuber injury or disease, especially the non-parasitic "net- necrosis" and the Fusarium "ring necrosis." Indeed, it A\as because of the evident confusion of frost necrosis with certain of these other types of injury that tlie scnioi- author's atten- tion was .first directed to this probU'iu. Fi'e in the ceHar. One striking exami)h' of typical net necrosis occui'i-ed in a certain lot of selected e.xhihition i)otatoes shown at the meeting of the Wisconsin Potato (Jrowers' Association in 1914. Tile exliibitor was confident that the tnhci-s were noi'nial when he started t'foni home hut they had been subjected to freezing temi)eratures in ti-ansit. Siuiila]- conditions wei'e found in several lots of i)ota1oes in the exhibition of 1!)1S. 'i'he matter presented so niucii ol' ])]'actical as well as scientilic interest that further observations have been supplemented by careful Fhos'i' Necrosis of Pota'io Tibkks 3 oxpciiiiiciits to (Ictcriuinc the effects of various freezing Iciiipci'nt HITS upon ])otatoes. Previous publications. Several previous publications have embodied the results of more or less extensive investigations upon freezing injury to potatoes. The most valuable of these is that of Miiller-Tliurgau (4, 5, 6), who undertook to determine the tempcratui-cs at which plant tissues freeze. His tirst con- cern was with till' ])henomena of svipercooling and the determi- nation of the ultimate freezing point, but in connection with this (5) he investigated the turning sweet of chilled potatoes. Since then, Apelt (1) in Europe, 1907, has approached these questions by somewhat different methods, while in America Appleman (2) published his observations in 1912. In general, where their conclusions have not been in agreement, our own results have confirmed those of Miiller-Thurgau. In none of these earlier publications, however, was critical attention fo- cused upon the intei-nal lesions or symptoms of frost necrosis and it is chiefly here that our own ett'orts have aimed to sup- plement those of previous workers. While the details must be left for later consideration it will be helpful at the outset to summarize the conclusions upon which there is general agreement. Plant tissues, in general, must be cooled to some degree be- low the freezing point of water before ice crystallization begins. With the potato it is the consensus of judgment that there is no killing of tissue or other permanent or injurious effect short of ice crystallization. Where tubers are held at temperatures near or slightly below the freezing point of water, but above the freezing point of the potato tissue, they turn sweet owing to the accumulation of sugar produced by the gradual starch conversion. It is commonly believed by potato handlers and has even been stated in literature by Norton (7, p. 70) that this is due to their having been slightly frozen. ]\Iiiller- Thurgau (5, p. 753), and others since, particularly Apelt (1, pp. 12-27) and Appleman (2. p. 330), have disproved this. By storing potatoes for long periods as Ioav as -1.66°C. (29°F.) Appleman (2, p. .333) determined that sugar accumulated most rapidly at 0°C. or below, and that freezing with potato tubers began between -2.2° and -3.3°C. (26° and 28°F.). Miiller- Thurgau (5, p. 753) stored potatoes at temperatures ranging 4 Wisconsin Research Bulletin 46 from 0° to -3°C. for two weeks and found them still unfrozen after that pei'iod. Our own results as will appear later, confirm their conclusions that there is a considerable range possible in this critically low temperature at wliich tubers may turn sweet before they begin to freeze. Furthermore, none of these men has ever found potatoes to become sweet as a result of freezing consequent upon rapid cooling. Instead they deterndned the rate of sugar accumulation to be veiy slow even under most favorable temperatures. Our own experience is in accord with this in that we have regularly tasted tubers frozen experi- mentally without discovering evidence of increased sugar con- tent in the potatoes which we have subjected to freezing tem- peratures enduring from 2 hours to 2 days. Hence, while sweet- ness indicates that tubers have been held for some time dan- gerously near their freezing point, it does not indicate that they have been frozen, Miiller-Thurgau (4, p. 147) showed that living plant tissues in general require supercooling to some degree below their true freezing point before ice crystallization begins. He found that the freezing point of the expressed sap of a potato tuber was -0.65°C. while the living potato tuber tissues in his ex- periments required supercooling to -3.2° to -6.5°C. before they began to freeze. Apelt's results with the potato, using a less reliable method we believe, are not in full accord with this, but our own trials confirm Miiller-Thurgau 's conclusions that supercooling is the normal course when potato tissues freeze. The earlier workers were led to define rather exact temperature limits for these phenomena with potato, generalizing, perhaps, from work upon a few tubers of uniform type, although they do not agree among themselves upon these limits. On the other hand, our w^ork shows that there is considerable range in variation ])etween individual tubers, even in the same lot of potatoes. The most interesting point and one of considerable pi-actical imi)Oi-tance in relation to symptomatology, is that 1h- apparatus^ which has furnished more accurate data with ability to satisfactorily control the temperatures. The general construction of the freezing cham- ber (fig'. 3) is similar to that described above but special de- vices are added for accur- ately controlling the rate and degree of cooling the freez- ing chamber. This is ac- complished through the in- sertion of an electric lieating coil with a regulating device ■fucli that the temperature can be made to fall at an exactly controlled rate and stoi)ped and held constant at any de- sired point sho]-t of the ex- treme temperature procurable by the ice-salt mixture. Since this latter point is much be- low the temperatures with which we were concerned in our potato freezing trials the apparatus proved highly effi- cient and satisfactory. In most of these trials the appar- atus was so adjusted as to drop the temperature in the experimental cham])er to 0°C. at the end of the first half hour and to lower it SVo de- grees each liour thei'cafter un- til the desired mininuim was reached. For determining the internal tcm]>eratures of freezing potatoes. Miiller-Thurgau's method was employed as descri])ed by him (], p. 168). Two thermometers were used, one of Avhich was sus- })ended in the air of the freezing chamber, the other in a cavity made in the end of a tubei- as shown in figure 4. FIG. 4.— LONGITirDIN.^L SECTION OP TI'BER AS USKD IN SUPERCOOL- ING EXPERIMENTS Thermometer bulb (B) is inserted in cavity (O made m stem end of tuber (T). 10 Wisconsin Research Bulletin 46 In order to prechule any undue pressure or the freezing of sap from the cut surface of the tuber upon the mercury bulb of the thermometer, the thermometer was so suspended that it did not press against the bottom of the pit and the cavity was made about twice as great in diameter as the thermometer and was carefully dried out with filter paper to rid it of free sur- face sap. No doubt the mercury bulb touched the walls of this cavity but the data (Table IX) indicate that the tempera- ture readings were not influenced perceptibly by i)ressure or the freezing of water upon the bulb. While the temperatures obtained in this may not indicate the temperatures of the whole tuber, they do markedly differ from the air temperatures and give some indication of what may be taking place inside the tuber. In the 1918 experiments carefully selected tubers were used chiefly of the Rural New Yorker variety. These had in all cases been harvested and stored without risk of freezing and suffi- cient iiunibers of initreated tubers were cut open to prove them to be generally free from internal lesions. This ena])led us to proceed confidently in their use without i)revious cutting of each experimental tuber since this exposure of freshly cut tis- sue introduces a disturbing factor. The later trials were con- ducted during the latter part of the normal storage period, February-July. In some cases the tubers were kept previous to trial in the wai'in, dry lal)oratory long enough to secure par- tial wilting in order to compare normally turgid with wilted specimens. In the latter part of tlie period ( March-July ) Triumph potatoes were introduced into the ti-ials. Tliese had been previously stoi'cd at .temperatures ai)pi'oacliing 0°C so that there had resulted a considerable sugar accumu- lation. In June and July recently dug. innnature soutlicrn samples of Triumphs were available for comparison with this old stock. Some Early Ohios and Irisli Cobblers were also tested at this time. In previous years trials had been made in- volving different varieties, degrees of turgidity, and stages of maturity. The details regarding these are given later in this article so that it will here suffice to state that in general neither variety, size, relative turgidity nor stage of development nor maturity of the tuber influenced in any marked degree the li- abilily to frost necrosis or the type of resull;mt injury. FkOST NpXROSIS (M' I'o'IATO TlUllKS 11 TnK Symptoms oi- I^'kost Nkckosis in Potato Tubers Effect of freezing- upon the potato. A potato tuber that has been completely frozen will upon thawiiio; be soft and watery and will quiekly collapse or- decay. If the tuber is cut open water drij)s freely from it and even befoi'e cutting the sap freed by freezing oozes through the skin so that the surface is soon wet. This soft, wet condition immediately indicates the trouble to one experienced in handling i)otatoes exposed to frost. Vei-y often potatoes are thus frozen and collapse on one side only (PL, fig. C), owing to one-sided contact with a frosty cellar wall if in storage or to a cold car floor if in tran- sit, or it may occur through partial exposure at or near the surface of the ground before harvest. If such a frozen potato is cut across soon after thawing the cut surface of the interior flesh, although Avatery, is not at first discolored. Upon ex- posure to the air it will, however, very soon pass promptly til rough pink, red, and brown discolorations to a uniform inky blackness. This, according to Bartholomew (3, p. 631), is due to the oxidation of certain elements in the freed sap upon their contact with the air. Evidently the absence of discoloration before the tuber is cut is due to the fact that in the process of freezing and thawing the sap passes from the inteiior of the cells to the intercellular spaces thus driving out the free air and making its reabsorption almost impossible until the tuber is cut. It is often the case in nature that the exposure to freez- ing temperature stops short of the time or degree necessary to the uniform or complete freezing of the tubers. In this case few or none of them may show^ the softening or the wet surface characteristic of the frozen tuber yet, when they are cut open, various types aild patterns of internal discoloration may be found. Since such frost necrosis may bear close resemblance to other types of internal discoloration of the potato tuber, and indeed necrotic lesions of different types may occur in the same lot of potatoes, avc have undertaken to induce frost necrosis by experimental methods in order to determine the vai'ious forms of lesions. Symptoms of frost necrosis as developed experimentally. As a rule, potatoes from the experimental freezing chaml)er which do not immediately show evidences of complete freezing, i. e.,. 12 Wisconsin Research Bulletin 46 become soft and watery, will thereafter develop no external evi- dences of injury even though extensive internal necrosis has resulted. In exceptional cases, however, upon tubers having a clean, smooth, white skin, locally darkened areas may gradu- ally appear where the interior discolored areas lie in the cor- tex close under tlie skin (tig. 7, B). This is not, however, a uniformly reliable symptom and even where detected requires confirmation throuoh cuttino- of the tuber. FIG. 5.— DIAGRAM OF LONGri'UDINAL Sl'-CTION OF A POTATO TUBKR The heavier black portions represent vascular elements, the stippling indicates trans- lucent tissjue of liifrli water content. The vascular ring (r) connects tlie stem <'n) or narrow and Frost Xi:(;k()S1.s of Potato Tiijkhs 1. intensely blackened (fi^'. (J, \>) and is often restricted to the stem end, (3) The blotch constitutes a less well-defined type where the discoloration appears as small ovoidai oi- larger irregular patches ranging fiom an opaque grayish color to sooty black. These occur most connnonly in the vascular ring and cortex although they may be located in the pith (fig. 7, A and B, and PL, fig. A, B, E). •l^jT*. FIG. 7— BLOTCH TYPE OP FROST NECROSIS A — Longitudinal section of a tuber exposed to temperatures ranging from 0° to — i° C. for nine hours. Blotehes Ti.ore abundant in stem end. B — Cross section of the stem end of a necrotic tuber. The intense blotches in the vasrular and cortical regions were evidenced by dark areas on the e.xterior of the tuber. When any considerable number of tubers are subjected to identical freezing conditions it will be found upon cutting them open that diflPerent types of frost necrosis may have resulted so that one cannot with exactness associate these different symp- toms with definite temperature exposures. Numerous observa- tions have, however, shown that some conditions of freezing give a preponderance of certain necrotic types. For example, with Rural New Yorker tubers held at -5°C. for two hours a high percentage of net necrosis resulted (Table 3), the symp- toms becoming more intense Avith prolonged exposure. This 16 Wisconsin Research Bulletin 46 m$f^ T^^i^lff.: \ PIG. fi.— lil.dTCH IVIM: (IF FKOSr M-.(H(IS1S ForM) IX STORAGE ^—erofs wctioa of the ^tvin ciiil oi a tubfr frozen in stonige. B-Longi-Hcction of the rciiuiindiT of iho same tuber. The lesions in this ease ar" eonflne.i to a reUitivelv small portion of the stem-end. Ihe growth cracks in the interior flesh have no relation to freezing injury. Fkost Necrosis of Potato Tibkhs 17 same symptom type oeeiirecl in the Triumph vai-iety as a re- sult of an exposure of -8°C. for less than two hours and prac- tieally never at higher temperatures. The rin<)^ type is but slightly less common than the blotch in tubers of all varieties subjected for long periods to high freezing temperatures. Both occur commonly in potatoes which have been frozen in storage. Less definite blotch discolorations of the opaque type predomi- nate in field frozen specimens (fig. 8), frequently being re- stricted to a suiihuiiicd side of the tuber. With Rural New Yorkers this blotching oceurs with prolonged exposure, 12 hours or more, at -3°f\ Tubers of the Early Ohio variety often in our trials showed a sooty ring, water-soaked and intensely black even when not subjected to extreme exposures. These observations, which are in the main deduced from a series of experiments with well-matured tubei-s during winter storage, are not presented as final evidence that varietal ditferences are constant factors. On the contrary, examination of hundreds of samples of several varieties of potatoes which were accidentally frozen do not indicate any such uniformity. They do show, however, that minor varietal differences appear where freezing conditions are accurately controlled. While we have learned to expect internal darkening of the tissues as a regular symptom of severe frost necrosis, there are mild types in which this may not shoAv much when the tubers are first cut oj)en. In some such cases, even with tubers which had stood for a number of hours after removal from the freez- ing chamber, the only evidence of frost necrosis upon cutting them open was that the injured areas seemed drier and filled Avith air, and they showed a grayish-white tint when first ex posed but within a short time turned red, then brown, mean- Avhile shrivelling somewhat. Although kept for a week or more none of these vascular or other injured tissues turned dark except on the cut surface. We have interpreted this as a mild type of local injury in which after certain cells were killed their freed sap was so absorbed by the adjacent tissues as to hasten their collapse and permit the entry of air into the intercellulars. In addition to the symptoms above described potatoes may begin to freeze on the outside before any internal injury has taken place. This occurs most commonly where potatoes are 18 Wisconsin Research Bulletin 46 touching a freezing sui-faee (PL, fig. C) but also often happens in the Triumphs whieli have a very thin corky la\'er. Rarely it occui's in other varieties and without anj^ apparent cause. Symptoms of frost necrosis as found in storage. The occur- rence of eai'ly autumn frosts in northern AVisconsin in ])oth 1917 and 1918 caught potatoes so frequently that there have been numerous opportunities for observing the resultant effects upon such potatoes during winter storage. In general, these B FIG. 9.— DRIED OUT NKCIiUTlC LKSIUNS Tubers found in storage in March which appeared perfectly sound externally. A— Net type of frost necrosis in which pitting has resulted from drying out. B — Ring type, very opacfue discoloration, also pitted. observations have shown that under good storage conditions and where only internal necrosis occurs the symptoms do not change much. As a result, tubers showing the milder degrees of internal frost necrosis may lie in the storage bin all winter practically indistinguishable from the normal tubers with Avhich they are intermingled. l1 is Inic thai if tlie internal le- sions are very extensive such tubers will tend to wilt or shrivel worse than the normal ones and show internal pitting when cut (fig. 9). Also, Fusarium dry rot attacks them rather more frequently, probably following up tlie dead vascular areas from the stem end tissues. So far as can be judged from general Frost Necrosis of 1'otato Tubkrs 10 observations, such Fusaiium invasion in its earlier stages merely intensifies the injuries, slowly increasing their area and giving the tissues a darker color, but not essentially changing their ty|)e. If this proceeds to the later stages of dry rot the distinguishing symptoms of frost necrosis are soon obliterated. P>hick heart symploins may also complicate those of frost ne- crosis particularly in storage. While it is probable that in many cases these symptoms may have resulted from other fac- tors than those which condition frost necrosis there is some evidence that they may occur as a result of freezing.^ In Feb- ruary, 1919, sonif tubers were found in Rhinelander, Wis., which showed l)oth the net type of frost necrosis and black heart. They had been stored in a w^ell-ventilated room held at temperatures constantly below 60°F., averaging nearer 40'^F., and had been subjected to one sudden freezing tempera- ture when a door had been left open on a very cold day. Rate of discoloration of frozen tubers. Since the lesions of frost necrosis result directly from the oxidation of cells killed during the freezing process, they are not evident in tubers when they are first removed from the freezing chamber but appear only after such tubers have been exposed to warm air for sev- eral hours. Tn order to determine the color changes which occur during tlK» oxidation process and the time necessary for their com- pletion, experimentally frozen tubers were thawed at different temperatures and slices cut from them at short intervals during several days. It was determined that the color cycle, like that described and pictured by Bartholomew (3, p. 631) for black heart, ranges through i)inks, browns, and grays and seems to * The difficulty of learning exactly the causal factors concerned with internal discolorations is well illustrated by recent observations with two lots of seed potatoes. In one case the grower stored his potatoes temporarily in pits in the autumn and found some "wet" tubers indicative of freezing upon trans- ferring later to the winter storage cellar. These were sorted out and the rest of the tubers, some of which were preserved for seed, kept well and be- gan to sprout normally the following May. When cut open during the winter storage period frequent cases of frost necrotic discoloration were detected. Preparatory to planting the tubers were disinfected in May and then left in the open for several days to dry and start new sprouts, being covered with blankets. Upon cutting this seed stock it was found to show much black heart in addition to frost necrosis. The grower suspected frost as responsible for all his injury but E. T. Bartholomew, who examined this with us, diag- nosed the black heart as resulting from heat consequent on exposure to the sun following disinfection. This was confirmed by similar exposure of an- other lot of seed tubers, known to be free of internal discoiorations. Leaving these a few hours exposed to hot .June sun was enough to induce a consider- able amount of black heart. V^Tiile this heat injur>- is less likely to occur at digging time it is nevertheless possible, especially with the early or southern crop. 20 Wisconsin Research Bulletin 46 develop simultaneously thi-oui>hout the injured tissues. The time required for the ultimate dark color to be reached de- pends in part ui)on the air tempcratufc ; thus, at temperatures of 10° to 15°C. from ten to twelve houi-.s were required, while at 25° to 30^C. only five or six hours were necessary. There was no evidence that the rate of ihawiiio- iuHueiu'ed the deo'ree of injui'y nor that tissues which had received severe freezing injuries blackened more rapidly than did those \\ith lesser in- juries. Frost Necrosis Symptoms Contrasted With Those of Other Tuber Maladies^ In freshly frozen tubers frost necrosis m;iy. in general, be easil}^ distinguished from other potato tu])er diseases by the distribution and color of the lesions. Sometimes it may happen that the lesions shown by a single tuber may be so little char- acteristic as to leave one in doubt, hut if several tubers are available, confident judgment is usiuilly possible. If, however, such tubers have lain for some time following the injury, sec- oiulary storage rots may set in and complicate matters. Since the same forms of storage rot may. follow secondarily after various other initial injui'ies the only recourse in such cases is to seek for as clear e\i(hMice as is obtaiiinhh' coiicci'iiiiig tlic character of the initial injui-ies and base final judgment upon this.- It is also helpful in diagnosis of injuries in stored pota- toes to know the r('- the Ignited States Department of Agriculture as a handbook of diseases of vege- tables occurring undei- market, storage, and transit conditions, prepared un- der the direction of W. A. Orton of the Bureau of Plant Industry and W. M. Scott of the Bureau of Markets. Fkos'i' Xi:("i<()Sis oi" INnwro Tii!i;RS 21 one ;ici|ii;iiiil rd with coih lit ions, lliis iiuiy ^ivr iiiipoi-Taiit siij^;- y'l'st ions ;is to the probahlc initial causes. Tlic coinnionest of siidi types of tiiIxT injury initiated hy faelors other than frcoz- ini;- a re as follows : 1. Dry rot. Of those, late hiiuht rot caused l)y /'lii/t'Jijhthord iitfisldus is distinuuished from frost neci'osis by the fact that the initial lesions are strictly superficial, the discoloration rarely i)roceedinu- deeper than the eamhial reht and normal in appearance whereas freezing injuri(>s show as persistent discoloi-ations. 2. Wet rot, soft rot. Following severe freezing injuries to po- tatoes all fully frozen tissues collapse immediatelrupon thawing. Often only part of a tuber is so involved, in which case the re- maining flesh if cut open may show the net or blotch lesions characteristic of frost necrosis. As a rule, however, bacterial wet rot immediately follows as a secondary trouble and pro- ceeds to the destruction of the entire tul)ei-. In case of severe attacks by the bacterial blackleg disease the tubers may show a soft rot either while in the soil or soon after harvest. In most cases, however, such rapid wet rot is a secondary devel- oi)ment following late blight or some other initial injuiy to the tuber, especially in heavy wet soils. 3. Ring- necrosis. Stem-end bundle l)lackening occurs in some degree in many potato tubers, showing as a darkening when the stem end is cut across. This may be very shallow (ixn-haps one-eighth inch or less) in which case it is considered non-pai'asitie in origin, or it may extend well through the length of the tuber, in which case it is usually attributed to Fusarium invasion. The former type should lead to no con- fusion with frost necrosis but the latter may. In general, it may be differentiated by its being more strictly linuted to the vascular elements of the cambial ring without the attendant n,'t necro is ( r blotch lesions of frost necrosis. 4. Brown rot. This name is applied to the bacterial disease, caused bv lidciJJus saJdiKiccdno)!. which mav cause a wet. slimv 22 Wisconsin Research Bulletin 46 rot of the vascular ring-. It is, however, readily distinguish- able, as a rule, by the showing of a typical grayish bacterial exu- date from the vascular elements in the earlier stages, by the wetter condition of the tuber in the later stages, and by its restriction to southern stock, whereas frost necrosis is to be ex- pected in northern stock. 5. Net necrosis. This name has been a|)i)lie(l to a condition whei'e the vascular elements brown more or less throughout the flesh of the tuber even during the developmental stage, i. e., be- fore digging. This is considered non-parasitic and is inherit- able from generation to generation. It seems impossible by ap- pearance alojie to distinguish confidently between this inherit- able net necrosis and the net type of frost necrosis. In prac- tice, however, where one is dealing with any considerable num- ber of examples of necrotic tubers, there will probably l)e little difficulty in correct diagnosis. In the case of frost necrosis only a part of such tubers should show lesions of the net ne- crosis rype, others showing ring and blotch diseolorations. Probably in most eases some significant evidence may be ob- tainable also as to the histoiy of the sample, including lialiility to ex])()sui'e to freezing temperatures. 6. Black heart. The typical black heart lesions, resulting from high temperature storage or asphyxiation thi'ough con- finement with insufficient free oxygen, consist of clearly de- limited intci'Mal diseolorations. In certain eases of frost ne- crosis as already cited (see p. 19) J. P. Bennett has found black heart symj)toms where the history of the tubers seemed to preclude the above types of asphyxiation. In any case, this is not likely to be common or seriously confusing. 7. Internal brown spot. This non-parasitic and non-infec- tious malady is chai'actei'izcd by definite Ijtowu spotting of th<^ intcT'ior flesh of the tuber. It is readily distinguishable by its brown color fi'oni the interiud gi-ayish or purplish black frost blotch necrosis. The distiiu'tion is nuide sui-er by the absence in this l)l•()^\n spot malady of any tendency toward vascular discoloration of the ring or net types so commonly associated with fi-ost necrosis. According to Home's description inter- nal bf(twn s])()t lesions nui\- be delinnted by cork t-ells in which case microscopical examination should assui-e theii- difl'ei-ent i- ation fi'om frost necrosis. Frost Nkckosis oi' I'ota'io TriiiiKs 2'S TiiK Amottnt Axn Types of Frost Nkcrosis Which Occir AT DiFFKHKXT TkMPKRATI 'RKS Uccaiisc lati'i' cxpcriniciits (pj). o^)-'-U)) show tliat llie I'alc oi fall of tcinpci-atui'e is one of tlie factors which seem to influ- ence the amount of injury tubers sustain when chilled, the fol- lowino; data are compiled entirely from the 1918 experiments in which the Potter freezing machine was used. They show a certain uniformity in the types of injury which occur at the same temperatures, but also indicate the striking individual resistance of tubers in many cases. Unless otherwise indicated, tubers of the variety Rural New Yorker were used in these tests, and the air temperature was dropped at the rate of 31/2° C. per hour after the zero point was reached. The per- centage of injury as shown in these tables is not very conclu- sive since only 10 or 15 tubers at most were ex])osed at one time. However, they correspond in general with the data ol)tained in the earlier experiments where larger numbers of potatoes were exposed under uniform conditions out-of-doors and where individual resistance also showed strikingly. Injury above -3.2 °C. Miiller-Thurgau (4, p. 147) held that the critical temperature at Avhich potatoes regularly began to freeze was -3.2°C. and Appleman (2, p. 333) stated that this process began at temperatures ranging from -2.2° to -3.3 °C. In our experiments, therefore, the attempt was made to recon- cile their results. In numerous experiments tubers Avere held at -2°C. for hours (in some cases for 48 hours) and no injury ever resulted. Similarly, temperatures ranging from -2.0° to -2.5°C. were tested and found to be too high to produce injury. Between -2.5° and -3.0°C., however, although frost necrosis did not always occur, it did in perhaps 50 to 75 per cent of the experiments, depending upon the length of the exposure and the individual suscei)tibility of the tubers under trial. The following table gives data as to amounts and predominating types of injury from several of the experiments in which tem- peratures of from -2.5° to -3.2 °C. were used. TVl'KS OF FltOST XKCKOSIS IX MARlvKT I'O TA I OKS A.— STEM e:xd injury Cross section of the stem end showing iircg-iilMr hlotches. The whitisli areas together with the wilted appearance of the .>;urfac tuljrr. wliidi is the stem end. See also figures A and B. Reproduced from hand-colored photographs made under the direction of G. K. K. Link and M. W. Gardnrr of the Bureau of Plant Industry. U. S. Department of Agri- Frost Necrosis of Potato Tubers 27 Tablk I — Tvi'Ks AND A.MOUNTS OK Injuky AT — 2.5° to — 3.2°C. '27.5° TO 2(i.2°F ) Exposure In.iuky Exp. No. Temperature "C. Period Frost necrosis Frozen solid 1 -2.5° fi hrs. 12 lirs. 12 hrs. 18 hrs. 18 hrs. 18 hrs. 24 hrs. none 2 .. -2.5°.. l)loti-h (fuiiil), 20% 3 -2.8° "• ,15% (sooty), 30% .. 4 -2.5° to -3.0° '• 1 5 -2.6° to -3. 2° and ring-. fiO% 6 -3.0° to -3.8° X0% 10% 7 -2.8° to -3. 2° ,100% 80% ' In exiM'rinient No. 4 two tubers which had been ueeled were included. These were frozen solid and the unpft'lcd were not. See further data bearinsr on this. Table HI. Exi). 8. Table I V. Exp. 2, anti later discussion of this point. From Table I it appears that long exposures to critical tem- peratures are conducive to the production of the blotch type of injury, and that they ultimately result in the tubers freezing solid. The fact that this blotch type of discoloration is found commonly in field frozen specimens and that it occurs so regu- larly from i)rolonged exposure at these high temperatures is significant. Occasionally, however, the ring tyix' is produced at these temperatures. Injury at -3= to -4.5 -C. The temperatures below -3°C. were employed in further experiments to determine the time during which such temperatures must l)e maintained in order to in-o- duce frost necrosis and also to furnish material for studying symptom.s of tubers frozen at these lower temperatures. In storage and transportation tubers are often accidentally sub- jected to dangerously low temperatures for short periods. Table II gives the result of several experiments in which these temperatures were employed. 28 Wisconsin Rp:search Bulletin 46 TAi;r,K II Tyi'Ks and Amounts ok Injury at — B.2° to — 4.4°C. (2G.2° to 24°F.) Exposure Injury Em). Tempera I life "C. I'eriod Frost necrosis Frozen solid 1 -3.2 to -3.7 -3.2 to -4.0 —3 () to —3.9 Blotch lO'o '> Blotch and rintr. 20% .... None 3 3 liours 4 -3.7 to -3.9 -S.H to -4.2 -3.5 to -4.2 —4 0 to —4.3 2 lioiirs, 30 mill None 5 Blotch and rinf. •"(!% None (i 12 houis 100% Net and rintr, ()0% Net and rine, '■tn 8 -4.2 to -4.4 2 hours, 3il miii None Injury at -5 to -5.6" and at 6 to -8 C. The results of both the 11)16 and 1918 expei'inieiits show that the highest per- centage of net necrotic discoloration occurs after sliort ex- posures to temperatures of -5° to -5.6°C. These are not exclu- sive of other types but they predominate. 'J\\i!i,K III -Types and Amount ok In.fury at —5° to o.6°C. (2:)^ TO 21.!)°F.) Frost Necrosis of Potato Tubers 29 Tabi.r I\'— Typrs and Amount of In.htky at — 6° to ~H°C. (21.2° TO 17.()°F.) Exp. No. Exposure Injitry Temp. "C. I'criod Frost no(M-()sis Frozen s(jlid 1 -6.0 -6.2 -6.r, -6.8 -7.0 -7.4 -7.8 -8.0 1 hi- 30 mill net 80% none 2 " and rintr. 20% ■• " blotch, 60% " rintr, 40% (2 peeled) 3 4.') ■ none 4 30 '■ 5 1 hi- • " blotch, 70% '• " ■• , 100%.... .. 6 .. 7 ' hrs 100% 8 1 •■ net and blotch. 100% 0 The net type seems almost as prevalent at these lower tem- peratures (-6° to -8°C.) as at the next higher (-5° to -6°C.) but in each case where it is recorded as being present the blotch predominated. In experiment No. 8 the net type occurred in Triumph potatoes while the blotch refers to the condition in the Rurals. Not only do these experiments show that net necrosis de- velops very commonly as a result of short exposures at rather extreme temperatures, but it has been found as the predominat- ing type in eases of freezing injury to storage potatoes where the temperature has been known to drop suddenly. On the other hand, it has rarely been observed in cases of field injury before digging. Injury at -10.5° to -11.7°C. Potatoes freeze solid at tem- peratures below -10°C. if they are exposed for any considerable time. Internal frost necrosis develops promptly in all such tubers with the blotch type predominating over the net. It is also of interest to note that at these extreme temperatures freezing begins more often at the surface and proceeds inward. Thus, in experiments 3, 4, and 5 of Table A' the tubers reported as frozen solid were not entirely frozen but had begun thus to freeze from the surface, and in some the peripheral half- inch was thus killed but the interior was intact. 30 Wisconsin Research Bulletin 46 Table V — Types and Amount of Injury at — 10.5° to — 11.7°C. (13.1° TO 10.9°F.) Exp. Etposure Injury No. Temp. °C. Period Frost necrosis Frozen solid 1 -10. -, Ijlotch and net, 70% •• ••, 70% l)lotch. 40% 2 3 -11.0 -11.2 30 minutes 30% 60% 4 -11.7 -11.7 " . 40% 60% 5 •• . 10% 90 9^ Relation of Tuber Condition to Susceptibility to Freezing Throug'hoiit the coiu'se of these investiii'ations individual Kiiseei)til)i]ity of tiil)ers to freezinf? injury appeared constantly in field and storage as well as in ('X]K'rinientally frozen specimens, and it seemed probable that it mi^ht be exi)lained by some in- ternal condition of the tuber which could be produced experi- mentally if the external factors wei'e conti'olled. Consequently, potatoes at different stages of growth which had l)een subjected to varying storage conditions were exposed to similar freezing temperatures and the results compared critically. Relative resistance of mature and immature tubers. During the seasoj) potatoi's of different stages of maturity were tested for resistance. Three plantings of the Rural New Yorkers were made on June 1, July 13, and August 10, respectively. All were dug on October 3, at which time the tubei's from the first j)lanting wei-e mature, tliose fi'oni the second jibout half- grown, while those fi'oiii the lliird measured from one-half to two-thirds of an inch in diameler. Soon aftei- liai'vest. when these tubers were still turgid and uiniioditied by storage, trials were made in which several tubers from each of these plantings were exposed to the same freezing temperatures but no consistent difference in snscei)tibili1y appeared. To be sure, in some trials a larger iiUMd)er ol' inalure than immatui-e tubers remained normal, but in others tlie immature tubers seemed more resis- tant to freezing temperatures. .\s is common with turgid Rurals the net symptoms predominated in all of tliese tubers. Fros'1' i\i:cH<)Sis oi' I'o'iA'K) Tri'.i:iis 31 Fi^iii-c 10 shows tlii'c'c of these potatoes, one from each plant- iiiji', w hich wei-e exposed together to -6.5°C. for about two hours. Influence of relative turgidity of tubers. It is a natural su(j[)ositi()ii that the relative turgidity of the tuber tissues may intJueiiee their susceptibility to freezing? injury. In some of the eai'liei- trials i);irtly wilted tubers wei'e exposed aloiij;' with lur^id FIG. 10.— INFLUENCE OP MATURITY UPON SUSCEPTIBILITY TO FROST INJURY Sections of three tubers of different stages of maturity which were exposed together to a temperature of — 6..5° C. for two hours. All WTre harvested on October 10: A from seed planted .June 1, B from seed planted July 13, and O from seed planted -August 10. ones, and no consistent differences developed. Such comparisons have been made at various times during three seasons with like results. Owing to the individual variations between tubers it is dif^cult to make as convincing comparisons as might be desired, and it is impracticable to use a divided tuber for such experi- mental purposes because of the possible disturbing effect of cut surfaces upon supercooling. In an effort to establish moisture conditions which were as nearly uniform as possible, in some later experiments turgid 32 Wisconsin Research Bulletin 46 Rurals were carefully paired off as to size aud weight, the pairs numbered as 1 and V, 2 and 2\ etc. Numbers 1, 2, 3, etc., were placed in a damp chamber and 1', 2', 3', etc., in a desic- cator and both stored at a temperature of 10°C. Several pairs of tubers, e. g., 1 and V. 2 and 2^ etc., were removed and ex- posed to freezing temperatures each Aveek for a period of two months and although the tubers used may have gained slightly or lost considei-ably in weight during storage their suscepti- bility to freezing was not consistently altered. Tables VI and VII show the results of two experiments which give an idea of the distribution of iujury in the two lots of potatoes. Table VI — Symptoms of Fkost Nechosis as Shown in Paths of Tubers Which Were Stored Under Different Moisture Conditions for 6 Weeks and Then Exposed Together to — 4° to — 7° C. (24.8° to 19.4° F.) for 2 Hours Tuber Weights In Grams Per Cent Gain (+) OR Loss (-) Frost Injury Oriff- nal 1 After 6 weelis Damp chamber Desic- cator Damp chamber Damp chamber Desic- cator Desiccator 31 55 34 55 34 50 51 29 48 29 44 48 4-9 0 0 0 0 -16 -13 -15 -12 -6 net (faint) l)lotch (sooty) 34 50 51 ' As indicated above each one of a pair of experimental tubers had the same orig- nal weisjrht. A comparison of these results shows tliat loss of turgidity does not consistently alter susceptibility. For example, the desiccated tuber of the second pair lost 13 per cent of its or- iginal weight (55 grams) and was injured upon exposure to freezing temperatures while its turgid mate remained normal but in the fourth and fifth pairs the opposite condition ob- tains. There the desiccated tuber of the fourth pair lost 12 per cent of its original weight (50 grams), that of the fifth l)air 6 per cent of its original weight (51 grams), and yet both Frost Necrosis oi' I'otato Tubers apparently <^aine(i in resistance to freezing, same type appear in Table VII. Results of the Taiu.k VH — Symptoms ok Fkost Nk( iiosis as Shown in Pairs of Tu- uKKs Wiiioir Were Stokim) Under Different Moisture Condi- tions FOK () Weeks and Then Exposed to — 2.5° to — 7.0°C. (27.5° TO ID. 4° F.) FOR 2 Hours Tuber weisrhts, grrams l'(M- ct-nt. grain (+) or loss (-) Frost injury After 6 weeks Damp chamber Desic cator Damj) chamber ()i-ii,'iiial Damp chamber Desic- cator Desiccator 23 23 4.J 28 40 61 ?1 n 9 rintr (faint) 43 28 40 61 41 +2 U) 0 3S 0 56 fl 2 -32 —a —8 rintr (opaiiue). . ■• (fainl).... net blaclv heart net (faint) Here the desiccated tubers of the first and fifth pairs seem less resistant and in the other cases the symptoms differ only slightly in the corresponding pairs. However, where the tuber was very much wilted, as was the desiccated one of the third pair, which lost 32 per cent of its water content, intense symp- toms were produced which resemble black heart. This is an ex- treme form but it occurs not nncommonly, and it is indicative cf the increase of sootincss of neci'otic symptoms with decrease of watoi'. Relation of sugar content. ]\liiller-Tlutrgau (6, p. 493) has shown that the relative sugar content in the tuber may influ- ence its freezing point. For example, by preliminary storing at low temperatures he raised the sugar content from 0.53 jier cent to 2.21 per cent. His trials then showed that the true freezing point with these tubers was lowered from -1.0° C. for those of the normal sugar content, to -1.5°C. for those of the excessive sugar content. It is to be noted, however, that in his trials he secured extremes of variation far beyond those which are ordinarily met with in normal potato tubers and eveii so the influence upon the freezing point w^as not proportionately great. This factor may. however, be influential in determining the relative injury to the different tissue elements in the tuber. 34 Wisconsin Resp:arch Bulletin 46 Influence of wounds and bruises upon susceptibility. The presence or absence of a film of moisture on the exposed sur- face of a wounded or bruised tuber seems to determine the in- fluence of such wounds and bruises upon susceptibility to freez- ing injury. When wounds or bruises are corked or healed over as in the case of connnon scab, dry rot, or mechanical injuries, they have no important influence upon the susceptibility of the tubers. Even freshly cut sni'faces often seem not to cause freezing to take place at hi.uhei' temperatui'es as Miiller-Thur- gau (4, p. 172) predicted. In his experiments with freshly peeled tubers he found that supercooling was in-evented by the presence of the surface film of exuded saj) on such tul)ers. He explained this as being due to the fact that this free sap began to crystallize at the freezing point of sap (about -1.0°C.) and that when the sap tlii-oughout the tuber was chilled to this de- gree the presence of crystals on the outside caused the freez- ing process to extend from the outside inward, without the usual supercooling phenomenon. In our experiments tubers were freshly cut in different ways, some were peeled and some split in half longitudinally, and from others slices were cut, most often from the stem end. It was found that peeled pota- toes usually froze solid at temperatures which produced only minor injuries, if any, in sound tubers. In a few cases, how- ever, typical necrotic symptoms appeared in these peeled tu- bers just as in the case of tul)crs with surfaces only partially ex])()sed. In some cases fi-eezing started on these cut surfaces and pi'ogi-essed inward for two or three millimeters Avhile tlie usual neci'otic sym[)toms a])i)eared in the deeper-lying tul)er tissues. Relative susceptibility of sprout and tuber tissues. Spi-outs have in oui- e.\perinu>nts always pi'oved more resistant 1o fi'eezing injury than the tissue, of lhe tuber fi-om which they arise. As a result, if a spi-outed tuber is exposed to freezing temperatui-es the pai'ent tuber may show considerable internal necrosis and have its spi-onts unaffected (fig. 11). Since this has an im- portant bearing upon the relation of frost necrosis to the value of potato seed sto(dv, numerous trial |)lantings were made, some in sand in the ^i-eenliouse bench and some in the field soil. In certain of Ihese experiments, in ordei- to make closer (-om- parisons, the trial tubers wer-e cut in halves, one half being Frost Necrosis of Potato Tubkks 35 FIG. 11 —EFFECT OF FROST OX VIABILITY OF TUBERS A and B (Upper) — Sections of two tubers which were stored at 25° C. lor three months, chilled at —5° O. for two hours, then returned to the 25° C. temperature. O and D (Lower)— Control tubers held constantly at 25° C. Sprouts had developed on all tubers wh?n A and B were frozen. Freezing produced necrotic symptoms in A and B without apparent injury to the sprouts which, however, continued to grow much less vigorously than did those of the control tubers, as is shown in this photograph taken three months after A and B wre cliilled. The photo- graph also indicates the lack of storage rots and drying out in necrotic tubers, even where stored at such a relatively high temperature. 36 Wisconsin Research Bulletin 46 held as a control, the other chilled after the surface was well dried off. In practically all cases such exposed tubers retained viability even where there was internal necrosis but the sprouts started more slowly, and where the frost necrosis was very ex- tensive the parent tuber rotted before the sprout developed in- dependent roots. As a result, planting frost-necrotic tubers in the field yielded only about 50 per cent of a stand. Those plants which survived, although they started more slowly, made rapid gains later and were ultimately as vigorous and productive as the normal controls. The tubers thus secured from this frost-necrotic seed were in turn all examined for any traces of vascular necrosis, and found to be free. While this was to be expected, it is worthy of note as again emphasizing the distinction between net necrosis induced by freezing injurv' and the hereditary net necrosis from whit-h the symptoms may sometimes be indistinguishable. While, thcrefoi'e, in general, it is inadvisable to plant tubers showing any large amount of frost necrosis, nevertheless slightly necrotic tubers may safely be used if one cuts them and rejects pieces which' show lesions extensive enough to predis- pose to rot.^ Supercooling and Ice Crystallization Associated with Frost Necrosis No attempt has been made in connection with these studies to follow the microscopic phenomena associated with the changes in the potato, but it has been the conclusion of pre- vious investigators that the formation of ice crystals in the sap is antecedent to the death of such plant tissues. So far as our evidence bears upon the matter, it is in accord with this idea. In most cases where frost necrosis resulted it was, indeed, pos- sible to detect ice crystals in the tissues either by their macro- scopic appearance if the tubers were iiinnediately cut open, or by holding the suspected tuber close to one's ear and pressing > Supplementing- IMiiller-Thurgau's 1882 work (5) Wollny (8) attempted to fletermine the inHucnee of pi-olonKcd cold storage upon the viability of tubers. He took normal tubers, divided them into longitudinal halves and stored one set of halves in a cold chamber at 0°C. and the controls at 10° C. After 35 days he planted each set separately and recorded growth throughout the sea- son. The aerial vegetative parts were ciuite vmiform from both kinds of tubers but at harvest time the hills from seed tid)ers which had been stored at 10" C. contained more and largei- tubers than did those from the parent seed tubers which had been stored at 0° C. Frost Necrosis of Pota to Tubers 37 s]iari)ly betweiMi lliuiiil) and tiii^vc, when tlie presence of ice crystals is revealed by a faint crunchinj[? sound. This is, how- ever, but a crude test and its unrelial)ility was shown by the fact that frost necrosis appeared in some cases where ice crys- tals were not so detected. Still more significant is the fact that in otlier cases ice crystals were heai-d when no evident injury resulted. So far as any conclusion was justified, therefore, it is that frost necrosis does not necessarily result from a slight amount of ice crystallization but that this must i)roceed to a certain advaiu-ed stage to produce death of the associated tis- sues. It is a matter of common experience concerning the efifect of freezing upon plant tissues that there are wide variations in susceptibility and vai'ious theories have been developed to ac- count for this. Since our experiments give no new evidence bearing on these we will simply record the facts without attempt- ing to relate them to such theories, attempting t^ relate them to sueh— theories; Another interesting phenomenon having relation to ice crys- tallization is that known as supercooling. On this some evi- dence was secured. It is a familiar fact that any liquid must be cooled to some temperature below its freezing point before crystallization begins. This range of temperature Ijelow the freezing point is supercooling. Following supercooling there is a sharp temporary rise of temperature to the higher degree, this latter constituting the true freezing point of the solution (fig. 12). Since potato sap carries considerable matter in solu- tion its freezing point is lower than that of pure water. Miiller- Thurgau determined it to be about -1.0°C. but in our experi- ments it often more nearly approximated -2.0°C. than -1.0°C. and varied widely with individual tubers (Table IX). Miiller-Thurgau found further that where he made compara- tive determinations of the supercooling points of living plant tissues and of the expressed sap, the living tissues had a lower supercooling point than did the expressed sap. He also found that when the potato was frozen, then thawed, and frozen again, the extreme supercooling was not required for the sec- ond freezing. This lowering of the supercooling point in living tissues he attributed to the resistance of active protoplasm. 38 AViscoxsix Rkskakch Bullp:tix 46 Relation of time element to supercooling. Miiller-Tliiirgau licld tliat the su])eiT()o]iiiu' point varifd directly witli the air tcmpcratufe to wliicli tlu^ tuber was expo.sed ; i. e., was de- pressed witli the fall of air temperature. He justi.ties this con- clusion by sucli data as are given in Table Vlll. Table VIII • — Muller-Thurgau's Resi^tlts Showing Relatiox op SrPERCooLiXG Point to Air Temperatures Exposure Potato Temperatures Xo. Temperature Time Supercooling point Freezing point 1 2 3 4 5 - 4 ,5 2 houi's not tri veil 4 houi's 5 '■ 3.2 -0.8 -5.0 - 7.2 -11.0 - 9 to -12 -3 5 -1.2 4.1 -1.4 -1.0 -6.1 -0.98 From our experience it requires some further expUination than is afforded by Miiller-Thurgau 's figures to understand why freezing should have occurred at tlie end of 2 liours at -4.5"C. and at the end of 4 or .l hours at the exti-emely low temperatures of experiments 4 and ."), Table VIIT. Miiller- Thurgau, in liis experiments, already explained, liad no way of regulating the rate of fall of the air temperatui-e in his freez- ing cliaml)er. Fortunately, with the Pottei- freezing apparatus we were a])h' to do tliis. We therefore nnch'i'took to repeat Miiller-Thuriiau"s experiment controllinu' this time factor. The results, as shown in Table IX, indicate that the rate of fall of the air tem])eratnre influences the supercooling point. Frost Nix^kosis of J'otato Tubers 39 Tablk IX— Ki;[-A'I'ion ok Supekcooi-ing to Ratk ok FAr,r. of Fkkkzimg TkMI'KRATUKES N'aiicly Air Temperature Potato Temperature Exu. No. Max. temp. Time to drop from o" to ma.x. temp. .Super- cfKjliriK' ix>int Time to sujH-r- cool Free/iiiK- point °C. 1 2 3 4 Rural -5.5 -5.0 -10.5 -11.0 95 mill 90 " 80 '■ 40 '• -4.0 -4.9.-, -4.2 -3.1 105 mill 174 '• 73 • 49 '• -1.25 -1.3 -1.8 -1.7 5 6 7 Irish Colibifi- -5.6 -6.0 -11. 0 20 •• 60 '■ 55 '■ -3.5 -5.5 -3.2 100 •■ 125 '• 58 " -1.7 -1.7 —2.3 8 9 10 Early Ohio... -4.4 -8.0 -11.0 120 ■• 40 •• 45 " -4.15 2.2 -2.8 112 " 75 '■ 55 " -1.9 -1.6 -1.5 From these data it seems evident that the supercooling point does not vary simply with the air temperature but that it is influenced by other factors, including- the rate of fall of the temperature. Comparing tubers of the same variety, experi- ments 'S and 4 show that in :} a slow drop to -10. 5C. gave a lower supercooling point (— 4.2°C.) than did a rapid drop in 4 to practically the same point. With another variety, in experi- ment 6, a slower drop to -6°C. gave a lower supercooling point (-5.5°C.) than did the rapid drop to -11° C. in experiment 7, (supercooling point -3.2°C.), It will be noticed that in gen- eral the supercooling points recorded in our trials (Table IX) represent about the same range as Miiller-Thurgau's (Table VIII). These are also in accord with our general experience; viz., that potatoes do not begin freezing until exposed to -3°C. or lower. It will be noted, however, that in two cases, experi- ments 9 and 10, the su])ercooling point was reached above -3°C. These are to be regarded as exceptional cases requiring ex- planation. In the first place, in this method (see fig. 4) muti- lated tubers are used and where freshly cut surfaces are ex- posed, even with precautions to dry them, the supercooling point may be raised. In the second place, the supercooling point may be influenced by such external factors as mechanical disturbance, as was indicated in some of our experiments. 40 Wisconsin Kesearch Bulletin 46 The ultimate freezing point. The ultimate freezing- temper- atures as shown in Tal)l«' IX are in general somewhat lower than Miiller-Thurgau's, Table VIII. In hoth eases it will be noted that there is a considerable variation. It will be evident that the method employed can give only approximate results at the best, and also that tliis varies with individual tubers. Relative temperatures of air and potato. Tables X and XT show in detail the comparative tciiijx'i'atures of air and the in- terior of the potato tuber and the supercooling range as fol- +10 -hQ +6 % l^ i-4 ,^ •*^ .^ i-2 ^ 5 0 ^ .1^ -P. ^^ s)> -4 v> .^ •K ^ -6 *4) ^ If -<3 ft -m -/z -/4 V \ \ \ \ \ \ \ \ \ \ > \ \ \ \ \ \ \ V \\ K N \ \ \ \ ^ \ \ \ / \ \ \ \ \ ^»J ^ \ S.^ X \ \ "^-- 10 zo 30 40 50 60 70 60 90 100 7/me /n Minutes FIG 12— GRAPH RK Pl{KSF.NnNG THE RET-ATIVE TEMPERATURES OF AIR AND TUBER IN SUPERC00LIN(i EXPERI.MENTS Tlie iiDDer curves roprosent the temperatures of the interior of the tubers and the lower represent corresponding air temperatures. The dotted lines indicate the tonpera- turcs in experiment No. 4 (Tabic 0) while the contmuous hues belong to p.\penment A (Table 9). A compari.^on of these curves shows that where the air temperature dropped rapidly as in experiment 4. the supercooling point of the tuber occurred more ciuickly and at a higlier temperature than where the air temperature was dropped .slowly, i^" periincnt 3. S«e further evidence of this in Tables 10 and U and accompanying text. Frost Necrosis or I'otato TrBi;Ks 41 luweil tliruui-li two exia'riiueiits, in one of which the toinpcra- ture fall was more gradual than the other. In l)nth cases the internal temperatures could not be accurately i-ecorded in the cai-lier stages owing to the fact that the thermometers were madiiated only for lower temperatures. These data are, how- ever, unimportant. The apparent influence of the rate of fall of air temperature upon the supercooling range is shown graphically in figure 12. Due to the sudden rise of temperature in the interior of the potato just following the supercooling period, all curves which represent the internal temperature of potato tubers have a pi-ofile similar to that i-e])resentcd in this graph (fig. 12). Taui-e X — The Internal Temperature Variations of a Potato When the Air Temperature Is Dropped Slowly to — 5°C (Table IX, Exp. 2) Temperature Time Air °C. Potato °C. 0 +10 0 -1.0 -1.8 -2.4 -2.7 -2.7 -4.6 -5.0 -5.4 -4.6 -4.8 -4.8 -4.7 -4.9 -4.6 -4.8 -4.7 -4.7 -5.0 -4.7 -5.0 10 mill 20 ■■ 40 " 50 " PO " 70 ■■ +2.0 80 •• +0.3 90 ■■ -0.9 100 ■■ —2.0 110 •■ '. -2.7 11.-, •■ -3.2 120 "' —3.3 1?.-, •• —3.5 130 ■■ —3.8 13.-> '• —3.9 140 •• —4.1 ur, ■■ —4.2 ino •• —4.4 15.j ■ —4.5 160 '• -4.7 165 '■ -4.7 170 '■ -4.85 171 '■ —4.9 172 ■ —4.91 17.S ■ -4.92 174 • —4.95 175 '• -2.9 180 •• -1.8 185 ■• —1.5 190 • -1.4 19-. ■ -1.3 200 •■ 1.3 205 ■■ -1.3 42 Wisconsin Research Bulletin 46 Tablk XI — ^TiiE Internal Tevii'eratukk Variations of a Potato When the Air Tempekatukr Is Dropped Slowly to — 10.5 C. (Faulk IX, Exp. 3. The Same Data are Graphed in Fig. 12.) Temperature Time Air Potato "C. (1 mill + 1.0 +0.0 -1.0 —2.0 -2.8 -3.3 -4.0 -4.5 -5.2 -5.5 -5.8 -fi.O —7.8 —8.2 —8.7 -S.8 -9.0 -J.l -9.2 -9.3 -9.5 -9.7 -10.0 -10.0 -10.2 —10.5 10 •• 15 •• 20 " 25 •• 30 '• 35 •■ 40 " 45 •' +2.0 50 •' +0.2 55 " —0.3 60 " —1.4 65 '• —2.9 70 •• —3.2 71 " —3.4 72 " —3.8 73 '• —4.2 74 •• —4.1 75 " —2.9 76 •• —2.5 77 " —2.4 78 ■' —2.35 79 '■ —2,3 80 " —2.1 85 ■■ —1.8 90 •• 100 ■• Summary 1. The potato ci'o]) suffers a considerable daiiiaye eacli year l)ecauso of ±'i"eeziiil- lander 70.8 63.2 Commercial Hollander 7.6 Broesch Plot Average 4 strains of Wisconsin grown Wisconsin Hollander Western grown Wisconsin Hol- lander 75 4 61.6 100.0 11.9 16. U 98.0 85.3 80.0 1.0 49.6 51.2 Commercial Hollander 1.0 In 1919 the disea.se was severe and any sign of yellows on the plants was recorded. Thus a comparatively high percentage of disease is shown in column 1 even for the resistant strains. This condition usually occurs in a warm season like 1919, but, as previously noted, most of the resistant plants are scarcely checked by this slight attack while a large percentage of the commercial strain, when infected, dies before the end of the season. The percentage of plants killed by yellows and the per- centage heading are therefore the best criteria for comparing the various .strains. From both the 1918 and 1919 figures it is evident that all of these strains of Wisconsin Hollander gave fairly good rcsuUs as to Fusarium resistance. The 1918 results are the more sig- nificant and they show (juite clearly that under the conditions of that trial the western grown seed was not quite so resistant as that grown in Wisconsin. Even in 1918, however, the west- ern grown seed made a satisfactory showing and in the 1919 trial it proved practically equal to the average run of Wiscon- sin Hollander. It is to be remembered that in both sea.sons the trial was made on ' ' sicker ' ' soil than will commonly be used for commercial cabbage culture and therefore that the differences are more pronounced than would be evident in general field usage. Considering, therefore, the commercial advantages of growing contract seed in the intensive seed-growing districts wo FusARiuM Resistant Cabbage 31 are approving this method with certain reservations aiming to reduce the dangers inevitably inherent in the procedure. The first of these dangers results from the fact that it seems inevitable that there is in all these resistant cabbage strains a tendency to progressive reversion with a consequent loss in dis- ease resistance which can only be met by continued selection from plants grown on ' ' sick ' ' soil. The commercial seedsman who ignorantly or for other reasons neglects to recognize this prin- ciple may therefore fail to keep his strain up to standard. There is also always the possibility of seed admixture and of cross pollination from adjacent seed fields, both of which will require greater attention from the seedsmen and contract grow- ers with such a strain as this than with the less specialized types. To meet the situation with the Wisconsin Hollander we have continued to urge Wisconsin cabbage growers who have espe- cially "sick" soil, and who have already learned how to select heads from their own fields for seed growing, to continue this practice. This will insure them at least enough seed for their own use and in certain cases the}' will have a surplus to sell to their neighbors or to seedsmen. Certain commercial seedsmen are already arranging to secure their ' ' mother seed ' ' of resistant strains from heads carefully selected from "sick" soil with re- gard both to disease resistance and type. We shall continue to cooperate with both local growers and seedsmen in the establish- ment of these practices on a sound basis. With the kraut varieties it is more difficult for the ordinary Wisconsin grower to succeed in seed growing. One reason for this is that the earliness of maturity causes a much greater loss of heads during winter storage. The chief initial requests for this seed moreover have come not from the growei-s directly but from the kraut packers who, in general, purchase and distribute to the farmers the seed from which their cabbage is to be grown under contract. Most of the kraut manufacturers of the coun- try are members of the National Kraut Packers' Association. Accordingly an arrangement has been made with this Associa- tion by which this Experiment Station and the Federal Bureau of Plant Industry have cooperated for the growing of resistant kraut seed. In this way a considerable quantity of Wisconsin All Seasons and some Wisconsin Brunswick will be available for distribution in 1921. Efforts will be made so to place this as to insure the use of as much of it as possible on "cabbage sick" 32 Wisconsin Research Bulletin 48 soil and so to provide as to insure the production of an adequate crop of seed annually hereafter. It is believed that through the state and national institutions proceeding thus in cooperation with the Wisconsin cabbage growers' committee, with the National Kraut Packers' Associa- tion, and with such of the seed firms as are undertaking to handle the resistant seed, it will be possible to place the produc- tion and distribution of this seed upon a permanently reliable commercial basis. Evidence has already come to hand, how- ever, that along with this legitimate trade development there will be some confusion through the offering of so-called disease- resistant seed of unknown origin by ignorant or unreliable deal- ers. Probably this is not a matter which need mislead any in- telligent cabbage seed dealer or grower. In any case, it will be greatly minimized if all reliable dealers offering these Wisconsin strains of resistant seed will use the names herein given to them and will so state the source of their seed supply as to make clear the essential facts as to its origin or history. SUMMARY AND CONCLUSIONS 1. The disease known as cabbage yellows, caused by the soil parasite Fusarium conglutinans, is widely distributed and seri- ously destructive in the United States. 2. Once introduced, it persists indefinitely in the soil and there is no known method of control except through the use of disease-resistant strains. 3. It has been found that of the commercial varieties the Volga is the most highly resistant and the Houser is somewhat resistant, but neither of these varieties meets important com- mercial needs. 4. The chief commercial cabbage industry in the sections where the yellows disease occurs is concerned with growing either a winter storage or shipping crop or a mid-season or au- ' tumn crop for kraut manufacture. To a lesser degree there is need for truck types. 5. Experience justifies the belief that these several needs can all be met by the selection of Fusarium-resistant strains from the standard commercial varieties now in use which are best adapted to these various purposes. 6. In undertaking such selection our first success was attained with the standard winter storage variety, Hollander or Danish FusARiuM Resistant Cabbage 33 Ball Head. From this was developed the resistant strain known as Wisconsin Hollander. Since experience showed that an ear- lier strain of this was needed, further selection was made and a resistant strain secured which combines with earlier maturity a rounder head and shorter stem. This has been distributed un- der the name Early Wisconsin Hollander, and for purposes of distinction the original resistant strain is now being called Late Wisconsin Hollander. 7. In order to meet the needs of the kraut industry, resistant strains have been selected from two of the leading commercial kraut varieties, Brunswick and All Seasons, and these have been distributed under the names Wisconsin Brunswick and Wisconsin All Seasons. 8. Other Fusarium-resistant selections are receiving attention as follows : Professors White and Close of the Maryland Exper- iment Station have secured and distributed a resistant strain of the Late Flat Dutch; Professors Melhus and Gilman of the Iowa Experiment Station are developing a resistant Copenhagen Market. In Wisconsin the Experiment Station, in cooperation with the Bureau of Plant Industry of the U. S. Department of Agriculture, is working with resistant selections of All Head 'Early, Glory of Enkhuizen, and Copenhagen Market. 9. By following the proper methods any skillful cabbage grower who has Fusarium-sick soil may either undertake with reasonable confidence to develop a resistant strain of his own, or having secured one of these resistant strains he can maintain its resistance and produce his own seed. 10. It is, however, important to note that the Fusarium dis- ease or yellows is often confused by growers with the bacterial black rot (Bacterium campestre), and that these selected strains have not proved to be especially resistant to this nor to the other common cabbage diseases such as black leg (Phoma) and club root ( Plasmodiophora ) . 11. In all cases the degree of resistance to Fusarium shown by these strains is relative, not absolute. The seedling plants are less highly resistant than they are after the transplanting stage. 12. Environmental factors, especially soil temperature, influ- ence the development of the disease and also the disease resist- ance of the host. High soil temperature favors the disease and low temperature inhibits it. It does not develop even in the 34 Wisconsin Research Bulletin 48 non-resistant strains at a temperature below about 17° C. (62°F.) and at high soil temperatures even the most resistant strains show a considerable percentage of infection, 13. In accordance with the temperature relations noted above, the best results are obtained under Wisconsin climatic condi- tions by starting even the resistant strains in a non-infested seed bed to avoid possible seedling infection. These strains are then sufficiently resistant following transplantation to mature a com- mercially successful crop even on badly diseased soil. 14. These resistant strains have proved resistant so far as tested in other states. It seems probable that the only limita- tion in this respect which might occur would be in cases where they were subjected to more trying conditions as to soil tempera- ture, especially in the seedling stage. 15. Should such conditions be met, our experience gives us confidence that through further selection resistant strains suited to any localized conditions could be secured. It is our belief, therefore, that the cabbage industry can be permanently main- tained in any section of the country, in so far as the Fusarium or yellows disease is a limiting factor, through the selection of disease-resistant strains. 16. It seems pix)bable that in case the resistant strains are propagated through successive generations without repeated se- lection, they will tend to lose to some extent the disease-resistant character. 17. When, therefore, it seems desirable for commercial pur- poses to grow the seed crop under contract in non-infested re- gions, it is urgently recommended that the mother seed for each such contract crop be secured from plants carefully selected for resistance and type from Fusarium infested fields. By this method it is believed that the present standards may be essen- tially maintained and seed successfully produced on any desired scale, by the commercial contract method. 18. Work on the disease-resistant cabbage strains will be con- tinued by this Experiment Station in cooperation with the Bureau of Plant Industry of the U. S. Department of Agricul- ture and with certain other state experiment stations. While it will not be practicable for these institutions to grow or dis- tribute seed other than for trial purposes, they will advise or cooperate with growei's or seed finns in securing an adequate supply of resistant mother seed. Research Bulletin 53 July, 1922 The Influence of Soil Temperature on Potato Scab L. R. JONES, H. H. McKINNEY AND H. FELLOWS AGRICULTURAL EXPERIMENT STATION OF THE UNIVERSITY OF WISCONSIN MADISON CONTENTS Page Introduction 1 Experimental work in the greenhouse 2 Methods 2 Soil 2 Seed 2 Pathogen • 3 Planting 4 Manipulation of experiments 4 Methods of determining amount of scab 5 Experiment I 6 Experiment II 8 Experiment III • 9 Experiment IV 10 Experiment V 12 Results 14 Experimental work in the field 16 Temperature apparatus 17 Results 18 General observations 21 Discussion 23 Influence of soil temperature on certain organs of the potato plant 27 Underground parts 28 Above ground parts 31 Summary 33 Literature cited 34 The Influence of Soil Temperature on Potato Scab L. R. Jones, H. H. McKinney and H. Fellows THE COMMON SCAB of the potato caused by Actinomy- ces scabies (Thaxter) Giissow is probably the most gener- ally serious potato disease of America and with continued potato culture on the same soil the disease seems to increase stead- ily. Some years ago the senior author (3) had an opportunity to contrast this condition with the situation in northern Europe where potato scab is generally a minor disease, in many sections practically negligible, in spite of the highly intensive culture of this crop and the abundant use of stable manure from animals fed on cull potatoes. This difference in the prevalence of the common scab disease in the established potato districts of Europe and America led to the conclusion that the explanation must be in the difference in environmental conditions rather than the accident of introduction of the parasite. After making full allowance for other variable factors, such as soil reaction and moisture, it seemed as though temperature variations must be an important factor in the devel- opment of the disease. Other observations have tended to strengthen this idea, some of which have been listed by the senior writers in a preliminary note (5). The causal organism belongs to a fairly high temperature group, as first shown by Shapovalov (9), who found that it thrives best in pure culture at tempera- tures ranging from 25-30° C. In order to determine the influence of soil temperature on the development of scab, an effort has been made to control soil tem- peratures experimentally and grow potatoes in scab infested soil held at various temperatures over a reasonably wide range. Trials have been undertaken in both field and greenhouse, but thus far the best progress has been made in the greenhouse, be- cause only here have we been able to control conditions satisfac- torily. While the experimental work must be continued upon all aspects of the problem, sufficient has been learned to justify a re- port of progress. 2 Wisconsin Research Bulletin 53 EXPERIMENTAL WORK IN THE GREENHOUSE Methods P^ive experiments have been conducted using the Wisconsin temperature tanks (4) for controUing soil temperatures. All plants have been grown in round, galvanized iron pots 6 inches in diameter and 9j/2 inches deep. These pots were surrounded by the water of the temperature tanks, all of which were located in one greenhouse where the air temperature was kept within as limited a range as possible during any one experiment. Soil tem- peratures were maintained by heating or cooling the water in the tanks by means of electric heaters or steam and cold water or ice. Soil A rather fertile loam soil has been used in all of the experi- ments. This was always sterilized by live steam under about 1 pound pressure for four hours. The hydrogen-ion concentration of this soil after sterilization was found to have a Pji value of 7. The moisture content of the soil after inoculation and planting ranged from 18 per cent to 20 per cent (based on weight of water- free soil) in the various experiments. Both of these conditions have proved favorable for the development of the disease. Seed A number of varieties of potatoes were tried out in connection with the earlier series and the Irish Cobbler variety was found especially satisfactory. This variety is highly susceptible to com- mon scab ; it has a smooth white skin upon which scab lesions show distinctly and it develops a large number of tubers in a shorter time after planting than any of the other varieties tested, Owing to the necessity for potato seed to pass through a period of dormancy before sprouting, it has not been practicable to use northern seed in experiments started in the autumn. For this reason, early grown southern seed was used in all but one experi- ment. From the outset of the work it was planned to use Cob- bler seed from the same source (Warsaw, N. C.) in all of the work and this was done in the first three experiments. In the fourth experiment, since it was not possible to obtain this Caro- lina seed, use was made of some of the same variety from Madi- son County, Illinois. In the case of the fifth experiment, which IXFF-UKXCF. OF Soil, 'l"i:.M I'KHATUUK ().\ I'oTATf) ScAl! 3 was not started utitil the spring period of the year, Wisconsin seed (Cobbler) was used. All seed was treated at least two hours in a 1-1000 solution of mercuric chloride previous to planting in order that the work might not be complicated by Rhizoctonia and other tuber-borne diseases. Pathogen In all of the experiments reported in this paper, the same strain of Actinomyces scabies was used. This organism was isolated from a scabby potato from Door County, Wisconsin. It has been increased on various types of media, but it has always been car- ried in stock culture on potato glucose hard agar. While two methods of inoculation have been used, only one has been found entirely successful. In one case the tubers were grown in sterilized soil and the inoculum, consisting of a water suspension of the spores and mycelium of A. scabies, was applied to the injured and luiinjured surfaces of the uncovered tubers. In later experiments, however, the organism was increased on leaf mold and fme cut straw and added to the sterilized soil at the time of planting, and finally it was found that very satisfactory results could be obtained by merely adding the organism in a water suspension to the sterilized soil before planting the seed. The latter method consisted in increasing the organism on the surface of potato glucose agar.' After the organism had sporu- lated abundantly, the surface grow^th was removed and well macerated. This was put in water and the resulting suspension of the organism was sprinkled on the sterilized soil. A sufftcient amount of water was used to bring the sterilized soil up to the required moisture content. Two methods have been employed in increasing the organism on agar. The first consisted in introducing the potato glucose agar medium, containing 3 per cent agar, into Erlenmeyer flasks to a depth of ^ of an inch. The flasks were then plugged and sterilized for 20 minutes under 8 to 10 pounds steam pressure. Inoculations were made before the condensation water, which collects on the inside wall of the flasks during the sterilization process, was absorbed by the solid agar. The condensation water made it possible to prepare a suspension of the organsim w-ithin the flask. This suspension was then well distributed over the sur- face of the agar by revolving the flask at the proper angle to en- 4 Wisconsin Research Bulletin 53 able the suspension to pass over the whole surface of the medium, thus giving a uniform growth. The second method consisted in pouring the glucose agar into Petri dishes. The agar was then inoculated by spraying the agar surface with a water suspension of the spores of A. scabies. This method can be used successfully when a culture room or other suitable place is available in order to reduce contamination. Soil was always shoveled and screened from five to six times after the addition of the organism in order that uniform distribu- tion might be obtained. Soil for the control plants was always handled the same as that used in the inoculated series except that the organism was not added. Planting The seed tubers used were of medium size, ranging from 2 to 3 inches in diameter. Each tuber was cut once through the stem and eye ends and one piece was planted in each pot. The seed piece was placed 6 inches below the surface of the soil with the cut surface down. Manipulation of Experiments Soil moistures were adjusted by weighing the pots at frequent intervals throughout the experiments and replacing the water lost. At the high temperatures this was done daily or oftener if neces- sary, while at the lower temperatures it was done less frequently, depending upon the conditions. In practice it has been found that the temperature of that region of the soil where the greatest number of tubers develop remains practically the same as that of the surrounding water. In the case of the first three experiments, temperatures were adjusted three times during 24 hours. The water in the tanks which were operated above 18° C. (about room temperature) was raised one degree above the scheduled temperature at the time of adjust- ment, while in the case of those which were run at temperatures below 18° C. the water temperature was lowered one degree be- low the scheduled temperatures. This method allowed for a drop of one degree below or a rise of one degree above the scheduled temperature during the eight-hour period. Influence of vSoil Temperature on Potato Scab 5 In experiments III, IV and V all tanks operated at tempera- tures above 15° C. were equipped with electric heaters and thermostats, which made it possible to reduce temperature varia- tions very materially. The average temperature variation in any of the tanks was less than one degree centigrade either up or down from the tempera- ture at which the soil was intended to be held. Methods of Determining Amount of Scab Two methods have been used in connection with determining the amount of scab developing at the various temperatures. One consisted in determining the relative number of tubers scabbed and expressing the factor as a percentage of the total number of tubers produced in the infested soil and the other consisted in de- termining* the relative proportion of the total tuber surface scabbed. While the first method is the one which has heretofore been used most commonly for determining the amount of scab, there are some objections to it since it gives no indication of the sever- ity of the disease on the tubers. For this reason it was consid- ered advisable to record the results by both methods. The data secured in this way show that the maximum amount of disease as expressed on a basis of the number of tubers does not always coincide with the maximum as expressed upon the basis of per- centage of tuber surface scabbed. This brings up the question as to which of these factors expresses more nearly the truth as to the optimum soil temperature range for the development of the disease. While this may depend to some extent upon point of view, the writers feel that it is not possible to decide this point at this stage in our knowledge of potato scab. It may be that the percentage of tubers scabbed expresses more nearly the optimum temperature for infection, while the percentage of tuber surface scabbed may express more nearly the optimum temperature for the development and progress of the disease after infection takes place. In determining the percentage of tuber surface scabbed, it was necessary to measure all tubers and determine as nearly as pos- sible the surface area of each tuber developing at each of the temperatures. After this process, it was necessary to measure or estimate the area of all scab lesions. From these data the per- 6 Wisconsin Research Bulletin 53 centages of seal) surface were determined for each temperature. In cases of doubt this work was done by two people working to- gether and independently. In all of the greenhouse work herein reported, it is believed that all factors have been suf^ciently well controlled to justify the conclusion that the differences in scab secured are due pri- marily to the recorded variations in soil temperature. Experiment I This experiment was started November 8, 1918, and ter- minated January 25, 1919. While the main object of this series was to determine suitable methods of inoculation for further work, the experiment was so planned that some results might be obtained on the influence of soil temperature on the disease. The soil used in this experiment contained 18 per cent moisture based on the weight of water-free soil. The seed was of good quality ol)tained from North Carolina. The experiment was divided into two parts on a basis of the method of inoculation. One method consisted in inoculating the soil at the time of planting with 50 cc. of a spore and mycelium suspension of the scab organism. The tough leathery surface growth of the fungus was scraped from a thin layer of agar con- tained in six 750 cc. Erlenmeyer flasks. This material was thor- oughly macerated and put in two liters of sterile water and thor- oughly agitated. before applying to the soil. The second method of inoculation consisted in planting pota- toes in sterilized soil and allowing plants to grow at room tem- perature (15-20° C. ). When tubers commenced to develop (7 weeks after planting), part of the soil was removed from the upper roots and tubers of the plants. When the tubers were lo- cated, they were inoculated either on the uninjured or on a scratched surface with the sporulating growth of the scab organ- ism. Soil was carefully replaced over the tubers and roots and the pots were placed in the temperature tanks. Four temi)eratures were maintained, 12° C, 18° C, 24° C, 30° C. Three inoculated pots and one control pot for each method of inoculation were placed at or reserved for each soil temperature. The pots in the soil inoculation series were allowed to remain at 18° C. for a week after planting, then placed in the Influk.nck of Soil 'I'i;mi'i;k.\tiikI'; on Pot.\T(j Scau 7 temperature tanks. The pots in the tuber inoculation series were not placed in the tanks until after inoculation (7 weeks after planting). On January 25 (eleven weeks after planting) all of the tubers were removed and the final data taken upon tuber development and the amount of scabbiness. The results of this experiment showed the superiority of soil inoculation over tuber inoculation. The influence of temperature upon the development of the tubers and plants was quite marked in the series which was carried through the entire period in the temperature tanks. In the case of both series there was a marked temperature influence on the development of scab. In the direct 90 80 ^ 60 % ^ §^ ^'0 > ""^ >^ / _y y / b-«-»* 1 ».^ ••«,. . 10 12 14 16 IQ 20 22 24 26 28 50 50IL TLMP. in degre:e:5 ccnr. FIG. 1. THE INFLUENCE OF SOIL TEMPERATURE ON SCAB IN EX- PERIMENT I. Solid line represents the percentage of tubers scabbed. Dotted line represents the percentage of the total tuber surface scabbed. tuber inoculation series the greatest amount of scab developed in the tank held near 24°, but owing to the exceedingly small popu- lation of tubers these data are not included with the rest of the scab results. In the soil inoculation series the largest percentage of scabby tubers were produced in the tank held near 30°, but the greatest percentage of tuber surface scabbed was in the tank held near 24°. Reference to Table I, Fig. 1, and Plate I will give a clear idea concerning the development of the disease at the different temperatures in the soil inoculation series. 8 Wisconsin Research Bulletin 53 Experiment II This experiment was started February 3, 1919, and terminated March 24, 1919. In this series seven temperatures were main- tained, 12°, 15°, 18°, 21°, 24°, 27°, 27-30° C. The high tempera- ture tank was operated on an alternate rather than a constant basis. For five days the temperature was held near 27° and for two days near 30°. This plan was followed throughout the ex- periment ; a fair development of tubers was obtained from the ex- perimental standpoint. The soil moisture content in this experiment was 18 per cent based on weight of water-free soil. The inoculum consisted of leaf mold cultures and a water sus- pension of the spores of A. scabies added to the soil previous to planting the seed. The seed was obtained from North CaroHna and was a part of the supply which was used in Experiment I. Two plants were grown at each temperature in uninoculated sterilized soil and with the exception of the 24° temperature six plants were grown in inoculated soil at each temperature. In the case of the 24° temperature two of the plants failed to grow, making only four plants in this tank. The plants in the tanks held at temperatures above 15° C. were removed on March 24. In the case of the 12° and 15° tanks one and two plants, respectively, were removed on this date and on account of the slow development at these temperatures, the re- maining plants were allowed to develop for 18 days longer when they were removed from the soil and the tubers examined. The results of this experiment were much the same as those obtained in the previous one, as shown by Table I and Figure 2. There was a marked temperature influence on the host, which will be taken up later, and also on the development of the disease. While the disease developed at all temperatures, the most favor- able one was near 24° C. In spite of the fact that most of the tubers from the 12° and the 15° tanks were not removed until 18 days after the removal of the plants growing at the high tem- peratures, very little scab developed in these two low temperature tanks. In order to get some idea as to the influence of time on the de- velopment of scab at the low soil temperatures, most of the po- tato plants were not removed from the 12° and 15° tanks until 18 days after the removal of all the plants in the other tanks. At 1 .\l'l,ri-..\C !■, Ill' Si)||, I l.M I'i-.K A I I Kl', U.\ I'dlAlo .>(A|; 90 80 CQ 70 S 60 K.SO Q. 10 } \ / \ / > \ / \ y ^ — / > \ / •T. \ / # 0 ••. / ^ • • * • .^ / • • '•.^ ^^^^1 J... • « /O /2 14 16 /a 20 22 24 26 23 JO/L TLMP. in DEGRCE5 CEhT. 30 FIG. 2. THE INFLUENCE OF SOIL TEMPERATURE ON SCAB IN EX- PERIMENT II. Solid line represents the percentage of tubers scabbed. Dotted line represents tlie percentage of the total tuber surface scabbed. the end of this period, the tubers had reached sizes equal to or larger than the tubers produced by the plants removed earlier from the higher temperature tanks, but the amount of scab on these tubers was very slight as compared with the amount de- veloped at the higher soil temperature. This observation indi- cates that the reduced amount of scab at the low temperatures is due to a direct temperature relation and not altogether to the de- layed tuber development. Experiment III This experiment was started January 20, 1920, and terminated March 20 and 21, 1920. It was practically a duplication of Ex- periment II. The amount of inoculum used was reduced slightly on account of the extreme infection resulting in the previous ex- periment. The soil contained 19 per cent moisture based on weight of water-free soil. As in Experiment II, seven .temperatures were maintained, 12°, 15°, 18°, 21°, 24°, 27^ and 28.5-30° C. The highest tank temperature was held around 28.5° until March 8, when it was decided to raise the temperature to 30°. This was done on account of the fact that there was very little ditTerence 10 Wisconsin Research Bulletin 53 between the development of these plants and that of the plants growing at 27° C. The greenhouse temperature ranged between 15.5° and 18.5° C, about 3 degrees lower than in the case of Ex- periments I and II. The results of this experiment are shown in Table I and Fig. 3. These results are in accord with those obtained in the two pre- vious ones in that a marked temperature influence was obtained both in the development of the host and in the development of the disease. While the disease developed at all temperatures, there 90 SO K. /O 12 14 IS IS 20 22 24 26 JOIL TEMP, in DEGREE J CENT, fig. 3. THE INFLUENCE OF SOIL TIOMPERATURE ON SCAB IN EX- PERIMENT III. Solid line represents the percentage of tubers scabbed. Dotted line represents the percentage of the total tuber surface scabbed. y\ / ^\ V r- / \ i / \ / ^^^ / #*• S s^ f .♦•• ...' •• • \ ,' .•' ^ ••• • • • • • •• •• , • •• • • ...• 26 30 was very little at the extremes. In this series the greatest amount of disease developed in the tank held near 21°. This is a little lower than was the case in the previous experiments. The exact cause of this shift in optimum is not known. This point will be taken up later under the discussion on the host plant. Experiment IV This experiment was started December 25, 1920, and termi- nated March 19 and 20, 1921. As in the previous experim.ents, seven temperatures were maintained in this series, but there was some slight modification in the temperatures at which the individ- ual tanks were operated. Tanks were held near the following tem- peratures, ir, 14.5°, 18°, 21.5°, 25°, 28.5°, and 30.5° C. I XKLUK.NCii (JF Soil 'I'k.mi'Kkatuke ox I'otato Scai! 11 'i he methods used were essentially the same as in Experiment 111. The inoculum consisted of a water suspension of A. scabies added to the soil. A considerably larger quantity of the organ- ism was used in this experiment than in the previous ones for the reason that it was feared the organism might have lost some of its virulence. The final results, however, show that this was not the case and a decided over-infection resulted. While the results (Table I and Figure 4) of this experiment show that soil temperature has a marked influence on the disease, the extreme degree of infection made it difficult to interpret the relative amount of tuber surface scabbed. As in the previous ex- periments, the disease developed at all temperatures with the least amount developing at the extreme temperatures. 90 "? eo '^SO ^,0 / / *'%. V -y -. ^^^. \/~ ' t ■ • vx • • at T« • -^ %; 10 IZ 14- 16 15 20 Z2 24 26 23 30 JO/L TCMP. Ih DEGREE 6 CENT. FIG THE INFLiUENCE OF SOIL TEMPERATURE OX SCAB IN EX- PERIMENT IV. Solid line represents the percentage of tubers scabbed. Dotted line represents the percentage of the total tuber surface scabbed. Taking the number of tubers scabbed as a basis, it will be noted from Table I and Figure 4 that the maximum percentage of dis- eased tubers developed at 21° C, as was also the case in the pre- vious experiment ; however, it will be noted that the greatest per- centage of tuber surface scabbed was at 18° C. This shifting of the optimum temperature for the development of the disease will be taken up in the discussion of the temperature tank results. The increased rate of scabbing shown at the highest temperature 12 Wisconsin Research Bulletin 53 is not looked upon as being very significant since only two tubers developed. Experiment V This experiment was practically a duplication of Experiment IV with the exception of a few minor details. The series was started March 23, 1921, and terminated May 27 and 28, 1921. As in Experiment IV, seven tanks were held at or near 11°, 14.5°, 18°, 21.5°, 25°, 28.5°, and 30.5° C. The amount of inoculum used was reduced considerably under the amount used in Experiment IV and a considerable reduction in the severity of the disease resulted. 90 .O 70 I. 60 > <^40 ^ JO K fO ^ «-^_^ > -^ \ ^/ y^ \ _^__^ „- — ^ ..••" • • \ ...•'' • ^ •, • \ ..•• •• '. * •• • '••. *. • • •* • • • • /O 12 14 16 IQ 20 22 24 26 25 JO JO/L TCMP in DEGREES CEfiT . FIG. 5. THE INFLUENCE OF SOIL, TEMPERATURE ON SCAB IN EX- PERIMENT V. Solid line represents the percentage of tubers scabbed. Dotted line represents the percentage of the total tuber surface scabbed. The results of this series are shown in Table I and Figure 5. The optimum temperature for the development of the disease as determined by the percentage of tubers infected was found to be at or near 24°, while the greatest percentage of tuber surface was scabbed at 21.5°. Plate II shows one fifth of the tuber popula- tion in the inoculated pots in Experiment V. These tubers were selected so as to show as near as possible the true variations in the development of the tubers and in the amount of scab at the various temperatures. [m'ij'I' xci', di' Siiii, Ti'.M im:k' \irui', o\ To'iaivi Scai-. 13 Table I — TabulatI'D data showing thi': rNFLUKXcic ok soil tkmpkkaiure on the dkvij.opmlcnt of potato sfab in thr te.mphkatukk tank expp:rimi;n'is. Kxporiment I. Soil temperature 'O. No. tubers produced Per cent tubers ecabbed.. Per cent surface scabbed. 12 15 18 21 24 27 ."30 11 n 10 2 0 12.5 43.0 50.0 0 .31 11.7 1..3 Experiment II. Soil temperature °C. No. tul)ers produced Per cent tubers scabbed- . Per cent surface scabbed. 12 15 IS 21 24 27 27 to 30 4 40 130 . 146 66 180 128 0 2.5 54.1 54.8 97.0 67.7 42.1 0 .0001 14.5 27.4 46.3 15.0 1.62 Experiment III. Soil temperature *C. No. tubers produced Per cent tubers scabbed. - Per cent surface scabbed- 28.5 12 15 1,S 21 24, 27 to 30 S2 80 49 46 55 58 69 28.0 65.0 61.6 85.1 50.0 40.9 25.3 2.7 21.7 18.5 33.3 5.5 1.8 8.6 Experiment IV. Soil temperature "C. No. tuljers produced Per cent tubers scabbed-- Per cent surface scabbed. 14.5 1 18 21.5 25 28.5 CA 40 .'53 49 66 28 2 .■J2.8 62.1 57.6 88.4 70.0 30.0 50.0 1.83 21.0 67.3 49.2 37.5 3.1 5.5 Experiment V. Soil temperature °C. No. tubers produced Per cent tubers scabbed.. Per cent surface scabbed. 11 : 14.5 18 21.5 26 60 42 34.5 40.0 45.2 .12 9.33 31.7 25 30.5 55 93 87 33 65.4 75.2 64.3 32.2 48.1 28.1 10.1 .48 14 Wisconsin Research Bulletin 53 Results Table I includes all scab data obtained in the preceding experi- ments and Figure 6 shows the average of all these data graph- ically. While the results have shown some fluctuation in the tem- perature optima for the development of scab, the average of all experiments indicates that while the disease operates over quite a wide range of temperatures, it develops in greatest abundance at soil temperatures ranging from about 20.5° to 23° C. (70-73° F.). As pointed out earlier in this bulletin, there is some question as to which factor (percentage of total number of tubers scabbed or the percentage of total tuber surface scabbed) is the most signi- 10 12 14 16 la 20 22 24 26 2& JO JO/L TLMPER/1TUF<€S Ih OEGRLES CCnTIGRADC: FIG. 6. THE influence OF SOIL TEMPERATURE ON SCAB IN EXPERIMENTS I, II, III, IV AND V. Dotted line represents the average percentage of the total tuber sur- f3,C6 SCcibbGd. Dashed line represents the average percentAge of tubers scabbed. Solid line represents the average between the percentage of the num- ber of tubers scabbed and the percentage of the total tuber surface scabbed. The latter is probably the most significant curve since it takes into account both the number of tubers scabbed and the degree of scabbiness. The authors are therefore accepting this as the basis for their final conclusions as to the relation of soil temperature to the development of potato scab. ficant in expressing the amount of disease developed at the vari- ous temperatures. However this may be, an average between the curves for both these factors, as shown in Figure 6, gives an optimum temperature of 22° and it is this temperature which the writers are considering for the present as the optimum for the development of the disease. It has been noted in this work that the temperature optimum for scab development has shifted from time to time. This shift- ing seems to indicate that certain aerial factors, not so well con- Influkxck ok Soil 'Ikmi-kkai ukk o.n I'otaio Scm; 13 trolled as the scjil factors, may have had an influence on the de- velopment of the disease. It is of interest to note that the fluctuation of the scab tempera- ture optimum in the several experiments is more or less corre- lated with the shifting of the temperature optimum for the rate of tuber development as expressed in terms of the average weight per tuber. These correlations are shown in Tables I and \' and in Figure 7. NUMBER or CXPCFUMCNT FIG 7 THE TENDENCY TOV^ARD CORRELATION BETWEEN THE SOIL TEMPERATURE OPTIMA FOR SCAB DEVELOPMENT AND THE RATE OF TUBER DEVELOPMENT IN THE FIVE TEM- PERATURE TANK EXPERIMENTS. Points connected by the solid line represent the temperature optima for scab in the different experiments. ^„^^of„r«. ^ntimn Points connected bv the dotted line represent the temperature optima for rate of tuber development in the different experiments. ,,.^^,„^ The rate of tuber development is calculated on a basis of the average weie-ht per tuber produced at each soil temperature The soil_ tem- perature giving the highest average weight per tuber is considered the optimum for the rate of tuber development. , , , ^ j tv.^ The tendency towards correlation between scab development and the average weieht per tuber suggests that scab development is dependent on tuber development and thlt rapidly growing tubers tend to be more scabby than those which develop slowly. While the optimum for the development of scab did not fall as low as that for tuber development in Experiments IV and V, it will be noted that the scab optimum seemed to be influenced by the extreme drop of the tuber optimum in Experiment IV. Ob- servations indicate that scab develops only on growing tubers. In all of the work carried on by the writers scab has never devel- oped on any seed piece planted in inoculated soil. In the case of experiments involving the application of inoculum directly to the 16 Wisconsin Research Bulletin 53 surface of developing tubers it has been found that a large per- centage of such tubers never develop after this disturbance. In all such cases scab has never developed on such tubers, whereas a large percentage of inoculated tubers which proceeded in their development did develop scab in a greater or less amount. These observations are in accord with the results obtained by Weiss and Orton (13) in connection with the black wart disease of potato, and there is also considerable evidence^ which indicates that the development of scab lesions caused by Venturia inaequalis on the fruit and leaves of the apple is dependent more or less upon the growth of the host. The tuber development and disease correlation in this work suggests that rapidly growing tubers are, within certain limits, more likely to become severely scabbed than tubers developing more slowly. In the event of the ultimate verification of this relation, it will not be dif^cult to understand how such aerial fac- tors as light, temperature, humidity, and gas balance may within certain limits influence the development of scab through a direct influence on the above ground parts and thence upon the tuber. In the experimental work involving the use of the scab organ- ism it has been found that the underground bases of the stems of the potato plant and also the stolons often develop severe scab lesions (Plate III D, E) which usually originate in the lenticels. This condition seldom occurs at soil temperatures below 24° C. and it becomes more pronounced as the temperature rises. As in the case with tubers, stems are clean and white in the control pots not containing the scab organism (Plate III C). Gussow (2) has also noted that the scab organism attacks the underground stems of potato plants. EXPERIMENTAL WORK IN THE FIELD Methods Soil temperature influences have been studied in a preliminary way under field conditions. In this connection two methods have been employed in studying variations in soil temperature. One consisted in maintaining three soil temperature gradations in small plots by means of special apparatus and the other consisted in planting inoculated seed at intervals throughout the season in ' Unpublished observations made by G. W. Keitt. of the Department of Plant Pathology, University of Wisconsin. iNi'LUENCii OF Soil TEMi'i:i / y '"" ^ J \ -- ^ < ^v ■~» \ •• ( ^ ^ ^' > V ^ \ 5 -- ') ) / \ \ ; \ ! \ : N r \ ) i V \ \ { ( (^ \ > ' } t>l. 'll 1. \ r a in /' \ i' I / J / ^ y" i \ \ V -. -s, ^ O SP CNi '■ V ^ r •'' ^ J V \ \. y > ' .• '■ > ^ t, m 1 o i* :»: « is « ■2 -o ^^e t~ o ■a CIS « « 'J- . 2 4 20 16.7 1.5 3.4 6 20.0 l.SO 6 24.3 1.12 4 16.5 1.87 6 30.0 .55 6 21.3 .22 ,_, Soil temperature °C. 12 15 IS 21 24 27 28.5 to 30 No. of hills - . 8 13.7 2.46 8 13.3 4.54 8 8.2 6.9 8 7.75 7.46 8 9.1 6.41 8 9.75 5.32 8 Av. no. tubers to liill Av. wt. i)€r tuber 11.5 2.31 > Soil temperature °C. 11 14.5 18 21.5 25 28.5 30.5 6. M No. of hills Av. no. tubers to hill 8 10.7 3.03 8 6.62 8.9 8 8.87 6.43 8 8.12 6.2 1?.1 3.9 5 4.6 .64 Lo 1.5 > Soil temperature °C. 11 14.5 18 21.5 25 28.5 30.5 6, M w No. of hills - Av. no. tubers to hill 8 4.33 3.8 8 10.0 7.02 8 ! 7.0 i 12.85 i 8 9.16 9.25 8 15.5 4.71 8 14.5 1.99 7 6.5 .34 •One and two plants were removed from 12° and 15° C. tanks respectively at the time the data was taken on the plants growing at the other tempera- tures in the experiment ; these data are represented by the numerators while the denominators are the data for the remaining plants grown at 12 and 15° C. which were removed eighteen days after the removal of the plants from the higher temperature tanks. The numerator figures are comparable with the figures recorded for each of the temperatures above 15°. 28 Wisconsin Reskakch Bulletin 53 Underground Parts Tubers ■ Variations in soil temperature seem to have less consistent in- fluence on the number of tubers produced per hill than on any of the other host activities thus far observed. However, there does seem to be a tendency for the greatest number of tubers to de- velop at 15° C. and at 25°-28° C. There is a reduction in the number of tubers per hill at the intervening and the extreme tem- peratures. This curve tends to be the reverse of all other curves representing the plant activities observed in this work. The size of tubers is influenced considerably by variations in soil temperature. During the period involved in the various tank experiments, the largest tubers were produced at soil tempera- tures ranging from 15° to 22° C. with the optimum at or near 18° C. However, when plants are allowed to develop beyond the period of the main experiment, as was the case with the plants held at 12° and 15° C. in Experiment H, the tubers progress rap- idly and increase in size as shown in Table V. It will not be sur- prising to find that the optimum soil temperature for mature tuber development is somewhat lower than just indicated, when a complete series can be carried through to maturity. Observations have not been made to determine the soil tem- peratures at which tubers commence to set first, but it is believed that this takes place at the temperatures which favor early sprout- ing and the emergence of the sprouts from the soil. The indica- tions are that the optimum temperature for the early setting of tubers is somewhat higher than the optimum for a yield of large tubers. The shape of tubers is influenced to a considerable extent by Table VI — The Ratio of Width to Length in Tubers Grown at Different Soil Temperatures. These data arc based on measurements mat ment V.' 0 on the whole population of Experi- Soil temperature '0. n 14.5 18 21.5 25 28.5 30.5 Ratio ot tuber width to length. 1:0.9 1:1.02 1:1 1:1.12 1:1.14 1:1.25 1:1.6 > The Irish Cobbler seod used in this work was typical of the variety which tends towards the globular shape. Most of the tubers were prac- tifaily circular in the cross section intersecting' the stem and bud axis at right angles, and for this reason width is designated as a single factor In Table VI. Influence of Soil Tkmfm:kature on Potato Scab 29 soil temperatures as shown in Plates J, II and IV'. At low soil temperatures the length of Irish Cobbler tubers is less along the stem-bud axis than along the transverse axis, while at the higher temperatures the stem-bud is much the longer and tubers tend to become egg or pear shaped, as shown in Plates I, II and IV. Ac- tual measurements made on the population of a complete temper- ature series show that the ratio of the width to the length of tubers grown at 11° C. is about 1 :0.9, while at temperatures near 30° C. this ratio approximates 1 :1.6. The ratios developed at the intervening temperatures gradually approach the latter ratio as the temperature^ rises forming a rather regular curve as is shown by the data in Table VI. Fitch (1) has noted that under certain conditions potato tubers tend to elongate and develop the pear shape. He associates this condition with drouth, the "running out" of seed stock and sea- sonal conditions. While factors other than soil temperature may influence the proportional dimensions of tubers, the results herein recorded sho\v definitely that soil temperature is an important factor in determining tuber shape, a point which is of economic interest to growers who produce exhibition seed stock. It is a matter of common observation that lenticel development on potato tubers is stimulated by certain moisture conditions. In these experiments where soil moisture was kept approximate- ly uniform, it has been found that lenticel character has also been influenced considerably by soil temperature. These organs have not been conspicuous on tubers developed at low soil tempera- tures, but have become large and protruding at the high tem- peratures. An evident suggestion is that their relative develop- ment may be associated wnth respiratory metabolism. The influence of soil temperature upon the chemical composi- tion of the tuber has not been a matter of direct inquiry in con- nection with this work. It may be assumed that the composition is so influenced and a type of evidence that bears upon this de- serves record. In Plate V is shown a complete temperature series of tubers grown in uninoculated soil. These tubers were photographed after storage in 70 per cent ethyl alcohol for five months after their removal from the soil. The tubers pro- duced at the low temperatures are jet black in color, whereas those grown at the medium temperatures show practically no dis- coloration. At 30° C. the tubers show a slight darkening, but to 30 Wisconsin Research Bulleti.x 53 Table VII — Data Showing the Influence of Soil Temperature on Certain Host Developments. These data are the average of the results from experiments II, III, IV and V. All data are based on detenninatious made at the close of each experiment. All weights are in gram.s and linear measurements in centimeters. Soil temp, degrees O.* 11 12 8.2 22 2.55 55 . 1 2.8 22.0 7.1 23.0 14.5 15 18 21 21.5 24 25 27 28.5 27 30.5 Av. No. of tubers in inoc. hills Av. Wt. of tubers per hill Av. Wt. per tuber 12.5 42.7 5.44 11.2 56.0 7.0 12.5 54.2 6.0 12.8 51.0 4.2 14.4 25.0 2.1 10.2 9.5 .80 Above ground parts Av. Wt. of green tops p«r hill (Exp. IV and V) . 77.0 3.S 23.0 8.5 17.5 82.2 4.2 22.6 7.9 14.5 89.0 4.6 23.5 7.7 12.8 87.0 4.0 25.0 S.7 12.8 70.0 3.21 30.2 3.9 16.5 50.0 Av. No. of stems per hill (Exp. II, III, rv and V) 2.8 Av. height of stems per hill Exp. II, ni, IV and V) Av. diam. of stems^ (Exp.V)-. Av. No. days for plants to come through soil (Exp. Ill, IV and V) 14.0 3.3 24.6 a less degree than is the case with the tubers produced at the low temperatures. The black coloration suggests a melanin relation involving the reaction of various enzymes, tyrosin and other com- plex proteins. While this variation cannot be interpreted at this time, it is evident that the chemical composition of at least the surface tissues of these tubers was influenced by the soil temperatures. The question naturally arises as to what relation such changes may have upon susceptibility or resistance to scab infection or development. Iliese are evidently matters deserving further investigation. Stem bases and sloloiis These parts of the potato plant seem to be greatly modified by variations in soil temperature, as noted previously by Richards (8). At the high temperatures they become relatively large in diameter and fleshy in nature. The stems are very much larger below ground than they are above the soil line, whereas at the low temperatures the reverse is true, the underground stems being more slender in proportion than the above ground stems. As the temperature advances toward the higher limits of endurance the basal portion of the main stem as well as the stolons evidently * The soil tomperature.s given arc Ihnso u.^cd throughout all the ex- periments. Siiiee ther(> was so little (liffereiice between the tempera- tures used in IOx])erimeiits II and III, and those used in lOxperiinents IV and V all data in tjiis table were averaged in the usual manner. iNKLUENCli ()]'• Soil 'I'i:.M I'KKATUKIi ox I'oTATO ScAIi 31 assumes a storage function. This may be correlated in some de- gree with an inhibition of normal tuber development and conse- quently with a disturbance of translocation processes. As in the case of the tuber, ienticel development is very marked on stems and stolons at high temperatures and much reduced at low ones. Above Ground Parts Richards (8) has published upon data which are essentially in harmony with those obtained by the writers concerning the re- sponse of the above ground parts of the potato plant. The results herein reported, however, cover somewhat longer periods of de- velopment and a larger number of experiments than those which he discussed. Stems Soil temperature greatly influences the length of time required for germination and the emergence of the sprouts from the soil as shown in Table VII. Sprouts emerge first in soil held from 21°-24° C. and in general emergence is earlier at the high than at the low temperatures. During the early life of the plants the height of tops is directly correlated with the time of emergence. A soil temperature of about 27° C. soon becomes the optimum for these organs with a sharp decline in the curve above this tem- perature. Later, however, these plants at the higher soil tempera- tures commence to slacken their growth and come to maturity, whereas the plants at the low temperatures come on slowly and finally surpass plants grown at the high temperatures. In general the life cycle of the potato plant is short at high and long at low soil temperatures. The green weight of "stems" is also influenced by variations in soil temperature, but this factor seems to vary more or less di- rectly with the number of "stems" produced at the various tem- peratures. Apparently the optimum soil temperature for the pro- duction of green weight and that for the production of number of "stems" per hill is at or near 21° C. The two curves differ, how- ever, in that the green weight curve tends to show a wider opti- mum range than the "stem" curve. Both curves decline grad- ually from the optimum point. The development of "wings" on the "stems" is much reduced and the swelling of the nodes is much increased by high soil tem- 32 Wisconsin Research Bulletin 53 peratures and in some cases aerial tubers and auxiliary branches develop at the higher temperatures. The leaves of the potato plant also show variations due to soil temperature. In general the ratio of width to length in leaflets follows the same trend as is the case with tubers. At the low soil temperatures leaflets are wider in proportion to their length, whereas they are longer than wide at the high temperatures (Plate III A, B). At these latter temperatures leaflets are decidedly lanceolate in form, but they tend towards the round type at the low soil temperature. While these observations have dealt primarily with the evident characters, chiefly morphological, it is realized that the more im- portant modifications may be those occurring in the physiological processes, especially in nutrition and including food translocation or storage. This is indicated not only in the nodal swellings and aerial tuber developments noted above, but also in the influence of soil temperature upon chlorophyll, the foliage tending in gen- eral to a deeper green at the lower soil temperatures and a lighter color at the higher. We have not the data to go far in the discussion of the details as to the relation of even this single variable factor, soil tempera- ture, to the normal development of the potato plant and its tuber- ization processes. At the same time it is evident that an adequate discussion of these must include some consideration of their in- terrelation with the other variables in the environment, especially air temperatures and light. While these observations relate primarily to the host plant, their significance must not be overlooked from the standpoint of the disease. As pointed out earlier, the evidence at hand indi- cates that potato scab as a disease is influenced by the conditions of tuber development. These include the suggestion that rapidly growing tubers may scab more severely than slow growing ones, and that there may be such differences in the chemical composi- tion of the tubers developing under diflferent conditions as may influence their relative susceptibility to infection. The rate of the tuber's development, as well as its composition is, of course, cor- related, with the metabolic and growth processes of the rest of the plant. It is obvious, therefore, that the ultimate understanding of the influence of soil temperature or other variable factors upon the development of a disease like potato scab is conditioned upon further studies concerning the relation of these to host as well as to parasite. Jnklukn'ce of Soil Tkmi'Kratukk ox I'iitato Scat. 33 SUMMARY 1. The development of potato scab, caused l)y Actinomyces scabies, is evidently influenced by several environmental factors. An attempt has been made to secure evidence as to such possible influence of soil temperature. 2. Five series of experiments have been conducted in green- houses using the "Wisconsin tank" method. These have included seven gradations of soil temperature ranging from about 11° to 30.5 C. In all cases under this method the aim is to maintain the other soil conditions including moisture alike and approximately constant, and in each experimental series all the plants are ex- posed to the same aerial conditions. 3. The results show that under these circumstances the devel- opment of the scab disease is influenced by soil temperature. 4. The disease developed at all soil temperatures, 11^ to 30.5° C,. but was comparatively slight at either extreme. 5. The optimum soil temperature for scab development as measured by the number of scabby tubers was found to be about 23° C, while the optimum for the percentage of the total tuber surface scabbed was a little lower, about 20.5° C. The conclu- sion reached is that all things considered 22° C. may be accepted as about the optimum soil temperature for scab development, where the "Wisconsin tank" method is used. 6. A field trial was also conducted in which three gradations of soil temperature were maintained during the season of tuber development. Of these the highest, approximately 25° C, proved to be most conducive to scab development. 7. Field observations seem in general to accord with the results obtained by experiments. They indicate that potato scab is com- paratively more prevalent in regions having high summer tem- peratures than in those of lower temperature, and also that in the same district in Wisconsin the disease development is greater dur- ing hot than during cool summers. It is to be noted, however, that such observational data is relatively meager and is to be con- sidered only as suggestive. Attention should be directed to this question by other observers. 8. As bearing more definitely upon this matter it is found that in the leading Wisconsin potato districts the mean temperatures 34 Wisconsin Research Bulletin' 53 for July and August during the hottest midsummers approximate those found in our experiments to be the optinnun for scab devel- opment. 9. Examinations of the scabby tubers from the controlled tem- perature experiments as well as of samples from commercial sources show that the scab lesions tend to be segregated upon the "stem end" portion of the tuber. 10. The evidence from the soil temperature tank series shows that such segregation is less evident at or near the optimum tem- perature for scab development and more apparent at the extreme temperatures. 11. The influence of temperature upon the development of the disease must obviously bear a relation on the one hand to effects upon the parasite and on the other hand to effects upon the host. 12. The available evidence indicates that the scab parasite as an independent organism is favored by relatively high temperatures, whereas the potato plant functions better in general at relatively low temperatures. It is, however, noteworthy that the influence of temperature upon the different potato organs is not uniform and that it varies also with the stage in their development. 13. The influence of temperature upon the prevalence of the parasite in the soil may be cumulative from season to season whereas the influence upon the host is immediate and temporary. 14. The immediate relation of temperature to the development of scab seems to be more closely correlated with its influence upon potato tuber development than with that upon the growth of the parasite. 15. It seems evident that a satisfactory interpretation of the relation of soil temperature to the development of potato scab must await on the one hand more critical study of its relation to tuber development and on the other hand a fuller knowledge of the details as to tuber infection and subsequent scab development. LITERATURE CITED (1) Fitch, C. L. 1914. Identification of potato varieties. Off. Pub. Iowa State Col. Agr., V. 12, no. 33 (Ext. Bui. 20), 32 p., 25 fig., 1 col. pi. (2) Giissow, H. T. 1917. The pathogenic action of Rhizoctonia on potato. In Phytopathology, v. 7, no. 3, p. 209-213, 1 fig. iNFLUKNCfc; OF SoiL TeM I'KRATUKK OX PoTATO ScAB 35 (3) Jones, L. R. 1905. Disease resistance of potatoes. U. S. Dept. Agr., Bur. Plant Indus. Bui. 87. 39 p. (4) Jones, L. R. 1922. Experimental work on the relation of soil temperature to disease in plants. In Trans. Wis. Acad. Sci., Arts and Letters V. 22, p. 433-459, pi. 33-37. (5) and McKinney, H. H. 1919. The influence of soil temperature on potato scab. In Phytopathology, v. 9, no. 7, p. 301-302. (6) Orton, W. A. 1913. Environmental influences in the pathology of Solanum tuberosum. In Jour. Wash. Acad. Sci., v. 3, no 7 p 180-190 3 fig. ' ' (7) ■ . 1916. Report on potato diseases in Bermuda. In Rept. Bd. Agr. Bermuda, 1914-15, p. 13-15. (8) Richards, B. L. 1921. Pathogenicity of Corticium vagum on the potato as af- fected by soil temperature. In Jour. Agrl. Research, v. 21, no. 7, p. 459-482, 5 fig., pi. 88-93. Literature cited, p. 481-482. ' (9) Shapovalov, Michael. 1915. Effect of temperature on germination and growth of the common potato-scab organism. In Jour. Agr. Research, v 4 no. 2, p. 129-134, 1 fig., pi. 15. (10) Smith, J. Warren. 1911. Correlation. In Mo. Weather Rev., v. 39 no 5 p 792-795 4 fig. . . , (11) 1919. The effect of weather upon the yield of potatoes. In Potato Mag., v. 1, no. 10, p. 11-14; no. 11, p. 15-17; no. 12, p. 7, 16-17, 27; v. 2. no. 1, p. 16-17, 33-34. 23 fig. References, p. 34. (12) Verrill, Addison E. 1901-02. The Bermuda Islands. In Trans. Conn. Acad. Arts and Sci., V. 11. pt. 2, X p., p. 413-[957]. 245 fig., pi. 65-104. Also separately issued, 1903. (13) Weiss, Freeman, and Orton, C. R. 1922. Progress notes on potato wart disease investigations. In Phytopathology, v. 22, no. — , p. (14) Wollenweber, H. W. 1920. Der Kartoffelschorf. Arb. Forsch. Inst. Kartoffelbau. Heft. 2, 102 p.. 11 fig., 2 pi. (partly col.). ( /2' 4, -^^^ 7(5' 2^' JO I'l^ATE I. I'AllT OF THE TUBERS FROM EXPERIMENT I SHOAVIXO T?IE INFLUENCE OF SOIL. TEMPERATURE ON THE DEVELOPMENT OF POTATO SCAB. Note the severity of the infection on the tubers grown in the soil held near 24° C. and the tendency towards tuber elongation and the de- velopment of pear-shaped tubers at the higher temperatures. we t //■ 21.5 4. W 14.5' ' ^ r <'^- \^0l .^ f C-f ^^ 25'' ( etc 285'' 305' % t ^ PLATE II. A REPRESENTATIVE ONE-FIFTH OF THE INOCULATED TUBER POPULATION FROM EXPERIMENT V. The greatest amount of scab is shown on the tubers grown at 21° C. (soil temiierature). Note the tendency for tubers to elongate at th« high temperatures and to thicken transversely at the lowest tempera- tures. PLATE 111. EF'FKCT OP TEMPERATURE ON PLANTS. A. A typical plant f;^■<|^vn at a soil temperature oi 15° C. Note the low stocky development and the wide rounded leaflets. B. A typical plant grown at a soil temperature of 27° C. Note the relative elongation of leaflets, some approaching' the lanceolate form, also the elongation of stems and the more upright habit of growth. C. An underground stem grown in uninfested soil. Note that this stem is free from any evidence of scab lesions. D. and E. Underground stems, stolons and tubers grown in disin- fected soil, which was infested with a pure culture of A. scabies previous to planting the seed. Note tlie scab lesions on the stems and stolons as well as on the tubers. Stem and stolon lesions seem to originate in the lenticels. # ^ 0^ ^fc* /r f" '« ^ f c r € * t 9 \ f>ef r #i € 2S.5' ^h^^ ^^- ^ REPRESENTATIVE ONE-FIFTH OF THE TUBERS FROM THE UNINOCULATED OR CONTROL POTS FROM RE PRE SENTATrV^E TEMPERATURES IN EXPERIMENT V. All the control tubers were scab free. The influence of soil tempera- ture on tuber shape IS shown in this series by the elongated tubers at trie high and the short tubers at the low temperatures. The dark pig- ment developrnent is shown in the tubers produced at the extreme ten peratures become Jet black #•• /2' Ql /s 2/ i-^*. €> 2^ 30° rL.AT10 V. ALL OK THK fXLXt )( 'UL ATKI > I'oXTUOL Tl'HKIlS FllM.M EXPEIUMEXT HI SITOWLXG THE IXFLrEXCE OF SOIL TEMPERATURE OX THE DISCOLORATIOX OF TUBERS. This material was photogiaphed five months after removal from the poil, during which period it was ])reserved in 70 per cent grain alcohol. These results indicate a difference in the chemical nature of the tubers produced at the various soil temperatures. Such a difference may be sisnificant in the final interpretation of the results obtained with scab at the various soil temperatures. All of these tubers were free fiom scab. Research BuUetin 58 May, 1924 Anthracnose of Cane Fruits and Its Control on Black Raspberries in Wisconsin LEON K. JONES Agricultural Experiment Station of the University of Wisconsin Madison Contents Page The disease 1 Description 1 F.coiiomic importance - „ 2 The causal organism 4 Cultural studies 4 Germination studies 5 Pathogenicity 6 Life history 7 Seasonal development 7 Production of spores 7 Source of inoculum in nature 8 Control Measures 9 Sanitation 9 SMrrV 9 Summary 24 Literature cited 25 Anthracnose of Cane Fruits and Its Control on Black Raspberries in Wisconsin^ A SURVEY of the cane fruit industry of Wisconsin by the writer (1920) in 1919 showed that anthracnose was one of the chief factors responsible for the drastic decline of the black raspberry acreage of the state in the last decade. Consequently, it was deemed advisable to make a careful study of this disease, with special reference to control measures. Field and laboratory experiments were conducted at Madison. Wisconsin, in the period of 1920-23. The results of the writer's studies as they relate to history and geo- graphic distribution of the disease, pathological histology, taxonomy, and morphology confirm those reported by Burkholder (1917). Burk- holder'g account of these subjects is so satisfactory that it appears un- necessary to treat them in this publication. THE DISEASE In the United States the disease caused by the fungus PlectodisccUa vcncia Burk, ( (Ucosporium vcnctuni Speg.) appears to be widespread throughout the north and also in hilly southern regions, coinciding with the ranges of its hosts. It is generally distributed on the following hosts in Wisconsin as shown by the writer (1920) : red raspberry (Rubus idaeus var. aculeatisshmis (Mey.) Kegel and Tiling), black raspberry (Rubus occidcntalis L.), purple-cane raspberry (Rubus ncgJcctus Peck), and black- berry (Rubus sp.). The relationships of the pathogens causing anthracnose on the above named hosts have not been definitely determined by cross inoculations, except as reported by Burkholder (1917) that the organism isolated from purple-cane raspberry through inoculations pro- duced infection on black and red raspberry. Description The symptoms manifested on the various hosts of P. vcneta are some- what similar, although they vary in the color, shape and size of the lesions produced, depending on the host and the severity of the attack. Canes, leaves, petioles, peduncles, pedicels and fruits may be attacked, although the symptoms on the canes are usually the most noticeable. Descriptions here given refer to symptoms on Cumberland black raspberry unless otherwise noted. On canes. Elliptical to circular greenish-brown lesions one-half to one millimeter in diameter usually appear on the young shoots in early spring when the latter are eight to ten inches high. These lesions are slightly sunken, with a stromatic development of the fungus in the center, which is somewhat darker and raised. Under binoculars the ' The \Nriter Avishes to express his indebtedness to Dr. G. \V. Keitt, of the University of Wisconsin, under whose direction the work was performed. 2 Research Bulletin 59 aflfected tissue has a slightly water-soaked appearance. The lesions enlarge slowly and the centers become a pale buff to white, while the advancing margin is raised and reddish-brown to purple in color. Mature lesions (Plate I) are circular to oval and seldom become more than one centimeter in diameter, although they often become confluent, making large irregular patches that may encircle the cane. In cases of severe attack, the canes may crack longitudinally (Plate I, B). These cracks, usually small, may split the cane to the pith for a distance of two to three inches. The lateral branches often become seriously infected. The resultant lesions are similar to those on the canes, but smaller, and often cause the death of the young branches. On leaves. The first spotting of leaves in the spring appears about the same time as that on the canes, although not so abundantly. The lesions first appear as yellow or straw-colored oval to irregular spots one-half to one millimeter in diameter. The center of the lesion is raised and brownish, while under binoculars the veins of the leaf at the outer edge of the lesion are slightly purple. The mature lesions are one to two millimeters in diameter with light colored centers and purple margins. These spots may drop out, giving the leaf a ragged or "shot-hole" appearance. In cases of severe infection of red raspberries, the leaves may turn yellow and drop. On the leaves the symptoms of this disease are often confused with those of common leaf spot (Mycosphaerella ruhi Roark). The Mycosphaerella leaf spot lesions differ from those of anthracnose in being irregular in outline and somewhat larger, with minute black pycnidia usually present. On petioles, peduncles and pedicels. Anthacnose has been found commonly on these plant parts. The lesions produced are similar to those on the canes, although smaller and often without the purple margin. On the peduncles and pedicels especially, they may coalesce into white scab-like patches (Plate II) that cause these parts to become brittle. These white patches often retard growth on one side of the attacked part, causing it to curl and crack. On fruit. Infection occurs less frequently on the fruit than on the other parts of the host. On one occasion the writer observed fruit in- fection on the variety Plum Farmer in Wisconsin. One or several drupelets become brown and sunken. Frequently the whole fruit be- comes brown, dry and woody, while the healthy berries are still green. Economic Importance Anthracnose is one of the most serious diseases of black raspberry and blackberry, although it seldom causes much injury to red rasp- berry. Burrill (1882) cites an instance of a plantation that had yielded a profit of $400.00 per acre, on which one attack of this disease re- duced the proceeds so that expenses were not met. Scribner (1888) estimates the losses in southern Missouri due to anthracnose on black raspberries at 10 to 12 per cent. Burkholder (1917) states that in certain localities in New York state growers have been obliged to discontinue berry growing due to anthracnose, and that it is evident Anthracnose of Cane Fruits 3 that anthracnose is correlated with reduction of yields. Anderson (1920) states that "Anthracnose has entirely eliminated the growing of raspberries in some sections of Illinois, and many growers are com- pelled to renew their patches after two years of bearing." He also estimates that in Illinois in 1908 the loss from anthracnose was 50 per cent of the crop, and that 25 per cent of the berry crop is lost there annually because of this disease. In a survey of Wisconsin the writer (1920) found that anthracnose was one of the most important diseases of cane fruits, and was found wherever raspberries were grown, al- though it was of very little importance on blackberries and purple-cane raspberries. Red raspberries usually show a light spotting of the canes but the writer has not noted important anthracnose injury on red varieties in Wisconsin except in the vicinity of Eau Claire. In this district in 1919 there was considerable spotting of the leaves, which caused yellowing and dropping of the foliage. The disease is most important in the state on black raspberries and, in association with crown gall injury, it is a limiting factor in the black raspberry industry. The disease aifects the canes and leaves in the first season of growth, thereby weakening the plant and causing a decrease in fruit yield the ensuing year. The diseased canes are also more subject to winter injury than healthy ones. A very important injury in Wisconsin is caused by the lesions on peduncles and pedicels. Abundant infection on these host parts causes the fruit to be small or to dry up before maturity (Plate II). In order to obtain data on the decrease in fruit yield due to anthrac- nose the writer made counts and weighed the fruit harvested from sprayed and unsprayed plots of Cumberland raspberries in 1921. The sprayed plot, consisting of 24 plants, had received two applications of lime-sulfur with gelatin as a spreader, as outlined on later pages. The disease had been very satisfactorily controlled on this plot the previous season, while the unsprayed plot, consisting of 12 plants in the same planting, had never received any spray treatment and the plants were abundantly infected with the disease. A sunmiary of the data obtained is presented in Table I. Table I. — Comparison of Fruit Yields of Spraveo and Unsprayed Cumber- land Raspberry Plants. H. I'lscH^iR Planting, Madison, Wis., 1921 Plot No. Treatment" Average number of berries Average weight of berries per plant Per plant Per pint 1 No. 143 217 No. 227 239 Pounds 0.42 3 L-S. -f-gelatin, 1,2. 0.62 "L— S=liquid lime-sulfur (1) delayed-dormant spray. 1-10, (2) second ap- plication of spray about one week before blooming, 1-40. One-half pound of gelatin was added to each 100 gallons of spray. These data show thai the sprayed plants on which the disease had been controlled satisfactorily lor two consecutive years produced about 4 Research Bulletin 59 32 per cent more fruit by weight than abundantly infected plants that had never received any treatment for the control of the disease. The loss caused by the disease was very noticeable during the season of 1921, due to the fact that the disease was very severe and was followed by a long dry period prior to and during harvest. The average number of berries per pint was slightly higher on the sprayed plants than on the unsprayed since the smaller berries of the latter part of the season on the unsprayed plants dried up and were not harvested, while all the small ones ripened on the sprayed plants. For a comparison of the control of the disease obtained with various treatments during the season of 1921 see Table VI. Photographs taken at the time of harvest in 1921 show the condition of fruiting branches which had been sprayed as compared with that of unsprayed fruiting branches seriously injured by anthracnose (Plate VIII). THE CAUSAL ORGANISM Cultural Studies Considerable difificulty has been experienced by investigators in ob- taining pure cultures of the anthracnose organism. Stoneman (1898) states that the fungus "....does not adapt itself readily to artificial cul- ture." The growth of the organism is so slow that contamination is likely to occur, but it is possible to obtain pure cultures by placing frag- ments of diseased tissue in poured agar plates. The exterior of the tissue from which the isolation is to be made is best sterilized by being dipped into 100 per cent alcohol and flamed. Fragments of tissue may then be removed from below the surface with a sterilized scalpel. On a 15 per cent dextrose-potato agar a straw-colored growth may be detected with a hand lens at the side of the fragment of tissues in five to seven days and may be transferred to an agar slant. The easiest method of isolation, however, is to place on the inside of the lid of a Petri dish fragments of a cane lesion bearing ascocarps. At Madison, Wisconsin, it has been possible to obtain mature ascocarps in the field from early March through June. If the cane tissue is moistened with water the spores are shot onto the agar in the lower part of the Petri dish. Germination of single ascospores may be watched and the resultant growth transferred to an agar slant in a test tube by the method described by Keitt (1915). The growth after 14 days on a 15 per cent dextrose-potato agar is light russet-vinaceous to maroon, with a reddish discoloration of the medium. There is very little aerial mycelium produced in the young cultures. The colonies are formed by a piling up of cells that have a glistening appearance. As the cultures become older, however, fine aerial hyphae are formed over the compact growth. In cultures that are a month old these hyphae give a white, downy appearance to the maroon mass of cells underneath. Conidia are seldom produced in culture. However, a sudden increase in humidity usually stimulates their production. The writer has been able to produce conidia by transferring culture fragments from dextrose-potato agar to the side of sterile sweet Anthracnose of Cane Fruits clover steins in tubes, the stems resting on a small amount of absorbent cotton and in an abundance of water. After three flays this culture may be removed to a tube of stcrik' distilled water, in which the conidia drop nff readily. In order to obtain an abundance of conidia for inoculation work it was found ad- visable to pour a spore suspension on poured agar plates. After ten days large pieces of agar bearing the fungus may be transferred to a sterile glass slide in a sterile moist chamber, consisting of a Petri dish lined with moistened filter paper. After three days the cultures may be removed to sterile distilled water and the conidia shaken from the fungal growth. Fig. 1. Camera lucida drawing of germinating coni- dia of P. veneta after 16 hours in sterile distilled water at 24 C. A study has been made of the relation of temperature to the growth of the organism on de.xtrose-potato agar. The most rapid growth was obtained at 22° to 26° C. while no growth occurred at 10° or at 32° C. Plate III, A shows the growth that was made in seven days at constant temperatures ranging from 11° to 32° C. The platings were made from a suspension of conidia in water, one loop of the sus- pension being removed to the center of each Petri dish, into which a 5 per cent de.xtrose-potato agar had been poured. Germination Studies Conidia germinate readily in sterile distilled water or on dextrose- potato agar or "water agar" (2 per cent agar in w^ater). In sterile distilled water on slides in moist chambers at 24° C. the conidia be- come twice their original size in 12 to 24 hours and a few may become one-septate, or produce short germ tubes (Fig. 1). During the ne.xt 24 hours elongation occurs and three or four septa and possibly a small amount of branching may be observed. Conidia are budded off from any of the cells, most abundantly, however, from those at the ends (Fig. 2). Further growth takes place with profuse branching and piling up of cells, forming a stromatic mass about 50 microns in diameter after 96 hours. From this mass of cells filaments radiate for 25 to 50 microns. There is seldom any further growth in sterile water. The germination on dextrose-potato agar is somewhat similar ex- cept that true germ tubes and conidia are produced after 44 hours seldom produced and that there is a greater in sterile distilled water tendency towards the massing of cells. After at 24 C. and the pro- 96 hours at 24° C. the colonies on this medium duction of secondary- have an average diameter of about 200 microns, conidia. Fig. 2. drawing conidia showing Camera lucidia of germinated of P. veneta the growth Research Bulletin 59 Experiments on conidial germination have been conducted at con- trolled temperatures in sterile distilled water and on dextrose-potato agar and "water agar." Six series have been run on dextrose-potato agar, five series in sterile distilled water, and two series on "water agar," at the following temperatures: 4, 8. 11, 15, 17, 19, 22. 24. 26, 30, 2)2, and 34° C. No germination has been observed on any medium at temperatures below 11' C. and only slow germination with slight growth at 15° C. The optimum for germination and growth was found to lie between 22^ and 26 ~ C. Germination takes place readily at 30° while no germination has been observed at either 32° or 34° C. The ascospores germinate readily in sterile distilled water and on dextrose-potato agar. In sterile water the spore becomes slightly swollen and from five to seven conidia are usually budded off within 16 hours. These conidia have not been observed to germinate in sterile water. On dextrose-potato agar the five to seven conidia are budded off and produce germ tubes 20 to 30 microns long within 24 hours (Fig. 3). Within 72 hours the germ tubes become branched and pro- duce masses of cells, making a colony about 50 microns in diameter with numerous strands of branching mycelium growing for a distance of about 35 microns from the outer edge of the stromatic mass. Modes of germination of the spores of this organism are very variable, depending (in such factors as temperature and media. The writer projects making this prt)blem the subject of a future publication. Pathogenicity Lawrence (1910) inoculated fruit of blackberries with conidia from leaves and canes and reports obtaining typical lesions after an incubation period of 15 to 48 hours. Burkholder (1917) inoculated young shoots with a water suspension of conidia from lesions on canes and from pure cultures. He obtained infection in 18 out of 56 inoculation trials, with an inoculation period of three to seven days. The writer made inoculation experiments on Cumberland raspberry plants at Sturgeon Bay, Wisconsin. The apical foot of each young cane was placed in a bag made of partially water-proofed, translucent "glassine" paper for seven days immediately preceding the inoculation, in order to preclude possibility of natural infection. At the time of inoculation the bags were removed and the young canes atomized with sterile distilled water (controls) or a water suspension of conidia from pure cultures of a single ascosporc isolation from a black raspberry cai'e lesion. The Camera drawing of germinating ascospores of P. 7'CHcta showing production and germination of secondary conidia. Antiiracxose of Cane Fruits 7 bags were replaced and the canes were kept moist by hanging inside of the bags Erlenmeyer flasks of water from which cheese cloth wicks were wound around the canes. The bags and wicks were removed five . days after the inoculations were made, at which time no disease »vas apparent on the young bagged parts. Observations were made every day after the wicks were removed and the number of resuhant lesions re- corded. The results of these inoculations (Table II) show an incuba- tion period of six to nine days. Table II. — Results of Inoculations made with P. venctu on Cumberland Raspberry Canes at Sturgeon Bay, Wis., June 6, 1921 Inocu- lation No. Inoculum Number of lesions o staled date bserved on s June 12 June 13 June 14 June 15 June 16 lA Control 0 0 0 0 12 0 0 0 8 0 0 17 2 0 0 8 8 0 17 2 0 0 8 16 0 17 2 0 0 IB Spore suspension 8 ■> 16 3 Control 0 4 Spore suspension 17 5 Spore suspension o 6 0 Black and red raspberry plants have been grown in the greenhouse in early spring and attempts to inoculate them have met with little success. Eight series of such inoculations have been made with coni- dial suspensions from cultures obtained through single ascospore iso- lations. Only one of these series gave positive results. Life history studies made during the seasons of 1^21 and 1922 indicat--^ that only the young growing canes are susceptible to the disease (Tables V and VII). The plants which were grown in the greenhouse during winter and early spring did not produce a succulent type of growth, which probably accounts for the lack of positive results from inoculation ex- periments with these plants. LIFE HISTORY Seasonal Development of Disease The disease first appears on the young growing canes and leaves in early spring, usually when the canes are eight to ten inches high. At Madison, Wisconsin, the first lesions have been observed on the following dates during the four years of observation: May 20, 1919; May 13, 1920; May 15, 1921; and May 20, 1922. The lesions on the canes, leaves, laterals and fruiting branches continue to increase in number on the young growing tissue throughout early summer. There appears to be little or no increase in disease after the middle of July as is shown in data obtained during the seasons of 1921 and 1922 (Plates V and VI). Production of Spores The immature ascocarps are first observed during the latter part of August. Some of the ascocarps are mature at Madison, Wisconsin, as 8 Research Bulletin 59 early as March 1, as the writer has been able to cause the discharge of mature ascospores from freshly collected cane lesions at this time of • the year. The asci and ascospores, however, continue to mature through the spring and early summer. Conidia are produced during the spring on the old cane lesions and abundant production of conidia follows the development of lesions on the new growth during spring and summer. On the fruiting canes the fungus probably dies after the production of conidia in the spring, as the writer has been unable to obtain conidia or make cultures from the lesions on these canes through the summer. Source of Inoculum in Nature The primary sources of natural inoculum are the ascospores, which are ejected forcibly from the asci, and the conidia from the overwintered lesions on the canes. The ascospores continue to be a source of inoculum through the spring and early summer. Burkholder (1917) reports that the ascocarps are very rare and probably do not play an important part in the dissemination of the disease. The writer has, however, observed an abundance of ascocarps (Plate III, B) on black and red raspberries in Wisconsin and considers the ascigerous stage an important factor in the overwintering of the disease and its spread in the spring under Wisconsin climatic conditions. The conidia produced in the lesions (Plate III, C) on the current year's growth form the source of secondary infection through the spring and summer. Experiments have been conducted at Packwaukee, Wisconsin, in order to obtain more definite information relative to the spread of the dis- ease. Three rows, each 250 feet long, of Cumberland raspberry plants were set out April 15, 1920. The rows were 12 feet apart and the plants were spaced five feet apart in the rows. The planting was one-half mile from any other raspberry planting, on land where grain had been grown for 15 years. A. careful survey of the surrounding country showed no wild hosts of the disease within one-half mile of the new planting. These plants were obtained from layered tips that were removed from the vicinity of the old plants one month before the appearance of the disease in the spring, and before the new shoots had appeared above the ground. Care was taken to remove all of the old cane stubs from the new plants, in order to avoid carrying any source of inoculum to the new planting. All of the soil was removed from the young plants by washing, after which they were dipped into mercuric chloride solu- tion, 1-1000, and then rinsed before they were planted. On April 14, 1921, one year after the planting was made, these plants were entirely free from anthracnose lesions. Observations made April 13, 1922, two years after the planting was made, showed an abundance of disease on these plants. Of the 125 plants that were living, 99 were diseased, 13 of them being severely in- fected. The remaining 26 plants, which were not infected with anthrac- nose, were scattered among the diseased ones. The ascigerous stage of the fungus had been found abundantly on the diseased canes in the old planting, one-half mile from the ex- Anthracnose of Cane Fruits 9 pcrimental planting. Ascospores carried by winds that blew over the old planting toward the new one were probably the source of infection for the new planting. Conidia are chiefly water borne, as emphasized by Burkholder (1917). The writer has endeavored to blow the conidia free from the coni- diophores with air from an aspirator, but with little success. However, the conidia drop off readily when the stromatic mass is placed in water. Consequently, when the fungus mass was atomized with water an abundance of spores was washed off. CONTROL MEASURES Sanitation Most writers have emphasized the importance of keeping the planta- tion free from badly diseased canes. Longyear (1904), Jackson (1913), Cook (1918) and Swartwout (1921) recommend cutting out all old canes and the most severely diseased new ones soon after harvest. This is a good cultural practice but it is of little value in checking the dis- ease during the current season, since the writer's observations show that little or no infection takes place after harvest. When thinning out the canes in the spring, it is advisable to prune out the more severely diseased ones, thereby reducing the source of early inoculum. Good cultural practices during the growing season are advisable in order to remove weeds from the rows. Weeds and compact growth of canes interfere with air drainage, and facilitate the collection of mois- ture, which is favorable to the spread of the disease. Six to twelve inches of the old canes are left on black raspberry nursery stock by nurserymen to facilitate handling. The disease is often abundant on these old cane stubs, and is therefore disseminated to the new plantings. Before nursery stock is planted these old canes should be carefully removed. Young plants obtained from the vicinity of old plants in the spring should be removed to the new planting before they are four to six inches high, since infection of the young plants usually occurs soon after they have attained this much growth. There is little possibility that the fungus lives over on fragments of plants on the ground. Observations by Burkholder (1917) and by the writer show that on the old fruiting canes the organism probably dies after the conidia have been produced in the spring. Therefore, it would appear that new plantings would not necessarily have to be made on land formerly free from the disease. However, it would be advisable to make plantings on new soil, because of the prevalence of the crown gall organism in soil previously used for cane fruit culture. Spraying Spraying for the control of anthracnose has been recommended by many writers, but most of the numerous attempts to control the dis- ease in this manner have given questionable or conflicting results. 10 Research Bulletin 59 Burkholder (1917) reviews the earlier literature on spraying and re- ports that a dormant application of lime-sulfur, 1-8, proved to be of no benefit in the control of raspberry anthracnose. After considerable experimental work he states that : "More data relating to the effect of diseased canes on the yield of fruit are needed, and until they are obtained no conclusive proofs can be furnished that spraying to combat the anthracnose of raspberry is a profitable practice." Button (1918) reports control of the disease from three applications of lime-sulfur before the blooming period, and further reports that in 1915 one dormant spray of lime-sulfur, 1-20, gave good control. There is considerable controversy as to the possibility of spray injury frorn the use of lime-sulfur and Bordeaux mixture. Most writers agree that raspberry plants are very susceptible to spray injury. With- out doubt, a considerable portion of the difference in the amount of injury reported as occurring in spraying experiments has been due to the fact that in their reports most workers have not differentiated between red raspberries, black raspberries, purple-cane raspberries and blackberries. There is certainly a difference in susceptibility to spray injury among these different kinds of cane fruits. Goff (1891) experimented with ammoniacal copper carbonate, and with mixtures of ammoniacal copper carbonate and copper sulfate. These had an injurious effect on the foliage of Cuthbert, Tyler and Gregg varieties, and Bordeaux mixture, 4-6-50, caused great injury to the foliage. He concludes that the foliage of the raspberry is very delicate and can not endure applications of a corrosive nature, and that the foliage of the blackberry, though more resistant than that of the rasp- berry, is more susceptible to injury than that of the apple. On black raspberries the foliage of old canes and fruiting branches is more susceptible to injury than that of young shoots, and injury is likely to occur in case either Bordeaux mixture or lime-sulfur is applied to the plants in hot, dry weather. From observations made in Wisconsin, foliage injury is to be expected if lime-sulfur or Bordeaux mixture is ap- plied to the plants after blooming. The writer has not observed injury to black raspberry plants from summer strength of Bordeaux mixture or lime-sulfur applications before the blooming period of the plants. The dormant strength of these sprays, applied to the plants after the leaf buds on the old canes had opened in the early spring and only a few leaves had unfolded (Plate IV, A), occasioned no material injury to the plants in the experiments conjiucted by the writer. In view of the conflicting evidence that has been presented regarding the effectiveness of spraying it was deemed advisable to carry on com- prehensive spraying trials. Preliminary reports of the results of these investigations have been made by the writer (1922 and 1923). A sum- mary of experimental treatments during the seasons of 1920, 1921 and 1922 appears in Table III. Antiiracnose ov Cane Fruits 11 Table III.— Summary of the Treatment of Experimental Plots of Cumberland Rasp- berry FOR the Control of Anthracnose, H. Fischer Planting, Mauison, Wis. Plot No. 1 2 3 4 5 5A 6 6A 7 8 9 9A 10 11 12 13 14 14A 15 15A 16 16A 17 17A 17B 18 18A 18B 19 20 21 21A 22 22A 23 23A 24 24 A 25 25A 26 28 30 31 32 33 34 35 36 37 38 39 No. plants treated 12 24 24 24 18 6 24 24 24 12 12 12 12 12 24 24 12 12 20 20 24 24 18 24 6 24 12 12 24 24 12 12 12 12 12 12 12 12 48 20 20 24 24 24 24 24 24 24 24 Treatment of plots in stated years" 1920 Unsprayed L-S. + glue, 1, 2, L-S. + gelatin, 1, 2 L-S. + gelatin, 1 L-S. + glue, 1 L-S. + glue, 1 L-S., 1 L-S., 1 B.M. + eal-cas., B.M. + cal-cas., B.M. + milk, 1 B.M. + milk, 1 B.M. + g?latin, 1 B.M.+gelatin, 1 B.M. + glue, 1 B.M.+glue, 1, 2 B.M., 1 B.M., 1 1 1, 2 1921 Unsprayed L-S. + glue, 1, 2 L-S. + gelatin, 1, 2 L-S.+gelatin, 1 l^S.+glue, 1 L-S.+glue, 1 L-S., 1 L-S., 1, 2 B.M. + cal-cas., 1 B.M. + cal-cas., 1, 2 B.M. + milk, 1 B.M. + milk, 1, 2 B.M. + gelatin, 1 B.M. + gelatin, 1, 2 B.M. + glue, 1 B.M. + glue, 1, 2 B.M., 1 B.M., 1, 2 L-S., 1, 2 I^S., 1 L-S. + glue, 1, 2 I^S. + glue, 1 L-S.+gelatin, 1, 2 L-S. + gelatin, 1 L-S.+gelatin, 1, 2 L-S. +saponin, 1, 2 L-S.+saponin, 1 L-S. +saponin, 2 L-S.+gelatin, 2 B.M.+gelatin, 2 B.M., 1, 2 B.M., 1 B.M.+glue, 1, 2 B.M. + glue, 1 B.M.+gelatin, 1, 2 B.M.+gelatin, 1 B.M. + cal-cas., 1, 2 B.M. + cal-cas., 1 Scalecide, 1; B.M.+gela- tin, 2 Scalecide, 1 Scalecide, tin, 2 Scalecide, tin, 2 Unsprayed 1; B.M.+gela- 1; L-S.+gela- 1922 Unsprayed L-S. + glue, 1, 2 L-S.+gelatin, 1, 2 L-S.+gelatin, 1 L-S.+glue, 1 Unsprayed L-S., 1 L-S., 1, 2 B.M.+cal-cas., 1 B.M. + cal-cas., 1, 2 B.M. + milk, 1 B.M. + milk, 1, 2 B.M. + gelatin, 1 B.M.+gelatin, 1, 2 B.M.+glue, 1 B.M.+glue, 1, 2 B.M., 1 B.M., 1, 2 L-S., 1, 2 L-S., 1 I^S.+glue, 1, 2 L-S. + glue, 1 L-S.+gelatin, 1, 2 L-S. + gelatin, 1 Unsprayed L-S.+saponin, 1, 2 L-S. + saponin, 1 L-S. + saponin, 2 L-S. + gelatin, 2 B.M. + gelatin, 2 B.M., 1, 2 B.M., 1 B.M. + glue, 1, 2 B.M. + glue, 1 B.M. + gelatin, 1, 2 B.M. + gelatin, 1 B.M. + cal-cas., 1, 2 B.M.+cal-cas., 1 I^S.+gelatin, 1,2,3 L— S.+gelatin, 1, 2, S^ B.M. + cal-cas., 1,2,3 B.M.+cal-cas., 1,2,3« Unsprayed L-S., 1 L-S., 1, 2 L-S. + glue, 1 L-S.+glue, 1, 2 L-S. + gelatin, 1 L-S. + gelatin, 1, 2 L-S.+saponin, 1 L-S.+saponin, 1, 2 L-S. + cal-cas., 1 L-S. + cal-cas., 1, 2 *L-S. = liquid lime-sulfur. B.M. = Bordeaux mixture. Cal-cas. = calcium caseinate spreader. Spray applications designated as: 1 =delayed-dormant, using lime-sulfur, 1-10, or Bordeaux mixture, 6-6-50; 2 = application about one week before blooming period, using lime-sulfur, 1-40, or Bordeaux mixture, 3-3-50: 3 = application one week after blooming, using lime- sulfur, 1-40, or Bordeaux mixture, 3-3-50, except as noted in footnotes following. For discussion of spreaders see page 13. *> Lime-sulfur, 1-80, plus gelatin was used in application 3. " Bordeaux mixture, 1 ^o-l 'a-SO, plus calcium caseinate was used in application 3. 12 Research Bulletin 59 Preparation of Sprays Lime-sulfur. Commercial liquid lime-sulfur testing 33° Baume was used in the experiments. The required amount of liciuid was added to the water to make the strengths outlined in the summary of treatment (Table III). Bordeaux mixture. Pound to gallon "stock solutions" of lime and copper sulfate were prepared. To make the Bordeaux mixture of the 6-6-50 formula, six gallons of the lime "solution" was diluted to 25 gallons, and six gallons of the copper sulfate solution was diluted to 25 gallons after which the two were mixed with agitation. Bordeaux mixtures of other formulae were made in a corresponding manner. Spreaders 'with Lime-Sulfur Gelatin. One-half pound of white gelatin was used to each 100 gal- lons of spray. The gelatin was placed in solution with a small amount of water aided by slight heating. This solution was added to the diluted spray mixture and agitated. Glue. One pound of finely ground high grade glue was added to each 100 gallons of spray. The glue was placed in solution and added to the diluted spray mixture in the same manner as the gelatin. Saponin. One ounce of soap tree bark was placed in one quart of water and boiled for 15 minutes. The liquid was strained and used at the rate of eight ounces to ten gallons of spray mixture. Calcium caseinate. A proprietary preparation of casein and lime was used at the rate of one pound to each 100 gallons of diluted spray. The powdered material was added to the diluted spray mixture slowly, with agitation. Spreaders with Bordeaux Mixture Gelatin. Added as outlined above. Glue. Addfd as outlined above. Calcium caseinate. During the season of 1922 the proprietary prepara- tion was used as outlined above. During the seasons of 1920 and 1921 this material was made as follows : 200 grams of powdered casein was mixed thoroughly with 480 grams of hydrated lime. The dry mixture was added to the spray, slowly and with agitation, at the rate of 150 grams to 25 gallons of the diluted spray mixture. Milk. Milk was added to the diluted spray mixture at the rate of two gallons to 100 gallons of the spray, as recommended by Lecomte (1913). Condition of Plots The experimental plots were located in the H. Fischer planting near Madison, Wisconsin. In 1920 four rows of 78 plants each were selected in a four-year-old Cumberland raspberry planting. The plants were four feet apart in the rows and the rows five feet apart. Plots 1 to 14, with the number of plants shown in Table III, were arranged consecutively in Anthracnose of Cane Fruits 13 the four rows. During the seasons of 1921 and 1922 an adjacent plant- ing of Cumberland raspberries was selected for additional plots, the planting being four years old in 1921 (Plate IV, B). The plots were square or rectangular and contained the number of plants shown in Table III. Previous to 1920 no spraying had been done for the control of the disease in the H. Fischer planting, which was heavily infected with anthracnose (Plate I, A). EXPERIMENTS IN 1920 Treatment A summary of treatment appears in Table III, and supplementary data follow. The delayed-dormant spray was applied on April 26 to plots 2, 3, 4 and 6, but a heavy rain washed off most of the spray before it could dry and made it necessary to discontinue the work On April 29, a partly cloudy day, all plots were sprayed, including the ones that had been sprayed on April 26. A "Perfection" hand sprayer was used. Since there was no foliage on the canes at this time it was easy to cover them thoroughly with a low pressure. An average of one-half pint of spray per plant was used. The buds on the old canes were showing from one-quarter to one-half inch of green tissue with no leaves unfolded. The second application of spray was made on May 26, a bright, clear day. A double-action pressure pump with a barrel attachment was used, and a pressure of 150 to 200 pounds was maintained on a single disc nozzle. An average of three-quarters of a pint of spray per plant was used. The plants were grown in the hill system and tied to stakes. The foliage was so dense that it was hard to cover the old canes thoroughly. Buds were forming on the fruiting branches and the new shoots were 12 to 15 inches high. Results Observations made May 4 showed no apparent injury from the dormant strength spray. Observations made May 26 showed very little infection on the plots. Primary infection occurred during the rain of May 10, appear- ing as lesions on May 13, although the infection was very light at this time. In order to obtain comparative data on the effectiveness of the different spray applications, a count of the number of lesions on each of 20 canes and 20 fruiting branches pet plot, chosen at random, was made on June 17 and again on July 13. A summary of this count work appears in Table IV. As the plants were heavily pruned in the latter part of July, it was impossible to obtain further data. However, very little infection occurred after the last counts had been made. The results of the counts made on the canes and fruiting branches are discussed in accordance with the objects of the experiments. 14 Research Bulletin 59 OO 005 b CO OS z" H O z 2 o < M o (N N — i> — I o t- w ^ M o> CO m CO 00CO00O5r-l-^C-N>O ooioNm-HC^(MTHcort O«000tCC0M(N-^-^-« CO.MC 00 to N 00 •S -S 3 -? bi] bfl bo bo + + + + " M C« W W W Jl Jl J Jl J) + + + + + s S S S S CO n PQ PQ n + + 09 m o o .-1 (N CO 00 a> o --I M CO c . 9 i' a c bo >> c a •5 I E ^ •— "— o M 01 " < W CQ Plate I A. — Anthraciiose lesions on two-year-old canes of Cuml.erland raspberry from the H. Fischer planting, Madison. Wisconsin, showing the severity of the disease when the control experiments were begun in 1920. B. — Longitudinal cracking of Cumher'and canes following severe anthracnose infection. C. — Anthracnose lesions on a Plum Farmer raspberry cane. Plate II 192l"'" Thr^'i '■^'?''^''''"-^- '""■^'■»« bi-anches from unspravcd plants Tulv 7 caused r^chir ^'s, ';;rf:::;[:^^"°^^ ^^^-- - id.„cj:;;^'£L^ 32** A Plate III A.— Seven-day-old growth of P. vciwta on dextrose-potato agar (x .65). Cultures made from a conidial suspension and held at various constant temperatures (Centigrade). B.— Photomicrograph of a cross section of an ascocarp of F. veneta from a cane lesion, showing one globular ascus with ascospores. C— Photomicrograph of a cross section of a lesion on a young cane, showing collapsed host tissue and the production of conidia. m to* »tr^A«'V- .( .a.^y.ff-' :ay< /*', s ^ 1 '' 8 2- 1 1 5 z "< _& 0 -z - p ! — 1 *> y 1 ; 1 0 i '•' ,,_ 1 ; ; -> 1 ; ^ 5J 9- f or ' ; y o f' / ^ i 0 >.. 1 lU : ^ ^ L »... \ of -u. 1 -5V "^~ — ^ ■; \ 1 ¥ ^= N /I - — ^ -J^ ; — ,i._ j> ~^ 1 f' ', i ' — ^ Oi 2 \ '•- ^ ^ «< — — s, . 1 ,X • 1 2 — ■ '• \ Off c ^^ — _^ J 1 ^ <• ■• ;\ — 1 ^ — ^ , — — — — — — \ — — 1 -Ac 1 i^ — — _J r- — — — — -r — — — — ' __i < — N i «- \ ■^ '• >, > / " 5 1 . s ^ ti. i c ^ % tf) S5 ^ M CM ^ •^.-^ ^ < '"■|-^ 2f| "^ Oi 5 ? 1- rfi X «»t V '^ Oo \ «« 5 t 0.^ t; "-,, - -1 > ■<' — "1 "SI 1 (00 90 flO TEMPERATURE 70 IN DEGREES FAHRENHEIT 60 50 40 30 20 « S S 2 0 'la 111 Ui II. - U Z^ :^<5 3 UJZ uZ< > < C£ •o "O c>i c»* ~" " 510 2 lO r: >: u- 0 < ii u z oc < 0: 10 0 10 0 tr « e« — — Z?o < ? 0- < 1 > - 0) ^ ' (S 0 s^ a. 0 ^-^ ji .p < ^!Sc35:J^^ss2o*«*«•.-oSsS!22•^o | \ • \ 1 0> — V. ■ 1 o> ^ .' J f f' R ^ « 5 3 1 = "^ > 1 ^ - 5 o -. / o 9 s / / 1 s s ? 1 s 5 SJ ^ M o «l \ s O ■o / 1 ,' o. / j >■ c *» ifcl >. ( 1 — 1 \,^ 2"* ► J \^ i«t 1 ) **3 2 ^ in s r 1 ( <0 ;? z> S ^ X — 1 S! 1 * 1 y 2 f CM s 2 J u - < -8 — o: y^ '- ^^ ' *v^ \ o ^ ' %.*' ^ OO 1 '.''"L , ^ 4 '^ < ^% \ vf — 1 o <0 -% V. ' K \ ;^\ / K, o cc \ Q. ^ N li \ '^ '■> \ "■i- s /J O < z ■ Z o IE ^*^ f csi ^ '. 11 / *>^ N 2 ' — --^, ; N / >• n i; iftin < ^ \ ■ Y tr> / t \ < <^ r^ ■■"' I o- o> ^-^ r^jrH — 1 — o~ — — — — — a N ^r n — ' — 1 — — ' — — ^ -K to ^ - "*~> r ,!" 1 J \ '^'-^ 1 OUJ \ ■^5 0 K <0 »- -O « ^« V K» N - 1 ^^ ^ f^ ^ ^ ■ 1 UJ r> yj ^ 1 KuiiJ ii^ >i;:^o fs ^ < bj l?5 OCOZ -oi i2< -S UJ < < z 1 lU fc t/1 •c K <-.-» c r^ u O P Plate VII A— A voung Cumberland raspberrv cane which received two thorough and timely applications of lime-sulfur spray with gelatni as a spreader. H. Fischer planting (plot 3). Photographed July 7, 1921. B.—A young unspraved cane irom the same planting (plot 1). showing the extent of the disease on the canes of unsprayed plants at harvest tmie. Photographed July 7. 1921. '— 'y. = b.-5-n ~ ■Z Z tl. b£j3i£ .r. '^ 9. 2 o £"5 < JSCQ bfi Anthracnose of Caxe Fruits 15 The effectiveness of lime-sulfur as compared with Bordeaux mixture. In gc'iu'ral, liiiic-sulfur was more ctTrctivc in coiitrolliiiK tlic disease than was Bordeaux mixture. The effectiveness of a delayed-dormant spray only. Lime-sulfur with gelatin as a spreader gave excellent control of the disease, and lime- sulfur without a spreader controlled the disease commercially. Lime- sulfur in combination with glue as a spreader failed to check the dis- ea.se in the latter part of the season. Bordeaux mixture with glue gave connnercial control of the disease, while poor control w-as obtained from the use of Bordeaux mixture alone or in combination with gelatin, calcium caseinate or milk. The effectiveness of a delayed-dormant spray followed by a second application about one week before blooming. Lime-sulfur with gelatin or glue as a spreader gave very good control of the disease. Bordeaux mixture with gelatin, glue or calcium caseinate as a spreader may be classed as having given commercial control. The effectiveness of adding spreaders to the above sprays. The ad- dition of gelatin to lime-sulfur distinctly increased the effectiveness of the spray. Glue in combination with lime-sulfur, when two applica- tions of sprajf were made, increased the etTectiveness of the spray nearly as much as did the gelatin. The addition of glue gelatin, milk or calcium caseinate to Bordeaux mixture increased its effectiveness in controlling the disease, glue and gelatin being the more efificient. EXPERIMENTS IN 1921 Seasonal Development of Host The first exposure of green tissue in the leaf buds was observed April 8, and the warm period from April 11 to 13 caused the buds to develop until an average of one small leaf was unfolded and one-half inch of green tissue was showing. The snow and cold weather of April 15 and 16 checked the growth. The first new shoots appeared above the ground about April 28. In order that the seasonal development of the host might be followed, 20 canes and 20 fruiting branches on the plants in the unsprayed plot were tagged on May 15. The length of these canes and fruiting branches was recorded at intervals of two to five days until no further increase was noted. These data are recorded graphically in Plate V in relation to the development of other factors important in the control of the disease.* From a study of Plate V it will be seen that most rapid growth of the host occurred between May 17 and June 4, while the fruiting branches had practically ceased growing by June 2. The young canes continued to elongate until about July 9, when they averaged about 35 inches in length. ' The climatological data are from the recoi-ds of the Madison station of the United States Weather Bureau ( riiniatoloaical data. U. S. Dent. .\ar. Weather Bur. Wis. Section 26: 15-32. 1921). The H. Fischer raspberry planting is located five miles east of the Weather Bureau station, and undoubtedly the climatological conditions vary somewhat from those recorded at the Weather Bureau station. 16 Research Bulletin 59 Seasonal Development of Disease The seasonal development of the disease was followed on the above noted 20 canes and 20 fruiting branches, the increase in number of lesions being recorded at intervals hi two to five days throughout the period of increase in infection. Additional records were made on August 17 and September 23 to ascertain whether any infection had taken place late in the season. The data were averaged and are recorded graphically in Plate V. Supplementary data as to the development of the disease are to be found in Table V. From a study of Plate V it will be seen that the first lesions appeared on May 16, infection having taken place during the rains of May 10 to 13. The greatest development of the disease occurred during the last half of May, when the raspberry plants were making their most rapid growth. There was a continued increase in the amount of disease until July 20, and the more important infection periods may be traced to preceding rains, allowing from two to seven days for incubation. The disease de- Table V. — Average Increase in Number of Anthracnose Lesions on Canes and Fruiting Branches of Unsprayed Cumberland Raspberry Plants, H. Fischer Planting, Madison, Wis., 1921. On canes by feet " Dates On fruiting observed branches^ lst= 2nd 3rd 4th Total foot foot foot foot No. No. No. No. No. No. May 1.5 0.0 0.0 0.0 0.0 0.0 0.0 17 1.5.8 0.0 0.0 1 .1 16.9 6.6 19 9.7 0.0 0.0 1.0 10.7 4.3 23 2 . 3 0.0 0.0 0.3 2.6 5.1 26 33.0 12.4 0.0 0.4 45.8 11 .8 30 11.3 110.0 1.0 0.1 122.4 13.5 June 2 8.4 42.2 3.7 0.1 54 . 4 9.0 7 5.8 31 .5 15.7 0.0 53.3 8.7 10 3.1 17.2 6.4 0.0 26 . 7 8.1 14 0.6 9.7 9.7 0.0 20.0 8.4 1« 0.2 5.1 29.4 0.0 34.7 4,1 22 0.0 2.4 16.0 0.0 18.4 2.6 2.'j 0.0 1.0 10.0 0.0 11.0 1.2 29 0.0 0.3 4.4 0.0 4.7 1 .1 July 1 0.0 0.0 1.5 0.0 1 .5 0.0 .5 0.0 0.0 0.1 0.0 0.1 0.0 9 0.0 0.0 0.5 0.0 0.5 0.0 11 0.0 0.0 0.6 12.2 12.8 0.0 M 0.0 0.0 0.0 7.9 7.9 0.0 16 0.0 0.0 0.0 5.4 5.4 0.0 19 0.0 0.0 0.0 4.3 4.3 0.0 21 0.0 0.0 0.0 0.0 0.0 0.0 Aug. 17 0.0 0.0 0.0 0.0 0.0 0.0 Sept. 23 0.0 0.0 0.0 0.0 0.0 0.0 "Average increase in number of lesions on twenty canes tagged on May of lesions on twenty fruiting branches 15. 'Average increase in number tagged on May 15. 'Basal foot of cane. veloped very abundantly during the season. The lack of disease de- velopment after July 20 seems to have been due, primarily, to the cessation of host development. This is further shown in Table V where in- creases in the nuinbcr of lesions on the young growing portions of the Anthracnose of Cane Fruits 17 canes are recorded, in contrast with the cessation of disease development on the older and hardened portions. High temperatures during the sum- mer (Plate V) may have been an important factor in checking disease development, since the writer's investigations have shown that the maximum temperature for growth of the organism on dextrose-potato agar is about 90° F. Treatment A summary of treatmeiit appears in Table III, and supplementary data follow. The delayed-dormant spray was applied on April 19, a bright day with a ten-mile easterly wind. A. barrel pump was used, and a pressure of 100 to 150 pounds was maintained on a single disc nozzle. An average of three- fourths pint of spray per plant was used. The leaf buds on the upper part of the canes had opened, averaging one leaf unfolded with two to four leaves folded but separated from the bud. The lower buds showed one-half inch of green tissue with an average of one leaf folded but separated from the bud. The second spray application was made Alay 10, a cloudy, cool day. A barrel pump outfit was used, and a pressure of 100 to 150 pounds was maintained on a single disc nozzle. An average of 1JS4 pints of spray per plant was used. The young shoots were three to four inches high, and the fruiting branches five to six inches long with the blossom buds well formed. Results Counts were made of the number of lesions on canes and fruiting branches on the various plots as in 1920, and the results appear in Table VI. Plates VII and VIII further illustrate the effectiveness of spraying for the control of this disease in 1921. The results of the counts are discussed in accordance with the objects of the experiments. Unsprayed. The disease was extremely abundant on the unsprayed plants and more so in the planting which had received no previous treatment for the control of the disease. The effectiveness of lime-sulfur as compared with Bordeaux mixture. There was little or no difiference in the effectiveness of these two spray materials. The effectiveness of a delayed-dormant spray only. On the plants which had been sprayed the previous season lime-sulfur in combina- tion with glue or gelatin as a spreader, and Bordeaux mixture with gelatin gave fair control of the disease. Lime-sulfur alone, and Bor- deaux mixture alone or in combination wMth glue, milk or calcium caseinate were not very effective in controlling the disease. On plants which had received no previous treatment for the control of the disease all spray materials failed to control the disease commercially when only the delayed-dormant application was made. Lime-sulfur with gelatin,' and Bordeaux mixture with calcium caseinate gave better control than any other spray combination in this test. 18 Research Bulletin 59 3 CX '- M C ° 6 P-2 3 P-l Ji,o M 5 a ^ m C QJ C p3 MOO > '^ w t- CO X-'-t cooooo-^«^t-'-^^^owco 00 T-l rH (M .1 COCOC>JIM»^t~XC^WC^O'^O^Tto-»n^»C OO^O^HCCt>CirHO^COOOiO»-HOiCOC CD t- r-c K CO 00 0CC5 :c a; t- t> Oi < r-^ ^ C^ t~ lO 00 00105 00 lO ^CO OiM O lO Tj< Tj* o 05 Oi 05 W CO 00 (M CO 1-1 CO i-> "i^ < M w a. . jO „ AxTHRACxosK OK Cane Fruits 19 The effectiveness of a delayed-dormant spray followed by a second application about one week before blooming. On plants which had been treated the previous season satisfactory control was obtained from the use of lime-sulfur alone or with gelatin or glue as a spreader, and from the use of Bordeaux mixture with gelatin or calcium caseinate as a spreader. Although no satisfactory control was obtained on the plants that liad not been treated the previous season, lime-sulfur with glue, and Bordeaux mixture with calcium caseinate were more effective than the other spray combinations. The effectiveness of a single spray application about one week before blooming. Lime-sulfur with saponin or gelatin as a spreader, and Bor- deaux mixture with gelatin as a spreader showed little effectiveness in controlling the disease when only the one application of spray was made, about one week before the plants were in blossom. The effectiveness of adding spreaders to the above sprays. Added effectiveness was obtained by using spreaders with the sprays during this season, which was one of extremely abundant infection. Greater benefit was obtained from the use of gelatin or glue with lime-sulfur, and from calcium caseinate with Bordeaux mixture than from any other spreader used with either of these sprays. EXPERIMENTS IN 1922 Seasonal Development of Host The seasonal development of the host was followed as in 1921, and the results are shown graphically in Plate VI in relation to the development of other factors important in the control of the disease.^ On April 18 the buds on the old canes were showing about three- quarters of an inch of green tissue, but no leaves had unfolded. The first leaves were unfolded on April 22 and the new shoots began to appear above the ground May 1. The development of foliage on the old canes on May 2 is shown in Plate IV, A. From a study of Plate VI it will be seen that the canes continued to increase in length until August 1, and that the most rapid growth occurred between May 15 and June 12. The fruiting branches had obtained their maximum length about May 27. Seasonal Development of Disease The seasonal development of the disease was followed on 20 canes and 20 fruiting branches as in 1921. The data are recorded graphically in Plate VI, and supplementary data are to be found in Table VII. From a study of Plate VI it will be seen that the disease first developed in the field on May 20 and that no increase in number of lesions was observed after August 1. The greatest development of disease occurred during the early part of June when the plants were making their most rapid growth. •The climatological data are from the records of the Madison station of the United States Weather Bureau as in 1921 (Climatological data. U. S. Dept. Agr. Weather Bur. Wis. Section 27: 17-32. 1922). 20 Research Bulletin 59 The disease continued to develop through a longer period in 1922 than in 1921, which may be correlated with the fact that the growth of the host plants continued for a longer period in the season of 1922. The fact that the temperature seldom reached 90° Fahrenheit during June and July of 1922 may have had some effect in favoring the longer period of in- fection. As in the previous season the greatest amount of disease de- veloped when the host plants were growing most rapidly. As in 1921 the older portions of the canes developed resistance to the disease while the younger portions were being infected (Table VII), which further in- dicates that the rapidly growing portions of the raspberry plant are the most susceptible to the disease and that resistance to the disease is de- veloped as the growth ceases and the plant tissues harden. Table VII. — Average Increase in Number of Anthracnose Lesions on Canes AND Fruiting Branchks of Unspraved Cumberland Raspberry Plants, H. Fischer Planting, Madison, Wis., 1922. On canes by feel a Dates On fruiting observed branches'" Ist'' 2nd 3rd 4th Total fool foot foot foot No. No. No. No. No. No. May 17 0.0 0.0 0.0 0.0 0.0 0.0 20 0..3 0.0 0.0 0.0 0.3 0.1 22 1 .0 0.0 0.0 0.0 1 .0 0.2 26 1 .0 0.0 0.0 0.0 1.0 0.1 29 2 . 5 0.0 0.0 0.0 2 . Ty 0.8 June 1 5 . 7 1 .6 0.0 0.0 7.3 2.3 4 2.0 10.3 0.3 0.0 12.6 1.0 7 0.5 11.3 1 .5 0.0 13.3 0.3 10 0.1 3.9 1.4 0.0 r).4 0.3 12 0 . 0 10.9 0.7 0.0 11.6 0.0 16 0.0 0.4 0.8 0.3 1 .5 0.0 1<) 0.0 0.9 0.0 0.0 0.9 0.0 22 0.0 0.3 0.3 0.0 0.6 0.0 24 0.0 0.0 0.0 0..^ 0.5 0.0 29 0.0 0.0 0.2 0.0 0.2 0.0 July 2 0.0 0.0 0.0 0.3 0.3 0.0 .'j 0.0 0.0 1.4 0.4 1.8 0.0 11 0.0 0.0 1.0 0.0 1.0 0.0 18 0.0 0.0 2.0 1 .8 3.8 0.0 24 0.0 0.0 1 .0 3.0 4.0 0.0 Aug. 1 0.0 0.0 0.0 1.3 1 .3 0.0 11 0.0 0.0 0.0 0.0 0.0 0.0 Oct. 7 0.0 0.0 0.0 0.0 0.0 0.0 17. 'Average increase in number of lesions on twenty canes tagged on May 'Average increase in number of lesions on twenty fruiting brancbes lagged on May 17. ■^^Basal foot of cane. Treatment A summary of treatment appears in Table III, and additional data follow. The delayed-dormant spray was applied on May 2, a cloudy day with a light easterly wind. A wheelbarrow spray outfit was used, and a pressure of ICK) to 150 pounds was maintained on a single disc nozzle. An average of one-half pint of spray per plant was used. The stage of de- velopment of the foliage on the old canes at the time when this spray was applied is shown in Plate IV, A. The new shoots were beginning to appear above ground. AXTHRACNOSE OK CaNE FrUITS 21 The second application of spray was made on May 17, a cloudy, cool day with a light breeze from the southeast. The wheelbarrow spray out- fit was used, and a pressure of 75 to 100 pounds was maintained on a single disc nozzle. An average of 1J4 pi"ts of spray per plant was used. The young canes were eight to nine inches high, and the fruiting branches seven to eight inches long with the blossom buds well formed. The third application of spray was made on June 1, at the end of the blooming period of the plants. The wheelbarrow spray outfit was used, and a pressure of 75 to 100 pounds was maintained on a single disc nozzle. An average of 1^4 pints of spray per plant was used. The new canes were 22 to 23 inches high. Results Counts were made of the number of lesions on canes and fruiting branches on the various plots, as in 1920 and 1921, a summary of which appears in Table VIII. The results of the counts are discussed in accordance with the objects of the experiments. Unsprayed. The disease was fairly abundant on the unsprayed plants, although not so abundant as in the previous season. Plants which had been sprayed in 1920 and 1921 but left unsprayed in 1922 showed con- siderable decrease in the amount of infection on them as compared with the amount of infection on plants which had been left unsprayed the three seasons (plots 5A and 1). This cumulative benefit from spraying is not evident, however, in a comparison of results from plots 17B (sprayed in 1921, unsprayed in 1922) and 28 (unsprayed the two sea- sons). The effectiveness of lime-sulfur as compared with Bordeaux mixture. In general, lime-sulfur gave slightly better control of the disease than did Bordeaux mixture. The effectiveness of a delayed-dormant spray only. Commercial con- trol of the disease was obtained from the use of lime-sulfur alone or in combination with glue, gelatin or calcium caseinate as a spreader, and from the use of Bordeaux mixture with calcium caseinate as a spreader. Lime-sulfur with saponin as a spreader, and Bordeaux mixture alone or in combination with glue, gelatin or milk failed to control the disease commercially. The effectiveness of a delayed-dormant spray followed by a second ap- plication about one week before blooming. Very satisfactory control o-f the disease was obtained from the use of each of the spray combina- tions, with little difference in their effectiveness. The effectiveness of a delayed-dormant spray followed by two appli- cations; (A) one week before the blooming period, and (B) at the end of the blooming period. Excellent control of the disease was obtained from the use of lime-sulfur with gelatin, and Bordeaux mixture with calcium caseinate, but extreme injury to the foliage resulted from appli- cation of the sprays after the blooming period. Little or no reduction 22 Research Bulletin 59 Oi £ C rj o o PhZ £ c M t; c > 3 O 3 t. t; e M c 0) c c c3 > 3 O O > 3 o 3 t- E c M c 0) C c n! > 3 o o T(iCT>t-CO uti c^ 00 »^ 00 oi 00 m(Mt> rt to ^■^ ^IM"500O5>nTllC0O5C~CJJ00 t-'XrtOOomiMTOeo-a'>o xiocoojo^maioioscoai o P3 <; o 00 -^ C^ t£ -^o^ t^ UD ^ W CO PhZ «!t-000100JOOm05(MC5 C!0'^00'H^ eoooooraO!^;o^'^ooN < < < < -H-HNoqiMoqNC ^ (Nco 00 -^ tn „ . EEC rr —"-^.S.S.S.S.S c c s rt S CO --H CO 04 CO lO f-H ;d « 00 Iff o C^] lO '^ji cc NN Si Si * * * S.a++ + + +++ +++++ . . • • • • ■ • • ■ • • • - • ij^DHiHiHiJhiHiiiJJiJjHiJjiJjJjfflmfflfflcdpammccmixcQm 3 C ii 3 3 J5 s « s ^ s & ft . 00 0) C M o o c M • ft .5 "aJ '^ S "S S .2 2 E E * '« '■" S »- £^ c ft S >. • * ■g E i ^ I p c s s X « i2 -2 > S ^ - ■ g 0) a, rt o, 5 ^ " S - a; o >- J'-S o c O O "-73 00 3 ^ '-' T3 B -^ ' „ E H M ^- 3 4J « M « O C ' M. N. W.^LKER, Plant Pathology Memoranda Research Balletin tt Aayoat, 1125 Experiments on the Control of Wildfire of Tobacco JAMES JOHNSON and HERBERT F. MURWIN Asfricultural Experiment Station of the University of Wisconsin Madison CONTENTS Introduction 1 Summary of Earlier Work 1 Overwintering Studies 3 Dissemination Studies 8 Spread of Wildfire in the Field 8 Seed Bed Infestation 10 Dusting and Spraying Experiments 11 Seed Disinfection 14 Loss of Virulence 17 The Wildfire Toxin 19 Practical Considerations 20 Summary 22> Literature Cited 35 Experiments on the Control of Wildfire of Tobacco' THK CONTROL of the wildfire disease of tobacco caused by Bacterium tabacuvt (Wolf and Foster) has been the subject of considerable investigation since the outbreak of the disease in North Carolina in 1917 (14). The outstanding observation, bearing on control, has been the fact that the disease originates in the seed bed and that practically all cases of field infections are traceable to this source. The prevention of seed bed infection is. therefore, the most logical aim of all methods of control. This naturally involves : first, the determination of how or on what materials the causal organism lives over winter or from one crop to the next ; and second, methods of pre- venting such infected material from being introduced into the seed beds. Once seed bed infection occurs and is discovered, the grower must choose between discarding the infected plant beds entirely or taking a risk in using some or all of the plants, relying on unfavorable weather conditions to prevent further serious spread of the disease. This latter method is economically hazardous, as it is likely that the disease may prove disastrous to a crop if proper weather conditions for the dis- semination and the development of the disease occur. Precautions to prevent dissemination in the field are of doubtful value as a means of control; their effectiveness is at least very limited, and probably more often they are effective onljr under relatively unfavorable weather con- ditions for the development of disease. The investigations reported in this bulletin are consequently mainly concerned with a study of the factors which may account for seed bed infection, together with methods of preventing such infection. The practical conclusions arrived at are also to a considerable extent in- fluenced by several years of observational studies made during field surveys. Summary of Earlier Work The control of tobacco wildfire has received some experimental at- tention in most of the tobacco districts in which it has occurred. While some difference of opinion exists as to the relative importance of the methods of overwintering of the causal organism, practically all in- vestigators agree that the causal organisms may survive from one crop to the next on infected and cured tobacco leaf, except that in flue- cured tobacco sufficient heat may be used to kill the organism. The subsequent dissemination of this infective material to the seed beds H;oc)pi*rative e.vperinu'iits with Office of Tolmcco liivestigiitions, Uiiicau of Plant Iiuhistiy, I iiilcd Stales Department of ,\g|-iciilture. 2 Wisconsin Research Bulletin 62 may naturally occur in several ways, the most unusual of which has been announced by Valleau and Hubbard (13) who claim that the wildfire organism is commonly transmitted through the spitting of tobacco juice into the seed beds. Wolf (15) and Fromme and Wingard (S) were first to point out the possibility of overwintering on seed and introduced the formalin and corrosive sublimate seed treatments respectively as control measures for tobacco wildfire. The importance of overwintering on seed in the Connecticut Valley has been questioned by Anderson and Chapman (1) and Clinton and McCormick (3). Similarly, overwintering in soil has been suggested by the earlier workers, but this again has been questioned by more recent observa- tions and experiments. Information concerning the possibility of overwintering of the wild- fire organism on seed bed covers (cloth and sash) and frames is espe- cially meager. The possibility has been recognized, however, and re- ported in some cases as occurring (15). Tobacco stems (leaf-midribs) both in commercial fertilizers and as untreated fertilizer material have been held responsible, by observa- tion, for some cases of overwintering. This seems least likely in the case of the manufactured fertilizers containing stems where heat treat- ment is used (15). Untreated stems and stalks, since they usually carry leaf fragments which may naturally be infected, are probable overwin- tering carriers as pointed out by Anderson and Chapman (1). Experimental evidence on the actual dissemination of the wildfire organism is small and fragmentary. Observational evidence is abun- dant but rarely convincing. Since almost any material which has been exposed so as to carry the causal organism physically may conceivably carry it from place to place, this subject is not a very fruitful one for satisfactory speculation or experimentation. It has been suggested by various workers that long distance dissem- ination may occur most often through transportation of infected seed, plants, or commercial tobaccos and by dry winds. With respect to transmission of the disease from plant to plant, in seed beds and in the fields, all investigators agree on the effectiveness of rain, especially when accompanied by strong wind. Heavy storms and hail which in- jure the leaf surface are especially favorable to subsequent heavy in- fections as well as for dissemination . The control of wildfire in the seed beds by dusting or spraying fre- quently with copper-lime dusts or Bordeaux mixture has been rec- ommended by workers in the Connecticut Valley. When properly ap- plied it is claimed to be an effective control measure. This method has not been generally adopted outside of New England and some question as to its value has already been raised in our work (8). Dust- ing and spraying in the field has received some attention by other workers (12) with negative results. Since the work reported in this paper had been practically com- pleted, Anderson (2) has published his results on overwintering of Control ok Wildi'ire of Tobacco 3 tobacco wildfire in New England. His results indicate that the bac- teria winter most successfully in situations where they are not sub- jected to keen competition from the growth of other organisms — prin- cipally in fairly dry situations — and that they winter least successfully under conditions moist enough for competing organisms to grow. He concludes the wildfire organism may overwinter on cured leaves in the barn, plants standing in the field, on boards, sash, and dry frag- ments of seed pods, but that overwintering in leaves exposed to decay or in the soil is least likely. Overwrintering Studies The overwintering experiments were designed to determine how long and under what conditions the wildfire organism is most likely to sur- vive the period during which its host plants normally can not be the source of its propagation. The main tests have been made with artifi- cially infested materials which are most likely to be concerned with overwintering and seed bed infection. These have been stored under different conditions in most instances, and tested from time to time as to their ability to yield infection when placed in contact with young tobacco plants. It has been assumed that the application of infected material to a unit area, in many cases hundreds of times greater in quantity than that which would occur under natural conditions, reduces the errors which might result from working with only a relatively small amount of material. Conditions for infection have been made as ideal as possible both by wounding the plants and by maintaining favorable environmental conditions. Considerable variation in this condition is evident, however, from the results. Tests were made soon after the materials concerned were infested and before storing away in all cases to make certain that the causal organism was pathogenic at the start of the test. The results are, therefore, believed to be reliable from the experimental standpoint. From a practical standpoint we have also tested out materials supposedly infected naturally, and made a considerable number of field observations, and these factors are also taken into consideration in drawing final conclusions. In the 1922 experiments artificial applications to seed, boards, cloth, soil, etc. were madd with both pure cultures of the organism and with the juice extracted from badly infected green leaves. Two different sets of applications were made known as Series I and Series II. These materials were divided up into separate portions each suitable for one test. It was planned to store one-half of this material out-of-doors in the winter months, but this was not done in some instances because the organisms were apparently dead on those materials most commonly out-of-doors in winter, before the winter months arrived. The cured leaf material was cured under normal conditions in the shed, and the buried leaves were, of course, outside all winter. The 1923 materials were inoculated artificially with dried crushed leaves, for the reason that this would seem from our 1^22 experiments 4 WISCONSIN Research Bulletin 62 to offer the best oppprtunit_v for the persistence of the causal organism over winter. Part of this material was stored at room temperature and part in a weather-instrument chamber out of doors where the material was protected from rainfall. Inoculations with these materials have been tried in several ways. Frequently they were made by scrubbing or washing of the materials in a small amount of water and making fifty wound inoculations on individual plants in pots with the washings. In other cases platings on agar were made from the materials and wildfire-like colonies used for inoculation. More reliable results are obtained by placing the materials directly upon young vigorously growing seedlings in "flats'" after wounding them. The flats were then well watered and kept covered with paper for one or two days, keeping the planFs and paper moist in the meantime. The infested materials were removed from the flats three or four days after the inoculation was made. Our exper- iments have led us to question any conclusions based on negative re- sults from inoculating individual plants with material in which the causal organism is not abundant and is in a latent state, even if such plants are vigorous, wounded and placed under good environmental con- ditions. Seedling inoculations in the greenhouse in which at least 100 plants are involved seems the most reliable test. Inoculations in out- of-door sections of seed beds are not apparently as reliable on account of the danger of dissemination of the organism from section to section, and less certainty in the control of the environmental conditions. Prac- tically all of our results are based on greenhouse tests. The first series of experiments were started in the midsummei of 1922, for the purpose of comparing the survival of the wildfire organism on or in seed, soil, cloth, boards, and dried, naturally infected leaves, cured naturally infected leaves, and green leaves buried about 4 inches in the soil, without direct contact with the soil and with mixtures of soil in proportions of 1 to 5 and 1 to 10. In addition the watery ex- tract from green leaves and the pure culture suspension used for in- festing the seed, soil, etc. was saved for comparative tests, as was the dried green leaf pulp from which the green leaf extract was made. An attempt has been made to present the data from inoculations made with these infested materials in condensed form. The percentages of infection given are not comparative throughout for the reason that different methods of inoculation were used in some cases and because of the variable conditions for infection which cannot be avoided. It is also to be expected that the dilution of the suspension of organisms recovered from the infested materials naturally varies greatly. Within certain limits, however, the percentages are believed significant and to these attention will be called. The principal value in the results, how- ever, lies in the outcome as to whether infection was or was not ob- tained after repeated trials. In this respect the results are believed to be significant to a high degree. Attention has already been called to the fact that the number of organisms involved in these tests are prob- ably infinitely greater than would be likely to occur under normal con- ("o.NTKOI. OK W'lI.DI'IKK ol- 'I'oMACCO 5 (litioiis, so that the small amount of material used as units (10 grams seed, 1 square foot of cloth, 16 square inches of boards, etc.) are comparable to a much larger quantity of these materials under practical conditions. From Table I it seems evident that the wildfire organism can sur- vive but a comparatively short period in liquids exposed to general contamination, and that its limit of survival on such material as moist soil and dried green-Ieaf-plup, cloth, and boards is only about one to two months, under the conditions of this experiment. On the other hand, on tobacco seed and on dried and cured leaves the organism was still alive after nine months. It is interesting to note the comparative survival on seed and cloth. There is some indication that seed tends to exercise some protective influence on the wildfire organism and that the cloth possessed some deleterious action. This suggestion is based partly on the reversal of behavior in the two cases between the results from the pure culture and the green-leaf inoculum. In subsequent attempts to prove up these relationships, however, we failed to obtain conclusive results. A second series of tests with infested materials similar to that in Series I was started in the early fall of 1922, and carried on simultan- eously throughout the fall and winter. The methods used were in all respects similar. The data secured in this series are presented in Table II. The results agree in general with those presented in Table I re- garding overwintering. Moist soil, cloth and boards inoculated with pure-cultures or extract from infected green leaves rapidly lost their in- fective powers, whereas seed retained it to a striking extent. The liquid extract from infected green leaves again became non-infectious after only a few days and the dried pulp from these same leaves re- tained infection less than one month. Tobacco stalks were infested and included in this test. We are unable to account for the fact that we were not able to get definite infection from these in three trials made before the seventh-month test, which gave good infection. Under questionable infections, however, we have been in the habit of includ- ing lesions which were so faint that they could not be readily identified as wildfire, and we are, therefore, inclined to attribute this result to conditions not being sufficiently favorable to bring out good halos with a weakened organism. It seemed logical to conclude from the first year's tests that the presence of moisture and the protective action of host tissue were im- portant factors in the overwintering of the wildfire organism, and that the most likely place of the survival of the parasite under practical conditions would be in dried or cured leaf material not permitted to decay. Since such dried leaf material might readily attach itself to the materials used in seed beds or in their construction, or be acci- dentally or otherwise transferred from buildings where it had re- mained dry, to seed beds, it was deemed advisable to repeat the whole set of experiments, using as a source of inoculum dried infected 6 Wisconsin Research Bulletin 62 leaves crushed or powdered. These were applied in the fall of 1923 by dipping the various materials into a water suspension of the infected tissue after which they were rapidly dried and stored under the desired conditions. As the previous season's experience indicated that the most reliable results could be obtained by direct inoculation to flats (about 22"xl4") containing several hundred young and vigorously growing plants, prac- tically all results for the winter of 1923-24 are based upon this method. At the close of the experiment, 100 plants were pulled at random from the flat and the total number of lesions on these counted. Flats showing no infection in the counts were carefully searched for any single lesion which might occur. In most cases the results given are averages of duplicate inoculations. A comparison of results with these materials, as shown in Table III, indicates that the wildfire organism survived most successfully on dried leaves, or dried leaves in dry soil, or on dry stalks ; apparently not so well on boards and on seed, quite poorly on cloth and in cured leaves. and was rapidly destroyed in moist soil and rotting leaves. The var- iation in amount of infection is no doubt in part due to differences in amount of inoculum applied in each case, although effort was made to insure that the inoculations would have some quantitative significance. The results justify the conclusion that the causal organism survives better on some materials than on others, under similar environmental conditions. This experiment and others have indicated that the organism sur- vives better in infected leaves rapidly dried at air temperatures than in those going through the ordinary curing process. Considerable vari- ation may be expected in this direction according to the condition of curing. Table III is of practical significance, however, in showing that dried or cured leaf tissue may harbor the wildfire organism over winter either on seed, cloth, boards, stalks, or in dry soil under conditions such as occur in buildings or wherever materials may remain dry. In moist soil or in rotting leaves the organism dies out in a comparatively short time. In order to determine the influence of different conditions of storage as to temperature and humidity upon the pathogenicity of the wildfire organism, the infected materials were stored outside all winter in an open chamber protected only from direct rainfall. The inside storage was in a relatively warm (about 75° F.) and dry room. Table IV shows a comparison of this material after several months. It may be seen from this table that practically no difference in pathogenicity occurred as a result of these differences in temperature and humidity in relatively dry materials. The influence of alternate freezing and thawing on or- ganisms in a moist condition in cultures is discussed on page 19. The possibility of the overwintering of wildfire in soil has been repeatedly raised by growers. While most workers are agreed that this j)robabIy rarely happens and is of little significance, sufficient obser- Control oi- Wildfire ol Tobacco 7 vational and experimental evidence is offered to keep the matter in doubt. The experiments here have repeatedly indicated that the wildfire bac- teria cannot survive more than a month in ordinary moist loam soil. Some difference may exist in different soils in this respect, but over- vifintering of any organisms in soil is very doubtful except as dry in- fected leaf tissue is lodged in dry soil, and does not become intimately associated w^ith it at any time for even short periods. This condition may occur in tobacco sheds or other protected places where the soil remains dry. Table V shows how well the bacteria survive in air-dry soil or in sterile soil whether wet, moist or dry, as compared with soil kept moist, or remaining moist for only a sufficient time after infesta- tion to permit drying. The readiness with which the wildfire organ- ism overwinters in sterile soil or dry soil as compared with unsterilized moist soil seems to be a good basis for the assumption that over- wintering is largely dependent upon competition with other organisms as already has been suggested by Anderson (2). To be sure, this ex- planation seems to account for the comparatively rapid deterioration of the wildfire organism in contaminated liquids, rotting leaves and in ordinary moist soil as compared with otherwise sterile or dry soil. On the other hand, when comparing the persistence of the organism on seed and dried or cured leaves with its persistence on cloth and boards under similar moisture conditions, it does not seem to satisfy the re- quirements wholl}^ Although this whole matter requires further veri- fication, we are inclined to include in the overwintering requirements the protective action of certain materials, generally host tissue, and perhaps the absence of injurious substances not commonly considered a<: such. On the other hand, as will be shown later, the wildfire or- ganism may deteriorate more rapidly in pure culture than in the dor- mant condition on seed, in which case competing organisms do not ex- plain their death or loss of pathogenicity. The effectiveness of decay in destroying the wildfire organism is more clearly shown in Table VI, but, on the other hand, conditions favorable for continued decay do not seem to be required for ultimate destruction of the parasite. Various miscellaneous experiments have been conducted with over- wintering which will not be presented in detail except to say that thus far infection has been obtained from air-dried leaves after eighteen months, from cured leaves after fifteen months and from artificially in- fected seed after twenty months, although in some cases the period of longevity of the organism on these materials has apparently been considerably less. It seemed likely to us that if other plants were subject to infection by the wildfire organism these might also prove to be an overwintering agent. To test out this probability a considerable number of other plants (common garden and field crops and common weeds) were arti- ficially inoculated. Most plants tried were found to be subject to the disease when succulent young plants were inoculated under favorable 8 Wisconsin Research Bulletin 62 environmental conditions. The results of this phase of the work have already been published (10). No evidence, however, has been secured that sufficient infection occurs in nature on other hosts than tobacco to warrant the belief that they ever play a part in overwintering, nevertheless it seems worth while for investigators of this disease to bt- oil the look-out for evidence of such cases. Dissemination Studies The (juestion of dissemination of this disease involves a considerable number of problems, of which some are now open only to speculation while others are apparently more likel}^ to be solved by observational than by experimental data. The problems involved are in some respects distinct, since they in- volve long distance dispersal, dispersal to adjoining districts, dispersal from farm to farm, spread from plant to plant in the field, and in the seed bed, as well as the original source of infection of the seed bed. Wildfire apparentlj' spread from North Carolina to thirteen widely sep- arated tobacco growing states east of the Mississippi River in five \ears. At present there can be speculation only as to the agency of dispersal, since manj^ might conceivably be involved. Although pre- liminary experiments indicate that active fermentation may destroy the causal organism, all portions of the tobacco leaf do not ferment actively. Some experiments indicate, for instance, that dry infected leaves could withstand a temperature of 100° F. for five days, although moist cured leaves could carry the organism only one to two days at this temper- .iturc. It seems likely, therefore, that commercial tobaccos of certain kinds may be a common long distance dispersal agent, since the or- ganism may quite likel.\ survive two years in tobacco leaf tissue. Dry wind storms may readily carry infested material for long distances and infected seed and plants may be involved in special cases. All of these, excepting wind dispersal, seem to have been excluded in certain cases of epiphytotics which have been observed. The spread from farm to farm within a given area is still a sub- ject of speculation. The more or less localization of the disease in districts, as in Wisconsin in 1922, seems to indicate local spread which cannot be attributed to infection from commercial tobacco, or even the use of home growii tobacco by the workers as suggested by Valleau and Hubbard (13), a practice which is quite uncommon in the north. A careful survey definitely excluded dissemination by seed or plants as a possibility in that year. Dissemination of infested material by wind, especially dry wind, within the district seemed the most logical ex- plantation, although in isolated cases other means accounted for the spread from farm to farm. Spread of Wildfire in the Field. The importance of rainstorms with wind as a dispersal agent in the field and in uncovered or cloth covered seed beds is recognized by all workers on wildfire. The actual distance and amount of dissemination Control of Wildfire of Tobacco 9 following rainstorms can only be assumed, however, from the area and number of new infections occurring, which are brought out by condi- tions favoring infection. The causal agent may have been spread in many cases before the storm. Unless the wind is especially severe it is not generally believed that rain storms carry the disease over wide areas. In the spring of 1924, some seed bed experiments were laid out to test this subject by placing flats of plants with bare ground be- tween them at varying distances, up to twenty feet, from a central source of infection, but, unfortunately, the results were not convincing owing to the small amount of infection obtained. To test the possibility of man carrying the disease about in any one field or distributing it to other fields, two experiments were conducted. An artificially infected pad of cloth was used with which leaves in the field were brushed sufficiently to break the plant-hairs in one case and touched lightly in another case. Infection occurred in both cases, but was more marked in the former. One experiment was conducted dur- ing a moist period of weather and the other during a dry period. The relative results were apparently the same in both cases. Where wet infected cloth was applied to wet leaves the best infection was secured, although good infection was also obtained with wet cloths applied on dry plants. Some infection was also secured from the dry infected cloth on wet plants, when the contact was sufficient to break the plant-hairs. When dry cloth was used on dry plants no infection resulted. These tests seem to indicate that the disease may be readily spread in an infected field by man brushing against the plants when the leaves or the clothes or both are wet, but not when these are both dry. An important question relative to dissemination relates to the in- fluence of the amount of infection that can be permitted to enter the field on the seedlings when transplanted, and the extent to which infection can be kept down by the removal of diseased plants or diseased leaves. On June 28, 1923, an isolated piece of ground was selected and di- vided into eight plots, each 30 feet by 36 feet. One hundred and twenty plants, three feet apart each way, were set in each plot. These plants were selected according to the amount of disease present on them, the "badly diseased" ones showing lesions on all the leaves and the "slightly diseased" ones showing no actual lesions at all, although they came from a section of a seed bed which had been inoculated about two weeks earlier, but upon which no infection had occurred, owing apparently to unfavorable conditions for infection. The "considerably diseased" plants showed a few lesions on the lower leaves. The different lots were pulled and transplanted by different individuals to prevent con- tamination in handling. The season was unfavorable for wildfire, and at times no signs of the disease were visible in the field. Following light rains, a slight upward spread on the infected plants was noted, but no general spread occurred until following a short rain storm with strong wind about the middle of September, when the plants were full grown. Following this a heavy infection developed. On September 25, the 10 Wisconsin Research Bulletin 62 mimber of leaves infected on each plant was estimated by two different individuals. The average infection per plant is shown in Table VTI. The data show mainly that the "slightly diseased" plants eventually gave almost as much disease as the "heavily diseased" plants. The spread of the disease into healthy plots was evident, more to the eastward than to the westward, and consequently the infection in Plot 1 is believed to be due largely to the organisms originally present. Sim- ilar plots conducted in 1924 corroborated the conclusions from the pre- vious season. The experience with the careful removal of diseased leaves at short intervals from a small center of infection in plots in 1923 and 1924 was of such a nature as to indicate little or no value resulting from this practice if favorable conditions for the disease develop later in the .season. While this work has been done on a large scale in Wis- consin, in the control work in 1922 in cooperation with the State De- partment of Agriculture the subsequent unfavorable conditions for the development of the disease did not give a true measure of its value. Under favorable weather conditions for infection and consequently reproduction of the parasite and for its dissemination, a very small percentage of disease in the field may rapidly develop into a large one. which may subsequently be very injurious to the crop. Transplanting of even a very small percentage of infected plants or of only slightly diseased plants is, therefore, not believed to be war- ranted in view of the damage which may result. It should be stated, however, that observational evidence on a large number of farms under apparently similar weather conditions indicates that there is no close correlation between the amount of infection in the seed bed, or the original infection in the field, and that subsequently occurring. The actual condition of the plants themselves, as a result of local field conditions, seems to play a large role. Seed Bed Infestation. All matters considered, the transfer of infested material into the seed bed is the most important problem to be taken into consideration in connection with the control of wildfire. It has been shown that it prac- tically can be taken for granted that overwintering will not occur in soil lying out-of-doors. It has been shown at this station that the wildfire organism can over- winter on artificially infested seed, and that artificially infested seed sown in seed beds may result in infection of seedings (Plate 2, bottom) A number of trials have been made however, in which infection has not been secured as a result of sowing infested seed, although conditions apparently ideal for infection to occur have been maintained. Although it is not generally believed that seed under field conditions is infested, experience here indicates that it is a factor which must be reckoned with. It is a wise precaution, therefore, not to save seed from in- fected fields, but if seed must be taken from such fields, it should be thoroughly disinfected before sowing. The subject of seed disinfection is discussed on page 14. Control ok \\'iLr)iiRi-: or Tohacco 11 Overwintering experiments here as well as those of others, have .shown that the wildfire organism readily survives the winter in dry or cured infected tobacco leaves. In the tobacco shed and stripping rooms a very considerable amount of refuse containing the living causal organism must exist following work on an infected crop. Here, apparently, lies the most important factor for seed bed infection. The dissemination of this material to the seed beds may occur in a number of different ways, unless precautions are taken to prevent it. The loose refuse should be burned or buried, followed by the precaution of placing the seed beds at a considerable distance from the tobacco shed. The use of lumber, cloth, or any other material on the beds which has been stored in the shed should also be guarded against, since this involves overwintering not necessarily on these materials them- -sclves but on pieces of infected leaves which may be attached to these materials and carriecj to the seed bed. The wildfire survey (Plate VII) in Wisconsin in 1923 brought out the following interesting observation bearing on dissemination from sheds. Out of about ninety cases in 1922, 60 growers placed their seed beds near their sheds and 27 developed wildfire in 1923. Twenty-three growers placed their seedbeds a considerable distance from the sheds, and only one developed wildfire in 1923. Out of nine new cases of wildfire in 1923 seven developed in beds placed near the sheds. This evidence seems to point towards the general importance of dissemination of infected material from the curing sheds to the seed beds. The survey in 1924 did not indicate such close correlation between location of beds and infection, but the season was unsual in many respects and other complicating" factors may have played a part in infection. On farms where wildfire has previously occurred, it is an excellent precaution to keep the seed beds a considerable distance from the building which may harbor the parasite in order that wind, animals, or man may not readily transfer even small bits of infected material to the seed beds. Furthermore, seed bed boards or frames, cloth or sash, should not be stored in sheds. If they are so stored or have been on an infected bed the previous season, they should be cleaned and disinfected if again used for seed beds, .\ccording to our results, infested lumber piled out of doors in such a way that it all becomes wet will not harbor the organism from season to season. Cloth covers are not likely to carry infected material unless stored under infected tobacco. These can be readily sterilized by boiling or steaming when desirable. Dusting and Spraying Experiments In the Connecticut Valley, efforts have been made to control wildfire in the seed beds by dusting with copper-lime dust and spraying with Bordeaux mixture (1, 3). Their experimental results in the green house have shown very marked reduction in the amount of infection on seed- lings following these treatments, and a high degree of benefit was like- wise obtained under out-of-door seed bed conditions. Neither in the 12 Wisconsin Research Bulletin 62 green-house nor in the field is absolute control claimed, however, by these investigators. Experiments along a similar line were started in connection with our work in the fall of 1922 (8). Most of the work has been carried out in the green-house with seedlings in flats (about 14 inches x 22 inches) which usually contained 300 or more seedlings. One set of experi- ments was' also conducted in 3 foot x 3 foot seed bed areas out of doors. The first experiment was planned to show the difference in control obtained in wounded as compared with unwounded plants, inoculating artificially before and after dusting or spraying, together with a rela- tive comparison of the effectiveness of spraying and dusting and their frequencies of applications. The flats were inoculated three times (in a few cases two times) with a water suspension of the wildfire or- ganism from cultures. The data secured are shown in Table VIII. The percentage of infection obtained is relatively high, and it may be objected that this experiment was not a fair comparison as to the value of dusting and spraying on account of the number of inocula- tions and the amount of inoculum used. In the absence of any method for duplicating natural dissemination only the inoculated controls can be relied upon for comparison. While these indicate infection approxi- mately twice as great as that of the sprayed and dusted flats they were not as badly diseased as may be frequently noted in plant beds imder conditions of natural infection. Table VIII indicates that only about 20 per cent more of the plants were infected, and only about two to four more infections per plant occurred in the wounded as compared with the unwounded seedlings. (It is estimated that each plant received on the average ten or more wounds.) In both the wounded and unwounded series, plants dusted or sprayed after the inoculation showed considerably more infection than plants dusted or sprayed before inoculation. No important con- sistent difference between spraying with Bordeaux and dusting could be noted in this experiment, as there was more variation between the "brand" of dust or spray used than between the methods of applica- tion. The Bordeaux paste spray used from appearance was apparently of inferior grade. "Fungi-Bordo" gave better results than "Nu-Rexo." Increasing the number of applications of "Nu-Rexo" reduced the per- centage of plants infected and the number of infections per plant. A second experiment showed that corrosive sublimate sprayed on the plants one-half hour before inoculation reduced the percentage of plants infected from 96 to 49, and the average number of infections from 5.79 to 1.44. Leaf injury was produced by the corrosive subli- mate which could be reduced, however, by adding lime without mater- ially influencing its effectiveness. Following this trial lime alone was tried in comparison with copper-lime dust. This test seemed to indi- cate that lime alone was as effective as the commercial copper-lime dust. An experiment was then conducted in out-of-door seed beds in the spring of 1923, running duplicates in 3 foot x 3 foot seed bed areas. Air- Control of Wir.iMiRfc: ok Tobacco 13 flaked lime. "Limate", "Niagara D-6," "Nu-Rexo", "Corona Bordeaux," "Sanders Dust.' "Fungi-Bordo." "Corona Sulphur, ' Bordeaux (4-4-50) spray and calicum caseinate ("Kayso") were compared. Six applications of the chemicals were made, two being applied before one light artificial inoculation of the wildfire organism made on June 14. On July 11 an examination of the beds seemed to indicate that the "Limate" and "Kay- so" plots were as free from wildfire as the uninoculated controls. Slight infection was found' in the others and considerable infection in the in- oculated controls. "Niagara D-6" and "Sanders Dust" gave some leaf injury but not enough to seriously affect the plants. On October 13, 1923, young seedlings in flats were dusted with "Kayso" and Limate" in comparison with "Sanders Dust", "Fungi-Bor- do", and dry soil. The percentage of plants showing infection are shown ill Table IX. "Kayso" alone apparently gave the best results, due probably in part to its adhesiveness. "Limate" was approximately as good as the copper-lime dusts. Soil dust for some reason increased infection above that of the inoculated controls. These results with spraying and dusting are believed to have some bearing upon the theory and practice of this method of control for wildfire, although corroboration of the results and conclusions may be necessar}^ to bear them out. Copper, the toxic constituent in Bor- deaux spray and copper-lime dusts, has never been regarded as a good germicide and its use as ai spray to prevent bacterial infection is quite unusual in the history of plant disease control, although its value in preventing fungus invasion is universally recognized. The experiments indicate further that copper is not the effective agent in the case of wildfire control. It seems more likely that the effectiveness of spraying and dusting is due in part at least to its physical rather than to its chemical action. While "Limate" or "Kayso" is not recommended for the practical control of wildfire, yet the latter could probably be used to advantage on account of its adhesiveness. Spraying and dusting, with any material used in our tests however, do apparently not pre- vent the occurrence of more or less wildfire in the seed beds when conditions favorable to the dissemination and development of the causal organism occur. The experiments on dissemination have shown that a very slight amount of seed bed infection, in fact an infection so small as to be undeterminable at the time of planting, may result in heavy field infection, providing conditions favorable for the dissemina- tion and development of the disease occur in the field. If, therefore, spraying and dusting do not wholly control the dis- ease, the question may be raised as to the actual value of this practice. If wildfire becomes annually a common and serious seed bed trouble in any given district, spraying and dusting, or some better method of control, may need to be resorted to. Under conditions where only a low percentage of the seed beds are infested in a district, it will probably be safer in the long run for the grower to discard infested seed beds entirely in preference to taking the risk of placing even a flight amount of infection in the field, such as may occur in infested 14 Wisconsin Research Bulletin 62 sprayed or dusted beds. The actual value of this practice however, must finally be determined largely by the resuhs which the growers obtain from its use rather than from experimental trials of the kinds described. Seed Disinfection It was pointed out earlier that while seed was not apparently a com- mon source of wildfire infection, it is regarded as an unsafe practice to sow seed grown one or two years previously in an infected field with- out thorough disinfection. Formaldehyde solution (1-16) was first used for seed disinfection against wildfire. Earlier experience here with this disinfectant indicated that it was injurious to germination in some cases and this observation has also been reported from other stations, particularly from Virginia. Corrosive sublimate H-IOOO) treatment was recommended as a substitute by Fromme and Wingard (6), their experiments having shown that no injury to germination of the seed occurred under their conditions. This treatment was recom- mended and used soon after in the northern sections where wildfire was on the increase. Experience here with the corrosive sublimate treatment like that of others (I, II) proved disastrous, for the reason that while the treated seed germinated in subsequent seed germination tests on filter paper, (Plate 3, bottom) it almost universally failed to germ- inate for the farmers. The injurious action of the corrosive sublimate treatment (Plate 3. top) was found to occur only when the seed Avas sprouted before sowing (either as mixed with rotten wood or as pure seed) as is a common practice in northern tobacco-growing dis- tricts. When sown directly in soil, the treated seed sprout's normally ; and this method of sowing is the common practice in Virginia and other southern districts. Corrosive sublimate treated seed also practically fails to germinate on potato agar. The failure of corrosive sublimate treated seed to germinate is believed to be due to the toxic action of the corrosive sublimate absorbed and retained by the seed, which in contact with filter paper or soil passes from the seed, but in contact with other seeds, in decayed wood or on agar, is not absorbed from the .seed. We cannot agree with Anderson and Chapman'.s (1) explanation of harde-ning of the seed coat in thi.s' respect nor that treatment with water alone iitay result in a similar injury-, although there have been cases where seed treated with water alone and lying in cloth bags in contact with other bags, treated with corrosive sublimate, fail to germinate, apparently due to the diffusion of the toxic property from one bulk to the other. Experiments were accordingly started with the purpose of finding some satisfactory method of disinfecting tobacco seed for tobacco districts where seed is normally sprouted before sowing. A large number of tests on modifications of the corrosive sublimate and form- alin treatments were first tried. Later calcium hypochlorite, "Bacilli-Kil" (B. K.), cupra-ammonium carbonate, electrically generated ozone (with possibly nitrous oxide), heat in vacuo, Seed-o-San. Semesan. Uspulun. Control of Wildfire of Tobacco 15 Bayer's Compound and other commercial compounds, and silver nitrate were tried. Following these treatments the rate and percentage of germination of seeds on filter paper and in bulk were determined, as well as the disinfection secured by sowing the treated seed on potato- dextrose agar plates. The data on this subject are too voluminous to present in detail here so the principal results only are given. Seed stored moist for as long as forty-eight hours, and then dried, showed no injurious effect on germination either on filter paper or in "bulk". Seed treated with corrosive sublimate, kept moist for eight hours or longer after treatment, retarded germination on filter paper markedly, and no germination occurred in bulk. Corrosive sublimate treatment at various temperatures from 0° C. to 30° C. did not appre- ciably influence the usual result (i. e. germination on filter paper but no germination in bulk). Two to eight washings after treatment with the sublimate did not measurably alter its normal behavior. Poor drying after treatment retarded germination only slightly as compared with moderate to good drying. Soaking seed in water up to two hours before treatment with the sublimate had no influence on the result. A twenty minute treatment with corrosive sublimate, 1 to 1000, re- tarded germination appreciably on filter paper, as compared with shorter treatments. Five, ten and fifteen minute treatments with corrosive sublimate. 1-1000 .gave good but not perfect disinfection of seed so far as wildfire was concerned, but did not permit germination in bulk. Corrosive sublimate (1-500) for fifteen minutes retarded germination somewhat more on filter paper than did the standard treatment. Cor- rosive sublimate (1-2000) was not effective as a disinfecting agent. Soaking seed in water after treatment with corrosive sublimate up to thirty hours did not favor its germination on filter paper or in bulk . but rather added to the injurious action secured. These, or other modifications of the corrosive sublimate treatment which have been tried, including those recommended by Anderson and Chapman (I), do not permit the germination of the seed in bulk, or in decaying wood with anything like sufficient certainty^ to warrant its recommendation in districts where sprouting before sowing is prac- ticed. The formaldehyde treatments did not prove particularly injurious to the particular lots of seed used in these experiments, either on filter paper or in bulk up to about 2 per cent formaldehyde with 15 minute treatments. The objections to the formaldehyde treatment lie primarily in the fact that its disinfecting properties are not so reliable as cor- rosive sublimate up to strengths which are not likely to be injurious to the germination of some lots of seeds. Our experience and that of others also has been that formaldehyde (5) is much more injurious to some lots of seed than to others for reasons not fully understood, and it is. therefore, not regarded as a promising tobacco seed disinfectant. Calcium hypochlorite (about 2 per cent CI. water) retarded germin- ation about 10 per cent only in treatments from two to twenty-four hours, but the seed was markedlv bleached. 16 Wisconsin Research Bulletin 62 'Bacilli-Kil" (B. K.), about 3.38 per cent NaCIo up to four hours, followed by washing, did not injure seed appreciably, but seed was bleached and it was not eflfective as a seed disinfectant in treatments of less than four hours duration. Cupra-ammonium carbonate (spray formula) did not injure seed ger- mination up to 1 hour treatment, but it did not give sufificient disin- fecting action to warrant further trials. Five hours treatment killed seed but did not satisfactorily disinfect it. Ozone (with perhaps nitrous oxide) generated electrically did not injure drj^ seed up to eight hours, but wet seed was killed in about four hours. Neither treatment was sufificiently eflfective as a disin- fecting agent in our tests. Heat treatments, even under reduced pressure, did not give satis- factory disinfection up to temperatures that killed the seed in these limited trials. A number of commercial seed disinfectants, mostly of the organic mercury compounds, have been tested both as dust and liquid treat- ments. These have included principally Seed-o-San, Dupont Semesan, Dupont Dust Disinfectant No. 12, Bayer's Dust, Bayer's Compound and Uspulun. None of these met all the requirements for disinfection of tobacco seed. The dust treatments as a rule do not permit germin- ation in bulk, and the liquid treatments retard the germination in bulk til such an extent as to render them unsafe to recommend in practice. The disinfecting value of these compounds against the wildfire organism on tobacco seed proved in all cases to be so low at any of the strengths recommended (and in some cases with increased concentrations and long treatments) that they cannot be recommended for this purpose. In tlic first experiments with silver nitrate as a disinfecting agent for tobacco seed a ^n solution (about .33 per cent) was used in treatments varying from two to thirty minutes. Germination was not appreciable- injured either when tested on filter paper or in bulk, and good disin- fection was secured in all cases. In a second preliminary test silver nitrate was used in strengths varying from 0.1 per cent up to 0.8 per cent for fifteen minutes and again germination was not injured appre- ciably even at the higher strength, and excellent disinfection was se- cured at all concentrations. A number of trials subsequently made with silver nitrate indicated that it is the least harmful of any disinfecting agent tried on tobacco seed, and that its disinfecting properties are as good if not better than that of corrosive sublimate (Plate IV). Ac- cordingly, it was suggested that silver nitrate 1-1000 treating for 15 minutes be substituted for corrosive sublimate treatment of tobacco seed, especially in districts where seed is connnonly sprouted before sowing. During the spring and summer of 1924, a decided outbreak of wild- fire occurred in Wisconsin, owing to very favorable weather conditions for its occurrence. In a few cases seed was suspected of being the agency of introduction into the seed bed, although the seed had been treated with silver nitrate. This led to further investigations on the CoNTROi- OF Wildfire of Tobacco 17 subject of seed sterilization, comparing particularly silver nitrate and corrosive sublimate treatments. For this purpose seed heavily inoculated by artificial means was used, which following treatment was plated out on potato agar, often using as many as forty dishes with around two hundred seeds included in each dish as a test for each treatment. As a result of these tests it appeared that occasionally a wildfire or- ganism escaped the recommended treatments, sometimes one in five or ten thousand seeds. The results again indicated that silver nitrate was somewhat more efifective than corrosive sublimate as a disinfecting agent. It was also especially noticeable that the former was nuich more efifective against fungus saprohytes than the latter. In this connection, it must be remembered however, that the seeds used in these experiments were infested with at least a hundred and probably with a thousand times more of the wildfire bacteria than commonly occur on seed under natural conditions, so that it is doubtful if more than one seed in several hundred thousand escapes disinfection in practice. Since there are, however, three to four hundred thousand seeds in an ounce, the possibility remains that occasionally wildfire may escape the present methods of seed disinfection. In order to reduce this possibility to a minimum or eliminate it en- tirely, a double seed treatment with silver nitrate was resorted to, permitting the seed to dry one or more days between treatments. At the same time the length ol the treatment has been reduced since early experiments showed that even a two minute treatment with silver nitrate was very efifective. With the double treatment, each treatment lasting ten or even five minutes, it has been possible to disinfect the seed so completely that no wildfire organism has been recovered from seed so treated after extensive trial. Therefore, it seems that the double treatment should be used in preference to the single treatment in cases where the most reliable disinfection is reciuired. The germination of the seed is apparently more retarded by two treatments than where only a single treatment is given, but this has not been found to be of more than one or two days duration and, con- sequently, is not to be regarded as a serious objection. The advantage (^f two five-minute treatments lies largely in the fact that germination is retarded somewhat less than with two ten-minute treatments. It has also been noted here that it is not advisable to sprout the seed in the same cloth in which it was treated. Loss of Virulence The experience of most workers with the wildfire organism has been that it may relatively rapidly lose much or all of its virulence in culture. This phenomenon is common with many bacterial parasites and is said to occur in nature also. A simple though fairly extensive experiment was conducted with B. iahacxim for the purpose of determining in the first place the best 18 Wisconsin Research Bulletin 62 cultural method for this organism in order to retain its virulence, and secondly to form a possible basis of reasoning in regard to overwinter- ing of the organism. Three virulent strains (isolated from different sources) were selected and transferred to three different media, potato-dextrose agar, beef- peptone agar and bouillon. These media were made up in suflficient quantity to last throughout the experiment and sealed in tubes with paraffin. In all, about 650 cultures were involved. Several original transfers were kept, and transfers were then made serially weekly and monthly, fifty-two weekly and thirteen monthly transfers being made in the experiment. The cultures were kept in the refrigerator at about 8-10° C. throughout the experiment. At intervals of about one month the non-transfers and the last weekly and monthly cultures were tested for their virulence by making 50 wound inoculations with a water sus- pension from each culture on the leaves of \'Oung tobacco plants in pots supplying suitable conditions for infection. The results are recorded as percentages of infection. Considerable difference occurred in the rate of infection and the size and appearance of the lesions, but these cannot be gone into detail here. The results on the percentage basis were on the whole quite variable when compared from month to month, due undoubtedly to variation in environmental conditions af- fecting infection. Studied in detail, the results also show occasional contradiction, i. e., a culture would at one time give a higher, and at another time a lower percentage of infection when compared to an- other. Taken as a whole, however, tlu' following conclusions seem warranted from the data. The degree of virulence showed a general downward tendency on all media with increasing age, most marked on beef-peptone agar and least marked on potato dextrose agar (Plate V). In the case of potato agar the greatest loss of virulence occurred when no transfers were made (Table 10) and the least when weekly transfers were made. In the case of beef-peptone agar the greatest loss of virulence occurred in the weekly transferred series and the least in the no-transfer series. At the end of 15 months the loss of virulence was complete on beef- peptone agar in all three strains used. In bouillon not much effect of the transplanting itself was noted, the evidence 1)eing somewhat in favor of monthly or weekly transplants above no transplanting as regards retention of virulence. Some difference existed in the three strains used in regards to their ability to retain their virulence under any one condition, strain 3, for instance, was over twice as virulent as strain 2 on untransferred beef- peptone agar. When transferred back into potato-agar, the cultures in all cases seemed to be approximately equal in vigor of growth, but virulence was not materially altered. None of the cultures, as a rule, gave as high percentage or as good infection as freshly isolated cultures from new lesions. Aside from this the best culture medium for /?. tabacitm seems to be potato-dextrose agar, with transfers at intervals of somewhat less than one Plate I 7'op— Typical s> luploms at wildlirc on portion of toljacco leal'. 'I'lic clilorotif arm or '"halo" sinrounding a whitish or bio%Miish central necrotic area is characteristic of lliis disease. Bo//om— Wildtlre infection in a seed bed often kills young plants. ? %?v, r Plate H lap l.:iil\ :iii(l l.:ilc s.Miiploiiis nf lihuklirc 2'{, and li»2l art" sho%\ n. It will bo noteti that on sonic I'ainis the disease rc-occuiTctl three years in succession while on other (arms no inleclicni ociiirreti alter 1022. A marked spreatl ol" the disease can he noteti in 1921. The survey was matle in tietail in only the townships of Rurke ami Sun Prairie in 1924. Tlie survey from which this map was ruatle was supporteil by the Wisconsin State Department of Agriculture. Control ok Wildifre of Tobacco 19 month. The wildfire organism may lose all or part of its virulence under relatively favorable conditions for the growth and storage of the organism. The fact that it has favorable conditions for multiplica- tion outside of the host is not necessarily conducive to continued patho- genicity. That the organism can and does retain life and pathogenicity upon such materials as seed or leaves in the dry and latent state is evi- dent from the overwintering tests. Under other conditions, whether the material which harbors it is dry or sufficiently moist to favor growth, it may rapidly die out or lose its virulence as evidenced by the results with soil, cloth and wood. It seems evident, however, that whether the wildfire organism is in all cases really killed or merely loses its virulence has not been actually determined in overwintering ex- periments conducted thus far. Studies along this line may explain some of the peculiar cases of behavior in overwintering studies. The Wildfire Toxin The common and characteristic halo surrounding the ordinary wild- fire lesion, the less common chlorosis of bud leaves which sometimes occurs, with few, if any, organisms in the chlorotic area is evidence that a toxic substance is produced by the wildfire organism which is apparently soluble. To obtain further information on this point, cultures of the wildfire organism on potato agar were suspended in water and filtered through a small Berkfeld filter. The filtrate proved sterile on plating and was used for inoculations on tobacco in the ordinary way. Typical halos were produced by the sterile filtrate in one to two days. Platings from these spots showed that they were sterile (Plate VD. These tests, with other modifications, were repeated five times with similar results. The wildfire organism produces toxin which, though greatly diluted, is very effective in rapidh' producing chlorosis in plants. The wildfire bacteria when washed free of this toxin required several days longer to produce typical lesions than did the toxin alone. This observation is of considerable significance in work with the wildfire organism and led us to reconsider some of the previous experimental work here as well as that of others, and explained some observations previously not un- derstood. It was always noted, for instance, that the wildfire organism required several days to produce any signs of infection from some of the overwintered materials, whereas 24-48 hours suflFiced for cultures. It is evident that the presence or absence of already formed toxin had much to do with this result. Most of the determinations on overwintering and influence of other environmental conditions on the pathogenicity of the wildfire organism have been based on inoculations of living plants. It is possible, there- fore, that these results apply more particularly in some cases to the effect of the toxin than on the organism itself. Anderson, for instance, found that alternate freezing and thawing did not kill the wildfire or- ganism, since inoculations with the exposed cultures produced typical 20 Wisconsin Research Bulletin 62 infection. In experiments here, cultures exposed to alternate freezing and thawing for short periods also gave infection, but these same cul- tures failed to give growth on other media when transfers were made from them some time after the exposure. It seems evident that the or- ganism was largely, if not wholly, destroyed, but that the toxin was not injured. The loss of virulence or pathogenicity, as noted in the previous chap- ter, is no doubt related at least in part to a tendency of the organism to continue to produce toxin in culture. Apparently the observation that old cultures readily produce symptoms on plants, as compared with subsequent transplants, is not the result of any greater virulence of the parasite, but is rather a consequence of the transplanting of the toxin previously produced. Similarly, the question may be raised as to whether the symptoms produced on a plant, as a result of introducing the toxin through a wound, justify including the plant in the host range of the parasite. In work with the host plants of B. tabacuni (10) no distinction was made as between the toxin and the organism, in all cases probably inoculating with both. Some of these trials have been repeated sufficiently, how- ever, to justify the belief that in most cases, at least, the organism was actually parasitic, although the initial symptoms may have been pro- duced by the toxin introduced from cultures. The records show, how- ever, that in practically all cases infection was obtained by spraying the inoculum on unwounded plants, as well as by wound inoculations. Practical Considerations With the appearance of a new disease of economic importance, the quick demand for control measures often requires the dissemination of such information as may be rapidh' gained from limited experimental data, together with deductions from what is known about similar dis- eases. When the problem is subsequently more thoroughly investigated, it is natural that the relative importance of the earlier recommendations will be altered, possibly some eliminated and others added. This es- sentially has been the history of the development of wildfire control measures. Control measures can l)est be applied by adequately under- standing a disease and selecting rind using control measures that apply best to the case at hand, rather tlian by l)lindly following directions. The results secured from the investigations described in this bulletin do not fundamentally change the principles, of control which have been previously recommended I)y this Station and by workers in (itlier states. They corroborate previous results based on meager data, justify or eliminate certain recommendations which were in doul)t, in addition to altering some of the methods. The fundamental consideration which should be kept in mind in controlling the wildfire disease is that it is of an infectious nature and that measures for its control are, therefore, based largely on efforts to. prevent its introduction in the first place into the seed beds, and failing Control ok Wildfire of Tobacco 21 ill this the necessary precautions should be taken to prevent its intro- duction into the field. When wildfire has occurred on a farm in the preceding season, it is evident from our experiments that any material which may harbor even extremely small pieces of infected plant tissue may be a possible source of infection to the new plant beds if permitted to reach them, and favorable weather conditions for the infection follow. It is believed, however, that the actual dissemination of infected material from the curing sheds, where the material has remained dry, to the seed beds is one of the most common sources of infection. Seed bed frames or covering, or any other material coming in contact with the seed beds should not be stored in the curing sheds, where infested tobacco hangs, without being thoroughly cleaned and disinfected before using. It is not believed that wood or cloth readily harbors the wildfire organism except as it carries pieces of infected leaf tissue. To further insure sanitary seed bed conditions it is believed advisable to locate the seed beds a considerable distance from the farm buildings or, at least, from the curing shed, since various agencies may easily carry infected material which may be harbored in or about buildings to the plant beds when they lie close at hand. The refuse from the preceding year's crop should be burned or buried to reduce danger of its dissemination. It is an unsafe practice to sow seed grown in fields infected with wildfire, without adequate disinfection. Corrosive sublimate cannot be used for disinfecting seed when seed is to be sprouted before sowing. Silver nitrate, one part to one thousand parts of water, is a satisfactory disinfecting agent, but two treatments of 5 or 10 minutes each, allow- ing the seed to dry between treatments, is believed necessary to insure complete disinfection. Some retardation to germination usually occurs. Seed should preferably not be sprouted in the same cloth used in dis- infection. The wildfire organism dies out in a comparatively short time in moist soil. Planting in previously infested fields is believed to be a safe practice, especially if the "stubble" from the preceding crop is plowed under in the fall. If infection occurs in the seed beds it is hazardous to use even ap- parently healthy plants from such beds. It is believed to be a good plan to construct small beds, separated by paths, rather than to use large, continuous beds in which infection can spread more readily. In this manner the seed bed areas not infected can be used with greater assurance of safety. At the first signs of infection in the seed beds, effort should be made to destroy the infected plants together with the immediate surrounding area. After trying out various methods, the conclusion has been reached that the most convenient and cheapest way to do this is simply to cover these areas with three or four inches of soil. In case a heavy early infection develops in the field, plowing under and replanting with healthy plants should be given serious consideration. If plowing is not done, all infected plants should be removed before re- 22 Wisconsin Research Bulletin 62 planting. Removing the infected plants only or picking off infected leaves is of doubtful value in checking the disease if the disease is scattered throughout the field, even on only a small percentage of the plants. Working in a vsildfire infected field when the plants are wet is con- ducive to spreading the disease, if the plants are of such a size that the leaves are touched consecutively. It is believed that the methods of control workrd out and suggested for wildfire apph' equally well for the similar tobacco disease known as blackfire (Plate II, top). Control of Wildfire of Tobacco 23 SUMMARY 1 Practically every case of wildfire infection in the field can be traced to seed bed infection. This is borne out by three years of observation of the disease in Wisconsin, as well as being sup- ])orted bv reports from various other states. The control of wild- fire is. therefore, almost entirely a matter of preventing seed bed infection by the wildfire organism. 2. Locating the materials which are the common sources of carrying. the wildfire organism to the seed beds is. therefore, an important phase of the development of control methods. This involves determining on which of the various materials likely to come in contact with seed beds the bacteria causing the disease are most likelv to overwinter. 3. Other things being equal, the wildfire organism lives over winter most readily in the dry and dormant condition. 4._The wildfire bacteria readily overwinter on infected to- bacco leaves which are cured or dried and which remain dry be- tween growing crops. Small amounts of infected tobacco trash may accidentallv reach tobacco seed beds in a number of differ- ent ways and this material is believed to be a common source of infection. . 5 To reduce chances of infection from overwintered material in the curing sheds, it is advisable to locate the plant beds a con- siderable distance from the curing sheds or other places which may have harbored infected dry tobacco over winter. 6. The wildfire organism can readily live over winter on the seed In fact, the experiments indicate that it can live in the dormant stage as long as two years on seed. Seed is not. however, apparently a common source of infection, although it may become so. It is not to be regarded as good practice to sow seed from infected fields, unless thev have been adequately disinfected. 7_The wildfire organism does not seem to remain alive as readily on wood or cloth, even when kept dry, as on leaf tissue or seeds but since these materials mav readily harbor infected leaf fragments, especially if stored in the curing sheds, the cleaning and disinfection of seed bed frames and covers may frequently be advisable. . . 8. As far as can be experimentally determined the bacteria do not over-winter in moist soil, consequently there is no danger, as far as known, in using land for tobacco which has grown a previously infected crop, especially if the refuse and stubble from the preceding crop are thoroughly plowed under. 9_Plants from infested beds preferably should not be used for transplanting. If weather conditions are favorable for the disease an extremelv small amount of seed bed infection may re- sult in heavv field infection. In a small percentage of cases this seed bed infection may be so small as to escape even careful in- spection. 24 Wisconsin Research Bulletin 62 10. — Spread of the disease in the field is almost entirely de- pendent upon rainfall, especially with strong winds. No satis- factory measures to prevent the spread of the disease in the field are known. 11. — While dusting with copper lime dust or spraying with Bor- deaux mixture in the seed beds, reduce the amount of infection, these procedures are not believed to prevent seed bed infection sufficiently to materially reduce the amount of subsequent field infection if conditions for the spread of the disease become favor- able. 12. — Seed disinfection with corrosive sublimate cannot be used where seed is to be sprouted before sowing. After trying out a number of seed disinfecting agents, it was found that a solution of silver nitrate (1-1000) gives the best results. It is believed, how- ever, that two 5- or 10-minute treatments (drying the seed be- tween treatments) is required to adequately disinfect infected seed for wildfire control. 13. — The wildfire organism may often lose its virulence in cul- ture in a relatively short time. Many factors seem to be con- cerned in this loss of virulence, among them being the nature of the culture medium, the frequency of transfer and "strain" differ- ences in the organism. 14. — The wildfire bacteria produce a toxin in host tissue and in cultures which is responsible for the chlorosis produced in plant tissue. This toxin is readily separable from the bacteria by filtra- tion and will in itself produce typical symptoms by inoculation. This fact should be taken into consideration in further studies on such subjects as host relatiotus, overwintering, loss of virulence, and culture studies. Control of Wildfire of Tobacco 25 c ; 1 , 1 ■' — < ^ o + ■ ; H - : +1"= B <- .2 2: 1 ■""^ 1 1 — cb"^ 1/3 i a-. + o o O o O o H I ff- " = l + l<= 3£; B 1 1 sl ^ 1 1 1 — g:* t3 1 1 . oe + + o o O o O o o H - ^ . = ., -i| u 1 1 o 1 c 3 3.5 0 CE j= ^ - r- 0 > »B o c^ o o o O o o o c , ^ > o 1 « 3 "~ >.£ £ 1- 00 ^ 0( ) « r 3 X ■"^ 1 — c" h es \Ci lo 1^ 00 o o O o o o cx ^" e ■^ ? ?; _c ftj ■" 1 ■*^ •<* o o o o O o o o c c c «: •fl c ! « OS's c u C-. n c 0( a -or u: c — ^^ < C^l !N C c > tc > sH "S 1^ a ff c C n u u S^l oc M o o r? C" a- X o c CN •^ •> (> 3 S- m 00 O M C>1 e 0- a _ £.^ """ te c^ « o X X ■t X n «c c ~1 c « c > c 0.5 ^=^ *-• ;c »— n •V M M ■^ c 0- c tr. a. >> »— JS CO oo CSI ■ e t^ c^ ■^ w ■^ re t^ •— 0- 3 0 CO - 1 c Q 'S' o o o X r£i ;S o T? "^ T? M T c > c t/; o 1^ o t>- CO in •^ 1 a BCJ^ 1 «i c _ 1 '2 re • Y. N o Tt o o M X tc o ■X ■^ IN c ) cv ^~x o r^ o o t^ hCC ■* cr ;^ "~ Cj/ e - •^ ^^ ^- 1 " '- — u: — ^ c- '— H ^ o M o o •^ o o X •* M M c > ■^ "* 3 .— M ■X rr t^ rr CT ^ - Z •^ ^^ -B-^ — t- " '— o| u u £ 6 a c Q. a c -B C X ^ 1 3 >■ 3 > 3 > > > > > « — CB C8 a C3 n « CO Cv; O 3 V U 0. u a >- ;, 3 — 3 3 — n 0 3; IT M 3 C 0 C u c o C O B u c c c U z B u V v V u b v 1. > > a u u V u Oi u 0. &. u a CI t. Oi t^ 5 3 u 3 3 u 3 u 5 c a " t< "" w C a fi. c CU c a. C c ^ £. a 0 X ^ ^ _ ^ fc ? 5c/: 1 « is ff^^ ■ 1 u ■ — j^ — *- f\ •^ 0 "5 c "■ < ^ I CI B 0 i u •* 0 0 B 0 5^ ii-B t^: c c CC CC Id _ ■ ^ llli Id f- ■M • •c m' 3 u. H U. id C • • • as 3 ^ 0 B 0= + - u. CC fc4 ■*' bH D Z = > <£ % CO 0! it z - Id t _c CO X B Cco = li 3 c e X ■SurS = a u c QC ■o u QC c "5 'ic 3 " BC z E o fi u a « > > C9 c X a CO > CO 5^ i.5 fij^x — ■' tl '. O ^ c« 'f\ < u C/5 (S w -!: c/2 £ ~ < u z-l a t5 u. a a Ul z o z a z III o K 7; O < »', Cu o S cc o a u H 1 U 1 ^ ,__, 05 1— 1 H a a w H Wisconsin Research Bulletin 62 I ;i I I I I 1 I I I I I I I ! ! l^-(.4.loioloioololoooooo-f- c- + + o ooolo -f -|- olo >■ O -|- + + ° !D OI i O O o- ■:- o o o o o o to I •* o o ;0 00 •* O O 00 O OJ.I •* I M C^I ® i + !=>! + => + o o I o 4- 004- ■* o -I- => + Control of Wildfire of Tobacco 27 Q Id 5 Q X 2:^1 o . 71 <^ b Q ^ w q H '•2 < o Z. u m Ifl o CO ;0 1^ o o so ?< O S oeo •< S so to so :? t^ 1 so so d 7. .2-c .« u » •* CO 00 00 00 O SO Test Age in months o ^ o so :5: in CO 6 Z_ u S • 1" CO o o w M oo a a OBO 1 -^ 1 o in re 'T d Z. 0) j3f 05 112 1 o o so :3? M CO so ; d Z .2* = M ■^ 'I' S5 CO 1 O O Tj< o 1 i _o a a aco ^ 1 :^ 1 : ^ i 1 1 Condition of storage Dry >1 •j: 'o "S. o 2 Q M 'o B op 5^ 3 2 1 : 91 0 a; "a o u o 3 ja ■2 il C SO '3 1 Yl 2 a -2- 'o OS a BO = 1 03 > s > J e ■ 28 Wisconsin Research Bulletin 62 c « 1 1 CA CO irt CO <=> 1 ^^ Tt i^ 00 • I^ « ^^ ■^ C5 0-. X Tt 1 CO X a r^ oe t^ C-. d ■«o Z OJO 1 v 1 tn h .CJ3 ::?: ::?: ^ :i^ ;?? ^ ' ^ :^ ^ \N CfiO CD <» CO CO in in in in in in <£ a M w CO to 05 to o •># o 00 00 CO §a in in CO p; O c^ o 'mO Z a>o o _ »> . h .Cj= ^ aoo in in in in ■^ ■ CO « >. ^ 1 t u : C C " I \ o i ■^ Control of Wildfire of Tobacco 29 > CQ < Z. SCO -Ox QOO .= 2^ ■r. a X ^ u a tsoo ooo ::s: :3? ::^ o « :^ h^: :s: :^ ^ , TJ ■o -o u O c a. a 0} X 'Jl v CJ > > s 3! i) 30 Wisconsin Research Bulletin 62 d H S «o CO d 2: if o o 05 o -.c o ■^ « 0} [-1 <:e o lO in ITS in in tT d o a J8 o o o - Ci M- o CO 0 .55 ^ :« ^ - :^ :^ ^ Condition of storage 'o 'S IS >> u Q >> Q u ■V c u 3 O x: a 'o C 4) J3 3 O 1 2 "o D a cs E ■o 6 c > a _o c u V t^ DO OD C O > a V O « u c e a u QD •o (U o •2 n '^ — "a o 0 C in — 0) 5^ °- >> • u • •O • 4* . o • (U ■ s ■ '$ : f « c u> T3 J3 ?- a w -> a) C32 "3 0 0 0 0 n c ,„ > •a Oi -0 0 0 0 « c „ > 4) Control of Wildfire of Tobacco 31 .u Wisconsin' Rf.sf.ar(H 1)Ulletin 62 Av. N... infections per plant j^J o CO co' 50 CO in 05 CM 35 05 00 1 C0-|° 1 i 1 00 c- ■ oo CO O CM in o - 00 CO CO o •a o o c _ c "a. o a . DC so c o u > in UO d 00 in 00 O in «o in 00 m IN 00 in d in in 00 in 00 05 in • o m 05 o CM o «o m in in in CO m o 00 in o CO in d m CO m 00 o o V3 ■■o M 00 00 CO CO CM 00 00 00 05 CD o OS Ti- er. CO o 05 o o in 1^ CD m 05 m in c^ 00 00 un r- t^ o 'u 4) '72 00 00 o a: 00 i ! c-j r- t^ 00 1 o 1 ■* t^ oo a: 05 1 1 ffl •* t^ S 00 3-. in ■■£ CO o CO o c _o CO o "H. 0. a o 6 Z ;^ '5 c 3 CO CO 1 CO CO to CM CO CO j ! ! i CO O O O CO 1 CO CO 1 T« 1 1 in CM CO 1 1 1 co' CO 1 o o V a a CO CO CO CO CO M CO CO CO CO CO c 1 CO CO CO CO CO C^l CO CO CO CO o v. Si 3 3 C o X u CC 3 z CO a i< X 3 CO 0) -o 0 CQ a. 3 7. >> CO a 72 . ^ S ^ M M X , B ,^ O X •■^ 0 ;o in fO Jm "o ^ d o 4)' 1 ,3 < ao >> 1 3 m to o w' -- > < V o> •* X D ^ ;D o O) ?. ? M 5^ in 00 o U« O s> r- 35 ^ C Z ■^ in to d ■£ « >^ X ^ j: M ■* X z u: ^ a ® Tf to Z'JI .3 o « in ro 2^ ca 1h s j t/3- a: ^ 3 M 05 "* < u rl o in ^ r^ in e^j ~ '* S "^ c »* 1 to o ~< o ^^ t-~ X -S z Tf irt M ?-. K 1 _ a „ _>;. •^ ^ X -" X .:t z S C o (N ^^ '- < 'n O ;C to T« >^r* b S ~ r"' 55 1 ^ H _>> ^ 2 3 O M p 2 V O 00 1 ^ ^ •^ L- ^ « 1 in (>t k".- < 1 z-^ V •^ 1 ^ to C 2 c H z o Z Tfl in OS z - _>. J3 r- i^ \r c ■*-» C c S to M t^ o 2 >n in tr: '^ oc in ^S ^ ffi in M £^ ll t^ X u cd * 0 ) > CO a V 03 n 1 z u gcs to C5 '. o 3 i3;c to 43 • CtO X2 3 X 1 Woe 1 O 1 T3 a*". \ J 4);0 o^ ^ § = , "« ft !-H a. ffi CD 1 34 Wisconsin Research Bulletin 62 Table IX. — Percentage of Plants Infected Following Dusting With Different Materials for the Control of Wildfire in Green-House Flats. Material used Percentage ot infection Series I Series II Average Sander's Dust 75 63 69.0 Limate 58 79 68.5 None (inoculated control) 84 79 81 5 Fungi-Bordo Dust . . . .' 64 84 74.0 49 65 57.0 Soil Dust 96 90 93 0 None (uninoculated control) 0 0 0 Literature Cited (1) Anderson, P. J., and Chapman, G. H. 1923 Tobacco wildfire in 1922. Mass. Agr. Exp. Sta. Bui. 213: 1-27. (2) Anderson, P. J. 1924 Overwintering of tobacco wildfire bacteria in New England. Phytopath. 14: 132-139. (3) Clinton, G. P., and McCormick, F. A. 1922 Wildfire of tobacco in Connecticut. Conn. Agr. Exp. Sta. Bui. 239 : 365-422. (4) Fromme, F. D., and Wingard, S. A. 1921 Treatment of tobacco seed and suggested program for control of wildfire and angular leaf spot. Phytopath. (abstracts) 1920 : 21. (5) Fromme, F. D., and Wingard, S. A. 1922 Blackfire or angular-leaf spot of tobacco. Va. Agr. Exp. Sta. Tech. Bui. 25 ; 4-42. (6) Fromme, F. D., and Wingard, S. A. 1922 Blackfire and wildfire of tobacco and their control. Va. Agr. Sta. Bui 228 : 1-19. (7) Jenkins, E. H. and Chapman, G. H. 1923 Wildfire of tobacco in 1922. Conn. Agr. Exp. Sta. Tobacco Exp. Sta. Bui. 2: 7-38. (8) Johnson J. 1924 Experiments with dusting and spraying for the control of tobacco wildfire in seed beds. Phytopath (abstracts) 14: 28. (9) Johnson J. and Murwin, H. F. 1924 Disinfection of tobacco seed against wildfire. Phytopath. (ab- stract) 14: 28. (10) Johnson J,, Slagg, C. M. and Murwin, H. F. 1924 Host plants of Bacterium tabacum. Phytopath. 14: 175-180. (11) Thomas, H. E. 1924 Tobacco wildfire and tobacco seed treatment. Phytopath. 14: 181-187. (12) Tisdale, W. B. 1923 Report of Tobacco Experiment Station. Fla. Agr. Exp. Sta. Ann. Rpt. 1923 pp. 125 R-140 R. (13) Valleau, W. D., and Hubbard, C. 1924 Angular leaf spot and wildfire infection of tobacco plants by spitting. Phytopath. (abstract) 14: 29. (14) Wolf, F. A., and Foster, A. C. 1918 Tobacco wildfire. Jour. Agr. Res. 12: 449-458. (15) Wolf, F. A. 1922 Wildfire of tobacco. N. C. Agr. Exp. Sta. Bui. 246 : 4-26. Research Bulletin 63 September, 1925 Transmission of Viruses From Apparently HesJthy Potatoes JAMES JOHNSON Agricultural Elxperiment Station of the University of Wisconsin Madison CONTENTS Introduction 1 Experimental Methods 1 Inoculations with Diseased Potato FoHage 2 Inoculations with Apparently Healthy Potato Foliage 2 Inoculations from Tubers and Other Organs of Potato 4 Symptoms of the Diseases 4 The Infectious Nature and Increasing Virulence of the Viruses 5 The Properties of the Viruses 7 Trials with Potato Seedlings and Other Healthy Plants 7 Other Host Plants 8 Transmission Back to Potato 8 A Combination Disease 10 Discussion of Results 11 Summary 12 Transmission of Viruses From Apparently Healthy Potatoes' DURING the course of crossMnoculation studies on certain virus diseases of solanaceous plants, it was noted that symptoms were secured on tobacco from potatoes selected as healthy controls, and that these symptoms did not materially differ from those secured when \ari()us virus diseases of the potato were used as a source of inoculum." An investigation of this matter, therefore, was undertaken and it became increasingly evident that extracts from potato plants which are healthy, within the ordinary meaning of this term, are capable of inducing symptoms of disease in tobacco and other solanaceous plants. Furthermore, the ability of inducing tliis disease is apparently universally present within most, if not all, of the standard varieties of potatoes. Three distinct syniptiinis have been secured which are associated with at least two and prol)ably three distinct viruses. These viruses behave like those of most virus diseases of plants as regards transmissibility, and have been not only transferred repeatedly through several generations of tobacco, but have also been used to infect a wide range of other solanaceous plants. In fact, one of these virus diseases when inoculated back into the potato, after having existed in tobacco for one or more generations, produces under the proper environmental conditions a most malignant disease. IMany problems have naturally developed during the course of these in- vestigations, centering around an explanation of these results. As far as can be judged at present only two theories appear to be logical. Either potatoes arc almost universally infested with viruses or they are capable of initiating virus diseases in other plants. Whether actual proof of one or the other of these theories can be established remains to be determin- ed by further experimentation. In the meantime it has been thought advisable to present the data secured up to this time in summarized form, with brief discussion of some of the more important features of the problem. Experimental Methods The potatoes used in practically all cases have been grown in a low temperature (17-22° C.) green-house suitable for a good development of the potato. In connection with other experiments similar plants have been subjected to a wide variety of environmental conditions, including high temperatures, and in no case have selected healthy potatoes exhibited any symptoms of a disease comparable to those to be described as a result of these chahges in environment. The Triumph variety of potatoes was used in all experiments unless otherwise mentioned. ^Cooperative experiments with Office of Tobacco Investigations, Bureau of Plant Industry, United States Department of Agriculture. -Johnson, Tames. A virus from potato transraissable to tobacco. Phytopath. (abstract) 15: 46-47, 1925. 2 WlSCd.NSlN ]\KSi:ARfll IJULLETIN 63 Tin- tobacco plants and other host plants used were grown in a high temperature green-house (27-32° C). The plants were grown in very fertile soil, transplanted to 4-inch pots and usually inoculated when very young, i. e., with only two to four leiivcs large enough to be inoculated. The potato or c>thcr foliage used for inoculum was crushed in a small sterile mortar, the juice strained through cheese-cloth, and inoculated by means of sterile needles wrapped at the end with a small wad of absorbent cotton to more readil_\ carry drops of the inoculum. I'wenty to forty punctures per plant were usually made in the leaf blade and midribs, al- tliough experiments showed that a fair ])ercentage of infection could he secured by a much smaller number of punctures. In the early experiments, ten plants were usually used in each serii'S of inoculations. In later experiments only five plants were used. This number gives equally reliable information except in cases of negative results, in which case, however, the experiment has always been repeated with the Same source of inoculum. Infection was sometimes evident in six days, although ten to fourteen days were usually required for final counts. The strain of tobacco used in practically all the experiments was the common commercial variety grown in Wisconsin, Connecticut Havana No. 3S. Several other of the more distinct varieties of Nicotiaiin tabacuni have been tried sufficiently to warrant the belief that probably no important \arietal ditTerences exist as regards susceptibility to these diseases. Inoculations with Diseased Potato Foliage In connection with the earlier work, and in later work where mosaic potatoes were used as controls, inoculations have been made from 2K different Triumph potato plants with mosaic symptoms. This involvcfl inoculation to 210 tobacco plants, infection being secured on 154 or 73 per cent of the plants inoculated. The potatoes used were mostly Wisconsin grown, coming, however, from several different farms. Inoculations have been made, from fourteen other diseased potatoes show- ing yellow-dwarf, spindle tuber, leaf roll, rugose mosaic, and other symptoms not clearly defined. These ])lants come from tub.^rs grown in such widely separated sections as Maine, New York, Wisconsin, and Oregon, as well as being of different varieties. One hundred and five inoculations were involved, infection being secured on 71 plants or 68 per cent. No consistent differences were noted between the synii)tonis secured from the different diseased potatoes nsed as sources of inocnluni. Inoculations with Apparently Healthy Potato Foliage The potato plants used in this group of inoculations have been largely of the Triumph variety grown in Wisconsin. In connection with a tuber indexing project single eyes have been grown from over 12,000 different tubers, coming from twenty different farms during the past winter. These potatoes grown at a low temperature (17-22° C.) in the green-house in fertile soil have given an exceptional opportunity for the selection of- Transmission ok Vikusks 3 iiurmal plants as tlu' sdurcc of innculum. From this stock, however, ap- proxiniati'K onl\' 170 iK'althy plants have hecn used, over fifty of these o'tning from stock indexed for mosaic the previous season and grown in is»3 , ^.li a 2 E ,. ij' "^ c !? o 2 2 o £ .2 o £ o.'o a >;• c r - M ^■u)2 . o *- S >.— ; 2Sl5 H ?■:&-■- rs ■' SJ 3 .S n fi " ::; o g c .2 3 2 ** ?i H-5 O- 3 C3 3 - O c ^ »j — - ■ >. ^ •:; 2 "J i; ^ -- w ^ .— ._ WJ ._ g 5 s s'-5's 0 o H"^5 « ■c ^ 5 -s. o o Transmission of Viruses 7 It should be noted in this connection that there is a decided tendency at times on the part of plants to recover following the first attack of the disease. This has also been noted in connection with other virus diseases of plants. Whether this is merely a form of partial or complete masking as a result of minor changes in environment, or represents a tendency on the part of the plant to develop less predisposition to the disease has not been determined. The latter suggestion is to some extent counteracted by recurring periods of attack by the disease. The Properties of the Viruses The properties of the individual viruses have not yet been studied in as great detail as they merit. Some of their more important characteristics, however, should be stated at this time, although it is expected that this subject will form the basis of a later paper. The viruses are apparently not as readily filterable as the virus of the ordinary mosiac disease of tobacco. However, they will pass through the coarser porcelain filters which yield a sterile filtrate. Transmission by aphids has given only very low infection from tobacco to tobacco. On the other hand, the potato aphid has given high percentage of infection of the "spot-necrosis" type from potato to potato or from potato to tobacco. These differences may be due primarily to the species of aphid used in relation to the host plant. E'^periments conducted to study the longevity of the virus outside of the living host, both in the liquid extract of the plant and in the desiccated condition, have given some variation in results, both as regards the particular virus used as well as between the separate experiments. In general these viruses are strikingly different from ordinary tobacco mosaic in that they are short-lived when separated from the living host. In most cases the survival is less ;han 20 days and frequently as low as 10 days. In general the virus survves longer in the drying leaf than in extracted plant juice. The viruses seem to be considerably less resistant to heat than is the ordinary tobacco mosaic. The "mottle" and "spot-necrosis" types are ap- parently destroyed at about 70° C. in 10 minutes. The "mottle" and "spot necrosis" type have been diluted up to one to five thousand, and have still yielded good infection. Further dilution could no doubt be tolerated. Judging from preliminary experiments the viruses in question are relatively resistant to germicides and other chemicals, as compared with bacteria, being more similar to tobacco mosaic in that respect. TriaJs with Potato Seedlings and Other Healthy Plants Repeated attempts to induce the diseases in question in tobacco by inocu- lating with juice from the foliage of potato plants grown from true seed have failed or yielded only questionable symptoms. The seedlings used have been small for the most part, but in a few instances plants from tubers grown from seed the previous season were used with like results. The potato seedlings can, however, apparently be infected with the viruses, so 8 Wisconsin Research Bulletin 63 that they are not entirely resistant to it. Nothing is known, however, about the source of the seed used in these experiments w-ith respect to variety re- lationship or to age. This phase of the problem needs further investigation as bearing on the origin of the viruses. It is evident from experience here with the Rural variety that potato varieties may be expected to dififer in this respect and because of the heterozygous nature of potato seedlings they may be expected to be very variable in their infectious properties. Further- more, it has been shown that the viruses obtained from potatoes are not resistant to desiccation, and that if they ever do exist within the seed coat they are probably destroyed by aging. The experiments preferably should be repeated with as young seed as possible secured from the Triumph variety. Extracts from eighteen different species of healthy plants, mostly of the sclanaceous family, have been inoculated into tobacco without yielding any symptoms of disease. So far as known, therefore, the potato is the only plant which produces infection on tobacco and other solanaceous plants when apparently healthy plants are used as a source of inoculum. In this category of "apparently healthy plants" there has not been included, of course, known susceptible hosts of mosaic diseases which may show no apparent symptoms on account of masking or other circumstances. Other Host Plants The diseases produced on tobacco from healthy potatoes are not specific for tobacco. It is believed that a number of widely different solanaceous plants might have replaced tobacco in these experiments. Infection, in fact, has been secured with one or more of the viruses in question on all of the solanaceous hosts tried. These include eight distinct varieties of Nicotiana tahacum, twenty-two species of Nicotiana and eight other species of the Solanaceae, namely tomato ( Lycopcrsectim escuJcntum) , Physalis, (Physalis pnbescens) , egg plant (Solanum nielongena), black night-shade (Solatium nigrum), jimson weed (Datura stramonium), pepper (Capsicum annum), petunia (Petunia violaceae) and buffalo burr (Solanum rostrafiim) . A marked variation in symptoms naturally occurs on the different hosts. "Spot-necrosis" is considerably more malignant on certain other hosts than on tobacco, and the "ring-spot" virus apparently shows no typical symptoms on such hosts as tomato or pepper, only mottling or general necrosis oc- curring. TransmiftAion Back to Potato In the earlier experience with the viruses secured from healthy potatoes, there was some reason to believe that the infection secured on tobacco might be potato mosaic, or some similar known virus disease of the potato existing either in the masked state or in a prolonged incubation period. It was con- sequently important to inoculate these virus diseases back to the potato from tobacco, in comparison with inoculations to potato from ordinary potato mosaic, as it commonly occurs on the Triumph Tariety. Transmission of V^iuusks 9 It was believed at this time that the potato would prove to be very difficult to infect with the virus of ordinary potato mosaic and that a prolonged incubation period would be required, with the result that current symptoms could hardly be expected. Consequently, it was decided at once to try to improve on the technique by varying the environment, with the hope of shortening the process. In this there was apparently success, since in the later trials 90-100 per cent infection was invariably obtained in 10 to 15 days with either ordinary potato mosaic or the new "spot-necrosis" type. The percentages of infection shown in the summarized results (Table II) are greatly reduced by low percentages of infection secured in the early experiments. The method used in these experiments consisted simply in placing the inociilated plants for a period of 8-10 days at a high temperature (27-32° C.) and then removing them to a lower temperature (17-22° C.) favorable at least for the expression of potato mosaic symptoms, and the general development of the potato itself. No doubt, this method can be further im- proved. Table II. — Summary of Result-; of Inoculations to Pot.\to (Tiuumph Variety) With Viruses of Type Secured From He.klthy Pota- toes IN Comparison With Other Mosaics Source of inoculum Number of plants inocu- lated Number of ijlants infected Per cent infection Symptoms Tobacco — "spot-necrosis" 80 65 81.2 Mottling and leaf drop, plants fre- quently killed •JO 7■^ 81.1 Di ease similar but not as viru- lent as above Tobacco — "mottle" 20 IC) One plant dead pr 0 b a b 1 y acci- dentally infected with '"spot-necro- sis" 20 0 0 None Tobacco — tobacco mosaic 30 0 0 Lesions on stems and petioles Potato^(Triuniph) potato mosaic. 40 25 62.5 Typical potato mosaic Tobacco — healthy foliage ,50 0 0 None Controls — no inoculation 50 0 0 None As shown in Table II good artificial infection with ordinary Triumph potato mosaic was secured on the Triumph variety. These inoculations were made simultaneously with the inoculations from the three forms of virus diseases secured from healthy potatoes. With the appearance of the symptoms it was at once strikingly evident that the ordinary potato mosaic was an entirely different disease from any of those secured from healthy (or mosaiced) potatoes on lohacco (Plate V^I). The "mottle" and "ring- 10 Wisconsin Kesearcti Bulletin 63 spot" tvpes of disease apparently produce no symptoms on the potato, al- though this may need further vertification (Plate V). The "spot- necrosis'" disease when inoculated into the potato from tobacco, however, produces a diseased condition of such a serious nature that not frequently the plant is killed in fifteen to thirty days. Following the first general chlorosis of the basal leaves, leaf-drop, and mottling on the younger leaves, the plant may partially recover, leaving a tuft of curled and mottled leaves at the top. (Plate VII). Tuber formation, however, is almost pre- vented in most instances when compared with the uninfected controls. There is some similarity in this disease to some of the known virus diseases of the potato, as, for example, late stages of "streak" or "stipple-streak", and probably to "leaf-drop streak", the latter not yet apparently fully de- scribed in literature. Whether this disease is identical with any known virus disease of the potato reuMins to be determined. Eighty potato plants have been inoculated with "spot-necrosis" from tobacco and infection was secured in 81 per cent of the plants. In the three last experiments involving ten plants each, 100 per cent infection was secured. If now potato plants are inoculated with the virus secured from potatoes infected with "spot-necrosis", infection is readily secured, usually in a slightly smaller per cent of the cases, and the disease is decidedly less malignant than in the transfer from tobacco to potato. The virus in this case apparently looses some of its virulence while in the potato. This also can be shown by comparative inoculations back to tobacco. In three genera- tions on potato it has not lost all of its virulence, however, and whether or not it will ever do so remains to be ascertained. Neither healthy tobacco plants nor tobacco plants affected with ordinary tobacco mosaic produced any symptoms on potato, with the interesting exception that tobacco mosaic produces brown or black necrotic lesions on the stems and petioles of potatoes apparently at the points of inoculation. Tobacco mosaic infection was not found to be systemic in potato, however, .'lud it cannot therefore be said to be a tvpjcal host of tobacco mosaic. A Combination Disease An interesting phenomenon occurs when the viruses secured from healthy potatoes are combined with ordinary tobacco mosaic virus and inoculated into tobacco, tomato or certain other solanaceous species. This is especially true when the "spot-necrosis" form is combined w'ith tobacco mosaic. The combined effect of the two diseases is much more malignant than either disease alone, in simultaneous inoculations. If, for instance, the "mottle" type of virus from potatoes and ordinary tobacco mosaic are combined, irarked necrosis, in the form of leaf spotting, may occur on tobacco whereas neither one of these alone produce necrotic symptoms on tobacco. The combination disease on some hosts like tomato has been noted at times to be so virulent as to kill the entire plant. The relation of environment or i)ther circumstances to this effect is not sufficiently understood. It is interesting to note that fre(|uently the combination disease on tobacco TkANSM [SSIOX ()!• ViKUSKS 11 pnicc'OfK witli iicnotic effect along the midrib or tlie principal veins of the leaf. Tiiis cciiuiition may sometimes be found to a less striking extent, h<^\vever, witli a single virus when it possesses necrotic properties. The plants if imt killed exhibit a special tendency to reccjver from the first effects of llie ciinil)iiiati(in disease. Discussion of Results The txperinunlal ividence summarized is believed to be sufficient to warrant the conclusion that luost potato varieties uniformly possess the property of inducing a disease in tobacco and other solanaceous plants, which is infectious in nature and belongs to the class of filterable viruses. '1 his ability is present regardless of whether the potato is healthy, as this word is generally applied, or affected with one or another of the common virus diseases of the potato. It is not meant to imply, however, that none of the virus diseases now known to occur naturally on potatoes may not be tiansmitted to tobacco or tomato, although the evidence that such has been done in the case of potato mosaic, as reported by Schultz and Folsom' and yuanjer', appear doubtful in view of the results secured with healthy pi tatoes. In view of the wide host range of these viruses within the solanaceous family it is rather surprising if these do not actually exist in nature occasionally as specific diseases. While it has not yet been definitely proven, it is possible that the "ring-spot" disease of tobacco as it occurs in nature is identical with the "ring-spot" disease secured from potatoes on tobacco under experimental conditions. It would be rather peculiar if the disease referred to as "spot-necrosis" on tobacco, which attacks potatoes with such \iruience, is not found to occur in nature on potatoes. It is difficult to r.nderstand, on the other hand, how potatoes can so generally harbor this \irus, without any apparent symptoms being expressed, assuming that this is actually a virus disease of the potato. In this connection it should be pointed out that the previously existing evidence of true virus carriers (meaning infected plants in which symptoms of any sort are never ex- pressed) is extremely meager, and that which does exist needs further verification. The potato has not yet been put into the category of a "carrier", although it may eventually prove to be an excellent example of ll)is phenomenon. If, on the other hand, it is assumed that the viruses secured from healthy potatoes are not actually present in the potato as a true virus, but merely as normal (or possibly abnormal) protoplasm another hypothesis, possessing many advantages from the standpoint of explaining what is now known about virus diseases, is at hand. It is not the purpose of this bulletin to present the apparent evidence one way or the other on this subject, h'urther experimental data must be secured to establish either of •'Schultz, K. S., ami Folsom, IJ. Transmission, Variation and Control of Certain Degeneration Diseases of Irisli Potatoes. lour. Agr. Research 25, p. 43-117 (Plates 1-15). 1923. ■•Quanjer, H. M. General Remarks on Potato Diseases of the Curl Type. Report Ir.ternat. Conf. Phytopath. and Ec. Ent. Holland, p. 23-28 (Plates I-IV). 1923. 12 Wisconsin Reskarcii 1)Ulij:tix 63 the hypotheses that liave hcen presented to explain the results obtained with healthy ; .atoes. By attacking the problem from new angles it is hoped that this may eventually he accomplished. SUMMARY 1. — 1'hc summarized results of inoculalinns from potatoes which are healthy as far as can be determined are believed to be sufficient to show that at least two different viruses are commonly, if not universally, present in most standard varieties of potatoes. 2. — The diseases ])r()(luced on tobacco are infectious and are characteristically of three types, which are referred to as "mottle", "spot-necrosis" and "ring-spot." The two former may be differ- ent expressions of the same disease. An increase and decrease in virulence of these forms is a])parcnt as they are transferred between hosts. 3 — ^rhe properties and nature of the viruses are similar to those of other well-known \irus diseases of plants, with respect to filtration, dilution, insect transmission and resistance to desicca- tion, putrefaction, heat and cheinicals. 4. — ( )ne or more of the viruses can be transmitted readily to a large number of different species of jjlants of the solanaceous family. 5. — The "spot necrosis" form can be transmitted l)ack to the po- tato where it causes a virulent disease, the "mottle" and "ring- spot" forms apparently give no s\-mi)toms on jjotato. 6. — No olher s])ecics of healthy i)lant h:is been found of which the extract will induce s\mptoms of any kind in tobacco. Potato foliage from true seed has also failed to give any definite infection on tol)acco. 7. — ( )r(linar\- tobacco mosaic combined with the virus from liealthy jjotatoes results in a combination disease with striking necrotic effects. 8. — The experimental results indicate that potatoes are either "true carriers" of viruses, or that potato protoplasm is actually the causal agency of one or more of the virus diseases of tobacco and other solanaceous plants. Research Bulletin 64 July, 1925 Pea Disease Survey in Wisconsin F. R. JONES and M. B. UNFORD Agricultural ELxperiment Station of the University of Wisconsin Madison Contents Introduction 1 Pea diseases and the canning industry 1 Studies of pea diseases in Wisconsin 1 General plan of the 1924 survey 2 Field notes 3 Records of cropping history 3 Rootrot caused by Aphanomyces Description of the disease 4 Environmental conditions controlling infection 7 Climatic conditions of 1924 in relation to disease 7 Occurrence and importance of rootrot 8 Relation of rootrot to number and frequency of previous crops of peas 9 Relation of rootrot to soil type and drainage 13 Relation of soil reaction and fertility to rootrot 15 Rootrot in first crop of peas 15 Persistence of the parasite in the soil 17 Resistant varieties and date of planting 17 Control of rootrot 19 Less important pea diseases Fusarium stem and rootrot 19 Footrot, a disease resembling Fusarium stem and rootrot .... 20 Seedling injury caused by Rhizoctonia 20 Seedling and root injury caused by species of Pythium 21 An undescribed wilt disease 22 Leaf and podspot or "blight" caused by Ascochyta 23 Leafblotch caused by Septoria 24 A Septoria leafspot new to Wisconsin 24 Anthracnose 25 Downy mildew 25 Bacterial blight 25 Mosaic 26 Summary 28 Literature cited 31 Pea Disease Survey in Wisconsin' AVOIDANCI^ of (liscase i.s a major consideration in the production f)f peas, whether by the canner. the seedsman or the market gardener. Each of these growers when entering new territory has usually found it to his advantage to grow peas repeatedly on the same ground, and each of these, outside of certain irrigated territory, has observed sooner or later that peas failed to grow as successfully as formerly on his older IK-a fields, and has been obliged to remove the crop to new ground or to adopt a long rotation. The rea.son for the failure of peas in old pea fields seems always to have been due to pea diseases which have increased rapidly with the intensive culture of the crop. Thus it has come about that pea diseases have largely determined the cropping practice wherever peas have been grown for a considerable time. This is especially true in the production of peas for the canning factories in Wisconsin. Pea Diseases and the Canning Industry From the time of estal)lishnient of the first canning factory in the state ir. 1889 until about 1912, nearly every company that entered the business owned land upon which it grew at least a part of the peas which it canned. In certain instances, as many as ten successive crops are said to have been grown on company owned land. However, conspicious crop failures upon land that had grown repeated crops of peas experienced by some companies in 1910 and following years were called to the attention of the State Experiment Station and led to the beginning of an investiga- tion of pea diseases. The opinion was soon expressed that peas could not be grown successfully on the same ground indefinitely as the companies owning the land had hoped. Following the advice of the Experiment Station, intensive growing of peas upon company owned land was gen- erally abandoned by 1915. Since that time, the most of the peas have been grown for the canning companies under contract by farmers with the supervision of the company field agent, though a few companies do their own farming on land leased for two or three years. Since this dis- persal of pea growing over a large territory around factories, disastrous crop failures from pea diseases have been less frequent, though they have occurred here and there where farrflers have been permitted to re- peat the mistake made by the companies earlier. Loss from disease is no longer the inhibiting menace to the industry that it was felt to be in 1912; and the canning business has grown until the number of plants in the state reached 135 in 1924, producing peas on about 102,000 acres. Studies of Pea Diseases in Wisconsin From the time of the first alarming failures of peas until the present, the department of Plant Pathology of the Experiment Station has been active in studying the several diseases which have been found contributing to these failures. At first, the conspicuous foliage diseases were regarded as the chief cause of loss, and effective control measures were devised and *By Fred Reuel Jones, Pathologist. Bureau of Plant Industry, Unittd States De- partment of Agriculture, and Maurice B. Linford. Industrial Fellow in Plant Path- ology, University of Wisconsin. The field survey was supported by members of the Wisconsin Pea Packers" Association. 2 Wisconsin Research Bulletin 64 generally adopted. When these foliage diseases subsequently declined in importance, it began to be apparent that in many fields root diseases were present which were not only able to destroy the crop effectually as the foliage diseases, but which were remarkably persistent in the soil. Further- more, it appeared that trouble of this kind was by no means a local problem, but that it was being encountered by growers in many parts of the United States. Thus it came about that in 1919 the U. S. Department of Agriculture began an intensive study of these root diseases in coopera- tion with the Wisconsin Experiment Station. In this later stage in the study of pea diseases several root diseases have been distinguished and described, but one of these appears to be so much more important than the others that it will undoubtedly become generally known as "the rootrot disease" of peas. This dis- ease can be distinguished in the field without great difficulty, and considerable information regarding the behavior of the fungus causing it has been gained. In fact, the study of this disease had reached the stage in 1924 where new and very specific recommendations for its prevention could be made. The new recommendations, however, made it necessary that the field agent of the canning company should be able to recognize this rootrot with certainty, not only in cases of conspicuous crop failure, 1)ut in less conspicuous beginnings. Thus the time had come when any as- sistance which could be given the field representatives of the canning companies in acquiring experience and skill in recognizing the disease was of financial value to the industry. GENERAL PLAN OF THE 1924 SURVEY Although the study of the root diseases of peas had reached a stage at which practical suggestions for control measures could be made, it had by no means reached a satisfactory conclusion. The relative importance of the several diseases recognized had not been adequately studied in the field ; an important effect of soil type on both the increase and persistence of disease was suspected but not thoroughly examined ; and the value of resistant varieties had not been tested. Further study of any of these things required wide field experience. This situation presented to the pea canners of the state early in 1924 led to a field inspection for disease in a part of the territory of 30 canning companies at their own expensed This bulletin is a fuller report of the findings of this survey than that given in a previous circular (11) together with a description of diseases of peas occurring in Wisconsin, and summaries of fragmentary studies by the senior author of some of these diseases. The first purpose of this survey was to bring assistance to the canning companies in recognizing rootrot in their territory, and especially in detecting it before it had become destructive in order that they might avoid losses which would come from replanting infested soil in the near future. ' The field studies reported in this paper were thus more or less secondary to this major purpose and were in some respects limited by it. In the course of the survey, 688 fields were visited comprising 5,416 acres, or approximately one-sixth of the total acreage of thirty-seven' factory districts. Nearly all these districts were visited twice in order to 'The 1924 Wisconsin pea disease survey was financed by 30 canning companies as listed in a previous circular (11), operating in 35 factory districts. Through W. E. Nicholoy, Business Secretary of the Wisconsin Pea Packers' Association, these companies subscribed from 300 to 600 acres each for inspection, paying fifteen cents per acre for the service. The fund obtained in this way was administered by the University under the general supervision of T. R Tones in consultation with R. E. Vimghaii. Pea Disease Survey in Wisconsin 3 examine both early and late plantings at the stage most favorable for inspection, and a few which contained considerable rootrot were returned to after the close of the canning season to obtain records of yields from diseased fields and to complete other important records. Field Notes Individual records were made of each field inspected (fig. 1). In most cases diseases both of foliage and roots were determined and recorded in the field, but in frequent doubtful cases samples were collected for microscopic examination or for making of cultures. PEA DISEASE SURVEY Company No Farm Owner Address Dale Variety Planted Area Stage Soil Type Roolroi y^ infestation FIELD HISTORY 1923 1922 1921 Ascochyta Colletotrichum Septoria Bac. Blisht Downy Mildew Powdery Mildew Fusarium Pythium Rhizoclonla Mosaic Nodule* Aphids FIG. 1- DATA INCLUDED ON FIELD SI RVEY CAHDS The soil type was determined from soil maps of the Wisconsin Soil Survey in mapped counties. In counties not mapped a few of the more common series were readily identified, but in some cases the soil class only was recorded. The classification and nomenclature of the Soil Survey are followed in this bulletin. Records of Cropping History The most necessary information to be secured regarding each field visited, and the most difficult information to secure with adequate accuracy was its previous cropping history, and especially the years in which all previous crops of peas had been grown. A cropping history was regarded as satisfactorily complete when the dates of all previous crops of peas were secured, but even this .limited field history was often lacking and' data of great value lost. In this connection, it may be pointed out that while the survey aimed to 'Data are included in this paper from two factory districts in addition to the thirty-five subscribed. 4 Wisconsin Research Bulletin 64 cover fields in which peas had previously been grown in preference to others, nevertheless records at the end of the season showed that 48 per cent of, all fields visited had never grown peas previous to 1924. The percentage of fields visited which were growing their first, up to their sixth crop of peas is shown in Fig. 2. Apparently at least half, probably much more than half, of the peas grown in Wisconsin in 1924 were grown on ground new to peas, and only a very small acreage had ever grown as many as two previous crops. A smaller acreage on land new to peas will necessarily be found in succeeding years. Per cent OF Fields ]0D 75 50 23 ^ -^ 1 2 3 4 5 6 or more Number of Crops of Pfas Grown {i92^ includbd) FIG. 2-DISTKlBUTION OF FIELDS ACCORDING TO NUMBER OF CROPS OF PEAS GROWN Showing the extent to which inspected peas were glowing in fields new or relatively new to peas. ROOTROT CAUSED BY APHANOMYCES Description of the Disease Inasmuch ai- this survey was designed primarily to discover the re- lation of rotation and soil conditions to the occurrence of rootrot it is necessary to describe in detail the disease and the fungus causing it. Rootrot is primarily a fungous soft rot of the primary cortex of the roots and cpicotyl of the plant (fig. 3), extending one or two inches above ground under humid conditions. Plants are susceptible at all ages, and the disease may begin at any point or at many points simultaneously' in the root system. Lesionf are at first water.soakcd areas yellowish to straw colored, especially on the epicotyl. From any point of entry, the fungus spreads rapidly in all directions through cortical tissue until that tis.sue has been completely decayed. Although the causal fungus does not I'i:.\ DisfciASi-: Sukvkv in Wisconsin FIC. 3— HEALTHY PLANT (A) L\ CONTRAST WITH PLANTS DAMAGED BY ROOTROT (Bi. Roots of diseased plants, and stem to just above the surface of the soil, are softened, darkened and shriveled. Stern, roots and nodules of the healthy plant are plump and white. Severe root injury has caused the death of the leaves of the plant at the right. (Photograph courtesy of U. S. Department of Agriculture.) 6 Wisconsin Rkskarch Bulletin 64 penetrate the endodermis of the mature root, it does cause the death of the meristematic tissue at root ends. Thus root growth is stopped. In the older roots the decay of the cortex exposes the endodermis to the attack of other soil inhabiting fungi. In a few resistant varieties a protective secondary cortex is formed from a cambium developing in the pericycle ; but in most varieties this has not been found to occur, and the endodermis does not appear to be an effective barrier against all invaders. In the epicotyl, unprotected by an endodermis, the vascular tissues are readily entered by several species of Fusarium which may quickly kill the plant. The effects of the disease upon the plant as a whole are not definitely characteristic, but depend largely upon the stage of development at which infection takes place. If the attack comes early in the development of the plant while the root system is small and incapable of supporting a large vine, the vines are stunted or killed immediately. If severe injury to the roots is delayed until the roots have attained nearly their full extent and if abundant moisture is never lacking from the soil, the plant may appear nearly normal, and mature to the canning stage at least a large fraction of a normal crop. Thus this disease is not always readily discernable in the field. When conspicuous injury to the vine is absent, it is necessary to examine roots to find the disease. When these are dug. the softened shrunken condition of the cortex of roots and base of the stem is usually readily observable. Sometimes when diseased plants are pulled the vascular cylinder of the tap root pulls out readily from the decayed cortex as a long string, while roots of healthy plants almost always break at the attachment of the seed. In cases which are not readily determined from superficial examination it is always possible to discover the character- istic spores (10) of the causal fungus in the decayed cortex by a micro- scopic examination. Since the parasitic fungus causing rootrot has been described fully in a recent paper (10) only the more important details of its life history need to be repeated here. The fungus, Aphanomyces euteiches Drechsler, is one of the few exceptional species of this genus of the Saprolegneaceae not strictly aquatic in habit. It is, however, dependent upon a period of submergence in water for the production of its asexual spores. It is most readily discernable in the host tissue as subspherical oospores 18 to 25 microns in diameter, surrounded by oogonial walls of unusual thickness. The mycelium in the host tissue is abundant, intracellular for the most part, but ephemeral in a living condition, since the contents are rapidly transferred to the abundant oospores. This mycelium is not readily dis- tinguished from that of other Phycomycetous species, especially of Pythium, which frequently accompany it. The oospores occurring in the host tissue have not been germinated, but those produced in culture germinate readily giving rise to non-spetate mycelium on a moist substrate or in water having sufficient nutrient material. In pure water they give rise more or less directly to zoospores. These zoospores come to rest after a period of motility and germinate giving rise to mycelium. The mycelium when young may function more or less completely without apparent differentiation as sporangia under aquatic conditions giving rise to numerous zoospores. It has not been possible to follow the life history of the fungus in nature. No evidence of conveyance with pea seed has been found. It is believed that the abundant oospores formed in the host tissue persist for a long time in the soil. Mycelium from germinating oospores is able to penetrate pea roots, and it is possible that zoospores formed when the soil is filled with water may also be a source of infection. It is also possible that the mycelium may persist for a time in the .soil as a saprophyte. No other plant than species of Pisum are known to be invaded by this fungus. Pea Disease Survey in Wisconsin 7 Environmental Conditions Controlling Infection The time of appearance of the disease in the field appears to be controlled chiefly by the temperature of the soil. Although laboratory study shows that the oospores may germinate giving rise to zoospores at a temperature as low as 9 to 10° C, very little infection of pea plants has been found either in controlled experiments or in the field until a temperature of 15° C. has been reached. The disease develops rapidly when soil temp- erature is between 15° and 30° C. The effect of soil moisture upon the development of the disease in the field appears to be important. Although greenhouse experiments have Degrees Centigrade On. A^£ /^ \ V, /.IT 4 /1 hi '\ ij V /-' * f H /il y Y / w C I ( 1 It lb 2\ 2b 3/ May W /5 20 25 30 June fig. 4— mean daily soil temperatuke at a depth of 2 inches Recorded at Madison, Wisconsin, from May 1 to June 30, 1923 and 1924. failed to show that there is any necessary relation between soil moisture and the severity of disease, yet in the field infection seems to be more prompt and abundant after heavy rains. Thus in the field the disease can be expected when rains have thoroughly wet the soil after it has attained a daily average temperature of 15° C Climatic Conditions of 1924 in Relation to Disease The climatic conditions during the pea growing season of 1924 presented peculiarities whicli need to be considered in interpreting the results of this survey. \ April, May, and June were cold and cloudy with an unusual number of rainy days. Planting was begun early in April but it was not com- pleted along the shore of Lake Michigan until June 13, and the peas planted early spent an extradordinary length of time between planting 8 Wisconsin Research Bulletin 64 ^nd harvest. July and August were both cool, and in the major pea sections of the state were marked by frequent rains. August 1924 was the wettest August in the climatological history of Wisconsin, and heavy rains in the fore part of the month caused great damage to the peas which were not then harvested. That low soil temperature restrained the develop- ment of rootrot until later than usual is indicated by a record of soil temperature kept at Madison during this season. The record for 1924 is charted with a similar record made in 1923 in Fig. 3. This figure shows that while soil temperature in 1924 averaged somewhat higher than in 1923 up to May 24, it was not high enough to permit of much infection by Aphanomyces to that date. After May 24, soil temperature was cooler in 1924 than in 1923. Not until June 10 did the soil become permanently warm enough to favor the development of the disease, whereas, that condi- tion was reached June 1 in 1923. From this and from other meager soil temperature records at hand it appears that the disease was restrained from development in 1924 until later than usual. Continued low tempera- ture through June was unfavorable for other soil fungi which often complete the destruction of plants injured by Aphanomyces. Thus cool weather probably accounts for the less destructive character of root- rot in 1924, — a condition believed to be a fact by several observers. Occurrence and Importance of Rootrot During the 1924 survey the njotrot of peas caused by Aphanomyces was found to be far more important than all the other parasitic dis- eases combined, causing losses amounting to approximately 8 per cent of the yield of the total acreage inspected. Of the 688 fields aggregating 5,416 acres inspected, 222 fields or 32 per cent were found to contain the disease in greater or smaller amounts. The infested territory was b\' no means distributed uniformly through the several districts. Two of the 2)7 districts appeared to be free from any traces of it while in one district 62 per cent of 366 acres examined were more or less thoroughly 68^ NO DISEASE 1 IG. .5— ROOTROT IN SURVEYED FIELDS Diagram showing the percentage of all fields examined which contained llio anioinits of rootrot indicated. (Sec discussion, page 9). infested. In general, the younger, canning districts were freer from disease than the older, though a few older districts had rotated the crop so carefully or had removed the crop to new ground so completely that little disease was found in them. I'^ields recorded as containing rootrot had widely different amounts from a mere handful of plants to complete infestation. In each case an estimate was made of the area infested as a certain percentage of the entire field. Pea Disease Survey in Wisconsin 9 The infested fields have been classified for convenience as follows: fields showing from a trace to 25 per cent of the area infested are regarded as having "light" infestation ; those having from 26 to 75 per cent are called "medium", and those with from 76 to 100 per cent are regarded as having "heavy" infestation. The result of this classification is shown praphically in Fig. 5. Of all fields surveyed, 68 per cent had no rootrot, 15 per cent had light infestation. 6 per cent had medium and 11 per cent had heavy infestation. The fields in which rootrot occurred were not always damaged in direct proportion to the amount of rootrot present for reasons which are discussed later. A more accurate picture of the extent and distribution of losses T.\BLE I. — Rootrot-Infested Acreage Classified According to Esti- mated" Reduction in Yield Due to Rootrot. Reduction in yield Total arrcage Percentage of total rootrot-infested acreage Percentage of total inspected acreage 0-5% 1228 57 22.7 6-25 % 356 16 6.6 26-50 % 186 9 3.4 51-100% .379 18 7.0 •The heavier losses were estimated by comjiaring actual yields from diseased fields with average yields from disease-free fields of the. same variety in the same locality. can be gained by grouping infested fields into classes based on the extent of crop reduction. Such a classification is given in Table 1. From this table it appears that in over half the infested acreage the loss was negligible but in 18 per cent of the infested acreage loss amounted to from 50 to 100 per cent of the crop. Relation of Rootrot to Number and Frequency of Previous Crops of Peas That the occurrence of rootrot is closely correlated with the number and frequency of crops of peas in a given field is one of the most widely recognized characteristics of the disease. An exact statement of this correlation, and a study of its characteristics were among the first objects sought in the survey. The correlation is very striking when the survey records as a whole are considered. There was almost no rootrot in fields which had grown no peas before, and a rapid increase in the frequency of its occurrence was found with each succeeding crop whether these crops were in succession or at short intervals. Increase in percentage of fields diseased. — Neglecting for the moment the interval that had elapsed in some cases between crops of peas — the records show that the interval was rarely long enough to be important — the fields may be divided into groups based on the number of crops of peas which they had produced. When this classification has been made, fields growing the first recorded crop had rootrot in but 8 per cent of their number, while every field growing the fifth crop had rootrot. The percentage of fields showing rootrot in each class is shown in the upper line of Fig. 6. The increase in infestation with each succeeding crop of peas rises regularly almost as a straight line from 8 per cent in the first crop to 100 per cent in the fifth. Since the last two classes are small, it must not be assumed that this curve presents an altogether accurate 10 Wisconsin Research Bulletin 64 picture of the rate of appearance of the disease under all conditions in the state. However, it can probably be regarded as an approximately correct average. One characteristic of the correlation between rootrot and the number of crops of peas grown in a field which has appeared in the records obtained this year is not shown in this presentation. This important characteristic is best seen when the fields thoroughlv infested with rootrot — those in Percent OF Fields too / Fields tj^orou^^ hor more Number of Crops of Peas Grown(/92^ incluped) FIG. 6— INCREASE IN PERCENTAGE OF FIELDS CONTAINING ROOTROT, AND FIELDS THOROUGHLY INFESTED, WITH INCREASE IN NUMBER OF CROPS OF PEAS GROWN which most of the larger losses occurred — are classified and plotted in- dependently (fig. 5, lower line). When the diagram produced in this way, showing the relation of thorough infestation to rotation, rather than the presence of the rootrot disease in any discoverable amount, is examined, its character is found quite different from the former figure. This curve rises very slowly until with the fourth crop of peas only 17 per cent of all the fields are in this class; but in the fifth year the percentage suddenly rises to 56 per cent of all fields in the class. Stated in another way, these figures suggest that under average field conditions in the territory surveyed it is possible to grow on new fields four .successive or nearly successive crops of peas without great peril from large loss from rootrot ; but that with the fifth crop there is about an even chance that the field will be severely damaged, and that with the sixth crop there is but one chance in four of a profitable yield. Increase of rootrot in average Wisconsin field. — TItc field records used in the preparation of the diagram previously discussed may be presented in a different manner that has some interest, even though it is not as clearly and directly significant to the pea grower. If all the fields covered in the survey are classified as previously into groups based on the number of the present crop in the field, and the figures representiiiL' Pk \ DisKASi-. Si'RVKV I x \\'is((i\m\ 11 the i)t'rccntat(c oi rontrot uiiOiatioii in the tieltU nt each group are average and plotted (fig. 7.), a diagram is produced which may be regarded as showing the percentage of the area of an average Wisconsin field invaded by rootrot with each succeeding crop of peas. This diagram is so similar in character to that of the preceding that it requires no added discussion. Influence of rotation. — Thus far in this discussion of the rate of ap- pearance of rootrot all the fields covered in the survey have been discussed without regard to any rotation that may have been practiced. This is justified by the fact that only in exceptional cases has rotation long enough to be significant been recorded. It now remains to examine these exceptional cases to determine whether they give any indication that rota- tion in any form delays or averts the appearance of the disease. From what has been said previously regarding the recent appreciation of the Percent OF Fields /OOi 75 50 2 3 4 5 Oor . more Number of Crops of Pfas Grown {I9z^ includf.d) —RATE OF INCREASK OF ROOTROT INFESTATION IN THE AVERACiE WISCONSIN pf:a field as indicated in the i!»2i survey necessity for rotation in pea growing, the youth of a large number of the canning districts covered, and the lack of adequate records of cropping history, it is not surprising that the number of significant cases found in ;i single year is too few to form the basis of conclusions. By way of comparison, it may be stated that only three fields growing their fourth successive crop of peas were found disease-free, while two fields which had been cropped in essentially a three-year rotation were disease-free in their sixth crop. Other instances of three-year rotation showed convincingly, however, that this is not long enough under other conditions to prevent the entry of disease even up to the sixth crop. Only a few instances of four or five-year rotations were found, and these were started so recently that they give no indication yet of their effectiveness over a long period of time. The extent to which care in rotating peas is avoiding loss from rootrot lunipaiu AABsq uinipaui o «»; ^qgl| ao AABSq uinipaui iqsir AAB3q uinipaiu >H8TI AABaq uinipaui 3 « O 4) [I. a ^qgil AABaq Q. a; uinipatu iqSii AABaq uinipaui ^^s^ ^Tf— --^ XABaq lunipaui 1M8i| N -i-llOCO NC^C0iCC-iO5 E CO — s ! — C S-E ra— sr= ffl c — —. '-^-o ■ c n ■ 5 o g co- co c C CO .— o — j-.-V III CO P ( = O 2 — t-T ' CO -,— ji , ■ E • r- ; CO c c ■ n, ^ m iS c O CO OB »-- ■S'e ^' co;5 o ^^L CO ^ CO I- CO g CO o^:: ■ C ^' iS^SiHE cj-a >,o CO in 5 CO , CO CO— CO O , CO « O tn trt— O) « cS c o— C o o;— o OMiWocojn^-cOtn 'u.S re'p £ co'^.^._ Cj CO ^ a'~ £ CO g^ ra -ra c « o C — ^ c sc: CO — f/5 •0:5 "• ■3 Ci 1; Pea Disease Survey ix Wisconsin 1.3 ill present practice may best be illustrated by comparing two adjoining factory districts (jn the same soil types and comparable in every respect. One of these districts operating in its thirteenth year was established by a company which had previous experience with the disease, and which had avoided planting peas repeatedly. In this district only three fields showed rootrot, and that in small amount. Two of these three fields were grow- ing their third crop of peas, and no field in the district was growing more than its third crop. The loss was negligible. In the second district, operating in its twelfth year, repeated planting of peas on the same ground had not been avoided. In this district 18 fields were found with rootrot, half of which were severely infested with heavy losses in five. All of these five fields were said to have grown many crops of peas be- fore, one having produced at least seven crops in ten years. Relation of Rootrot to Soil Type and Drainage Earlier observations. — In a previous paper (10) some scattered observa- tions have been recorded showing that in different localities where peas have been grown intensively in a similar manner, there is great difference in the time which has elapsed before the disease has made its appearance in fields, and also in the rate of its spread and increase in destructiveness. These differences appeared to be associated with differences in the capacity of soils to hold water, or with drainage and sub-irrigation. For instance, in Wisconsin the Superior red clay appeared to be more subject to severe injury from rootrot than contiguous loams. Some sandy soils in Maryland underlaid by impervious clays seemed remarkably favorable for the development of disease — an observation which seems to be supported by more recent observations by Drechsler (3). In irrigated districts of the Rocky Mountain States, peas on soils of low moisture-holding capacity rarely suffer from disease unless subirrigated, though on some of them occasional diseased plants can be found. There are several ways in which the water relations of soil might affect the development of disease in peas. The mvjst apparent of these is the favorable environment which abundant water in the soil might provide for the semi-aquatic parasite causing the disease. If, as has been assumed tentatively in this paper, the fungus is widely distributed in soils, it may be originally much more frequent in wet soils than in those which do not retain water. Studies in 1924. — Whatever cause or causes give rise to the observed variatii n in the behavior of rootrot, it was clearly of great importance in this survey to determine to what extent the several pea growing soils of Wisconsin do affect the behavior of this disease. Any differences which might be found would not only affect the cropping systems which must be used on the several soils to avoid disease, but might affect the direction of expansion of the industry. The method of classifying soils found most suitable for this study is that provided by the Wisconsin Soil Survey. To a considerable extent, soil types as distinguished by the soil survey are representative of a certain degree of drainage. Some entire series are characteristically well drained. Others are poorly drained. Within each soil type there are, however, many relatively minor, but still, from the point of view of this study, important differences in drainage. Fields of uniform soil type. — In the course of the survey, peas were examined on 27 distinct soil types besides seven groups of incompletely classified soils, making in all 34 groups into which the total of 688 fields are divided. There are, therefore, too few fields in many of these groups to afford a satisfactory basis for comparison. A complete summary of the data obtained is given in Table II. From this complete table, one important conclusion can be drawn. None of the soil types encountered shows any promise of furnishing an environ- 14 W ist oxsix Research Bulletin 64 ment where peas may be grown without danger from rootrot. Diseased fields are recorded on all but eight of the types. The eight exceptional types are represented by so few fields that the absence of disease in the location where they were found can not be taken as evidence that they are naturally less liable to disease than others. For instance, of the fifteen fields on Superior silt loam, only three were growing their second crop, and these were in a relatively new canning district in situations where disease would hardly be expected. The 13 fields on Fox silt loam were in a district producing its fifth crop of peas but none of these were growing more than its third crop. Under these circumstances, it may be said that this soil type has a more promising record indicating freedom from disease than any other. In contrast with the Fox silt loam, the soil type which shows the most unfavorable record with reference to rootrot is the Colby silt loam. Since only 20 fields were encountered on this soil, its present record with reference tc rootrot should not be regarded as convicting it of being the most favor- able soil for disease in the state. The table shows, however, that of the nine fields found which had grown one or more crops of peas previousl\ . all were diseased to a greater or less extent. This soil is characteristically compact with poor internal drainage and in many cases with faulty sur- face drainage. In view of the tendency of the canning industry to expand on to this soil, a more comprehensive examination of the behavior of the crop on this soil should be made. The two leading soil types. — The two soil types upon which sufficient numbers of fields were found for adecjuate comparison are Miami silt loam and Carrington silt loam. These show no important difference in behavior (Table III). The Carrington silt loam shows a larger percentage of infested fields, but a lower percentage of fields extensively invaded. When these two soils are compared with the total number of clay loams and clays summarized in the same table, it appears that the heavier soils show both greater percentages of total fields infested and of fields thoroughly infested. Since the average cropping histories of the fields on the heavy soils is not markedly dilTerent from that on the silt loam, the comparison appears to demonstrate a greater tendency for rootrot to become trouble- some on the heavy soils. TaBLH III. COMFARISOX OF C.XHHINGTON SiLT LoAM AND MlAMI SiLT Loam, the Two Chief Pea-Growing Soils of Wisconsin, With THE Total Clay Loams and Total Clays as to the Percentage of Fields Found Infested With Rootrot and the Percentage OF Fields Showing Light, Medium, or Heavy Infestation. Soil Total fields Number of fields with rootrot Per cent fields with rootrot Percentage of total fields (I Light nfcstation Medium Heavy C.arrinston .silt loam . . . Miami silt loam ^I'otal rla\ loams 180 7fi 51 57 :i7 :i3 21 .32 2.5 43 41 16 8 17 16 6 3 12 8 9 12 14 18 Comparison of soil classes. — In an attempt to condense Table 11 in significant manner preserving the summarized cropping histories of fields. all sands and sandy loams were placed in one group, all loams and silt loams in a second group, and all clays and clay loams in a third. Table IV. prepared in this way. reveals differences in behavior between the medium light and the heavy soils. In the group of clay loams and clays it appears that rootrot makes its entry more promptlx- and spreads through the field Pea Disease Survey in Wisconsin 15 more rapidly than in the silt loams and loams. All fields on clays and clay loams growing their fourth crop were diseased, while 29 per cent of such fields on loams and silt loams, and 40 per cent on sands and sandy loams were still rootrot free. An attempt has been made to condense Table II by grouping together all fields in the same soil series. The groups thus formed appear to be too small in most cases for satisfactory comparison. Rootrot in uneven fields. — The conii)arison of the behavior of soil types presented in the foregoing tables does not emphasize differences so much as examination of individual fields extending over two or more soil types, or fields not uniform in drainage. Fields of one soil type, with poorly drained portions, or of two or more types differing in tendency toward wetness were found in almost every district. In such fields, as a general rule, whatever the cause of the wet spots, whether lack of drainage, or seepage, or the texture of the soil enabling it to hold water, rootrot appeared to have entered the field first in these wet spots. Many of the apparent exceptions were probably due to accidental introduction of the rootrot fungus. In determining the presence of disease on very wet ground, microscopic examination was frequentlj' used to distinguish rootrot from drowning of roots from standing water. These general field observa- tions emphasize more than do the tables the differences between soils types, and at the same time strengthen the suggestion that the observed differences between heavy and light soils may be associated fundamentally with the natural wetness of such soils. Relation of Soil Reaction and Fertility to Rootrot No special study of the relation between soil acidity or fertility and the occurrence and destructiveness of rootrot was made in this survey. Neither in previous field experience, nor in that gained in this survey has it been obvious that any important correlation exists between these field condi- tions and the disease. There are. to be sure, a number of instances cited among growers in which high acidity and low fertility have been correlated with destructive occurrence of disease ; but when these have been examined they have not provided convincing evidence that correlation with either of these conditions was the essential factor causing loss. On the other hand, certain canning companies have attempted to render diseased land suitable for pea growing by carefully conducted liming and manuring experiments, but have failed completely. Rootrot in First Crop of Peas As indicated in figures 5 and (> a small percentage of fields was found infested in what appeared, from incomplete cropping records, to be the first crop of peas. Many of these had probabl\- grown some unrecorded crops earlier. There arc, however, eleven fields in which adequate records indicate that no peas have been grown before, and in which rootrot infesta- tion was found ranging from a mere trace up to 100 per cent of the field. A number of these fields clearly owed their infestation to inoculation from neighboring diseased fields in the following manner : two from diseased fields on the same farm where the diseases had long been established: one from surface drainage from an adjoining higher field ; one from an old barn yard included in the field ; one from manuring w^th uncured vines from the outside of a silage stack ; and one with root- rot along the roadside from passing loads of pea vines. The other five fields which contained rootrot in their first peas were all either poorly drained, wet soils, or contained the disease only in wet pockets. In addition to these there were a numlier of fields observed diseased in z o u c e h o ffi H o . O 73 2 z < ' Ui u o < I— I s. si: pa < 1^ as S 0) uinfpaiu )q8il AABaq oj'j^ lunipaui TqSlI XABaq Q, I uinipgiu iqSil 00C5 1 I^!*5 :^d «Oi AABOq 'o, i Lunipoiu -^ >!!<■ I T-. t^ 1—50 I CO t^ ^^m\ E ^ \.\BdH luiiipaiu iqgJI .)UOU AABOq lunipaiu I ^]l^ I — I :^I t^ I M .■^l 1 00 ;i I C<1> «5D I t^;o f3;o lO^ rl "-C I CD> C^sO I 0»5 s§ 5^ aS i '^^ Pea Disease Survev ix Wisconsin 17 their second crop which were said to have been diseased in their first. In these, as in the preceeding cases, the disease was mostly in wet soil. If Aphanomyces euteiches, occuring native in Wisconsin, was the source of infestation in these fields, it appears that it was originally restricted to wet locations. Persistence of the Parasite in the Soil Once peas have failed from rootrot and their decaying roots have released into the soil myriads of thick walled oospores, the parasite is able to persist for a remarkably long time. In a previous paper flO) instances were cited in which peas had failed from rootrot when planted on a field in which peas had failed six years earlier. Certain canners have reported experiences which indicate a survival of the parasite for a still longer period. During 1924 a number of well attested cases were encountered in which peas were growing on fields in which the last previous crop had "blighted" presumably from rootrot from one to ten or more years earlier. Assuming that in all of these cases rootrot was the cause of the earlier blight — an assumption which is undoubtedly true in nearly all of the cases — these fields have been classified in Table V. on the basis of the number of years that have elapsed between the previous blighted crop and 1924. The fifteen fields on which peas had blighted within ten years were found to contain more or less rootrot, the majority of them being still heavily infested. Of nine fields blighted more than ten years ago, only three were still thoroughly infested, while five were, as nearly as could be ascertained by careful search in the field, entirely free from the disease. This indicates that rootrot infestation does actually tend to diminish gradually with time, but the length of time required to free contaminated soil is discouragingly long. It appears unsafe to replant peas on infested soil within a decade. In this connection it may be added that field observation seems to indicate that the disease persists longer in heavy wet soils than in soils less favorable for the advent of the disease in the first place. This rela- tion is not obvious from the table, however, and requires more careful records for its confirmation. Resistant Varieties and Date of Planting in Relation to Injury from Rootrot Although it has been shown in experimental trials that no variety of peas is completely immune to rootrot. and that among the usual commercial varieties there is little difTerence in resistance, as measured in experimental trials, nevertheless search was made for evidence indicating that the slight differences between varieties is of any importance in averting loss on infested soil. The most resistant varieties that have been found, the Horal and Rice's No. 330, did not occur in surveyed fields. It may be added, however, that in a trial conducted by the Columbus Canning Company in cooperation with the U. S. Department of Agriculture during the summer these two varieties showed far greater resistance than has been shown by any commercial variety in general use. A study of the resistance in commercial plantings is greatly complicated by the fact that the date of planting is a factor to be considered when comparing the damage sustained by different fields. For instance, the record shows that there was a slightly smaller percentage of fields of Alaskas and Winners diseased and that they seemed on the whole to suffer smaller crop losses than other varieties. However, the peak of the planting season of these two early varieties was from three to 18 Wisconsin Research Bulletin 64 four weeks in advance of that of the sweet varieties, a fact whicii undoubtedly accounts in large part, if not completely, for their lighter damage. Thus a comparison of a few selected fields furnishes more reliable evidence of varietal resistance than a comparison of the rcc;)rds of varieties as a whole. From a comparison of suitable fields it appears that the Green Admiral alone showed appreciable resistance. One of the older companies operating in an infested district plants Admirals on all fields suspected of harboring disease. On one farm in this district, a uniform 18 acre field of thoroughly T.\BLE V. — Persistence of the Rootrot Fungus in the Soil Fields Which Are Known to Have Grown "Blighted" Peas Arranged According to the Period of Years Since Peas Were Grown Last and According to the Degree of Infestation Found in 1924. Interval since "blighted" peas 1- tu C c Z 2> •ea E rs >> > a X 3- a o Z -4 DC yea c s ■S rs > a di X 5- 4> C o Z -9 ! J3 DC I'ea E ■3 rs > a .X 10 ai a o Z or op mc E 3 -3 re > a 1 1 2 1 3 2 3 Wabitsh sill loam 1 1 2 2 •> Total 1 3 2 5 2 2 5 1 3 infested Miami silt loam was planted to Alaskas and Admirals on the same da}-. When first observed on June 30 both varieties were thoroughly irJested to the extent of about 90 per cent of all of the plants. The Alaskas were beginning to die, but the Admirals showed no evidence of injury above ground. On July 22, the Admirals were ready to harvest. The vines were short, pods poorly filled, and leaves dead on the lower half of the vines. However, the six acres of Alaskas yielded 166 pounds of peas per acre, while the Admirals produced 2,111 pounds per acre, though quality was not of the best. Here, the Admiral seemed to demonstrate marked resistance. In another similar field suitable for comparison the same varieties were planted on different dates — the Alaskas on April 11, and the Admirals 11 days later. On June 30 the root destruction had advanced far in both cases. The Alaskas were filling pods, though about 25 per cent of the plants were almost dead, while the Admirals were but 12 inches tall appearing perfectly healthy. Alaskas yielded 1,800 pounds per acre while Admirals yielded 1,035 pounds. The low yield of the Admirals in this case appears to be due to the later date at which they were planted, permitting the disease to attack them at an earlier stage of development. A number of other less closely comparable instances add evidence in favor of the view that under certain conditions the Green Admiral pea has .some degree of resistance, enabling it to produce a fair yield under conditions \\-hich damage other varieties much more severely. Until the l^iiA Disease Survey ix Wisconsin 19 I'cwer resistant varieties become available the Green Admiral appears to be the only pea showing a sufficient degree of resistance to warrant its use- on infested soil, although this variety may fail utterly under severe condi- tions of disease. The Control of Rootrot The findings of this survey suggest very clearly the control measure which must be employed to avoid rootrot — control measures which have been for the most part stated previously. In districts where pea culture on a large scale has been introduced recently and where there are few diseased fields already established, increase in disease can be avoided readily. First, poorly drained soil should be avoided for pea planting. The adoption of a long rotation on suitable soil should deter the appearance o' disease for many years, perhaps indefinitely. The length of rotation re- (|uired to prevent serious development of the disease appears to be dependent to some degree on the soil type, being longer on clay soils than on loams. A rotation of five or six years duration is suggested as probably adequate on most Wi.sconsin soils. If it appears advisable for commercial reasons to plant peas as long as possible on the same ground, the field records collected here show that under average conditions it is possible to do this for three years before serious loss from rootrot need be anticipated. Occasionally they may be planted for a longer term of years. Generally, the disease appears in s+ich fields for one or two years before it become.s- destructive ; and thus a careful examination of fields for disease can readily determine when such fields have become unsafe for further planting. No serious loss from disease need be incurred from such practice if in- telligent supervision is employed. In districts where the disease is already well established avoidance of disease is not so easily accomplished. Fields in which peas have failed from rootrot are not safe for replanting for ten years after the failure on most soil types. Soil from such fields can serve to carry disease to other fields during this period of time, and thus much new land that has never grown peas in infested districts is unsafe for peas. Since it will be impossible in most cases to determine in advance just where such injured areas are. it will be impossible to avoid loss in all cases, even where a suitable rotation is adopted. As soon as infested tracts are located, they must be abandoned for pea culture. Transfer of soil from diseased fields should be avoided. Uncured silage from pea vines should never be fed or returned to fields as manure. The use of resistant varieties of peas may become profitable on infested soils under some conditions. Such varieties should be planted as early as possible. Under conditions favorable for the development of the disease even the most resistant varieties known at present may be damaged greatly, and in any case their growth increases soil infestation quite as much as those varieties which are readilv destroved. LESS IMPORTANT PEA DISEASES Fusarium Stem and Rootrot A stem and rootrot of peas caused by Fiisariii}ii luartii App. & Wr. var. ;>is: F. R. Jones has been described (8) as occurring in Wisconsin and several other states. This Fusarium which was the only important parasite among several species and varieties tested produced typically its initial and most significant invasion at the base of the stem at or immediately above the point of attachment of the cotyledons. The resultant lesion becomes elongate. 20 W'iSCOXSIX Rl'.SF.AKCii UuLr.tlTIN 64 fxtciuling up the stem as a wedge-shaperl, dark bniwn or chocolate colored lesion, not appreciably shrunken until well advanced. This cortical rot rriay deepen and pentrate or even sever the vascular cylinder, after which, at higher soil temperatures the fungus invades the xylem for a short distance, producing a bright orange red or brown discoloration which may extend as far as the first node. Extensive vascular invasion is not a characteristic development. Rootlets may be attacked, in which case the symptoms are not visually distinct from the effects of several minor parasites. When this disease occurs alone as a stemrot it has been considered to be of easily recognized character. It has, however, almost always been found in association with rootrot where its presence rarely can be discovered except by the isolation of the fungus. In the course of the survej' only a few instances of the type of stemrot caused bj'^ Fusarium were discovered, and in all of these cases laboratory study showed the cause to be a phoma- likc fungus which is mentioned below. Thus it appears that the Fusarium stem and rootrot of peas did not occur in Wisconsin this year as an independently recognizable disease. Laboratory study was not made to determine whether it occurred in association with rootrot. The absence of this disease this year may not have been due to the absence of the parasite. A study of conditions which make possible the development of this disease has shown that a mean soil temperature of 18° C. is necessary before conspicuous lesions on stems appear, and that a soil temperature of approximately 24° must be reached before plants are killed or conspicuously injured. Reference to the record of soil temperature prevailing this year discussed previously will show at once that this disease must have been delayed in development even more than rootrot. and that there was little opportunity for it to become destructive. Thus it is possible that in a warmer season this disease may appear again, though its seems unlikely that it will be important under Wisconsin condi- tions apart from its association with rootrot. Footrot, a Disease Resembling Fusarium Stem and Rootrot Early in June, 1924, before soil temperatures were favorable to the independent parasitism of Fusarium martii pisi, plants were found showing lesions typical of Fusarium stem and rootrot. Such lesions, when plated < ut, yielded cultures, not of Fusarium, but of a Phoma or phoma-like fungus which the senior author has isolated from pea root and stem lesions many times before. Inoculation experiments in the greenhouse have demonstrated that this fungus is capable of producing lesions, which resemble very closely those produced by Fusarium. Haenseler (5) has reported frequent isolations of Phoma species from peas in New Jersey. Footrot symptoms were encountered widely but sparmgly in Wisconsin in 1924. Probably the disease will not prove of threat imporvance. Seedling Injury Caused by Rhizoctonia The sterile or Rhizoctonia stage of Corticium vagum B. & C. is another fungus capable of damaging the underground portions of the pea plant. Of wide occurrence in cultivated soils, this fungus is frequently encountered in pea fields where under some conditions it may assume considerable importance. Generally, however, it is of minor importance as a parasite of peas, Rhizoctonia may attack any underground portion of the pea plant, but it causes greatest injury when invading very young tissues. It may enter germinating seeds killing the embro or destroying the cotyledons, I'ka Dim; \ si Siik\i v in Wisconsin 21 rcmijving the loud rebcrvc oi the developing seedling. It may attack seedlings before emergence from the soil, injuring or completely destroying the growing points of roots and stem. When the stem tip is thus destroyed the pea frequently produces secondary shoots, one or more of which may escape similar destruction. Root tip injury may continue even after the plant is well established. This fungus may also produce lateral lesions on stems and roots of a type characteristic of this fungus on other hosts, being brownish, sunken and eroded, oval or oblong cankers. Coarse brown hyphae of the fungus frequently found on and around such lesions are helpful in diagnosis, but in general the injury caused by this fungus, par- ticularly upon roots, is not always readily distinguished under field condi- tions from that of some other parasites. Richards (13) has shown that the soil temperature most favoring the parasitism of Rhizoctonia on the pea is 18° C, but that it is able to operate in a less important way through a wide range of temperatures, beginning as lovvT as 9° and continuing up to 29^. The minimum temperature is thus below that of the major pea root parasites, and consequently Rhizoctonia injury occurs earlier than the more important root diseases. Late planted peas suffer greater injury than those planted early in cold soil. In the 1924 survey it was not possible in all cases to distinguish the injury produced by Rhizoctonia under field conditions. Injury attributed to this fungus was i-oted in 35 fields ranging from the killing of 30 per cent of the plants in rare cases to reductions of stand that were negligible, and from reduction < f vigor that would amount to 25 per cent of the crop to that which wouid escape detection. Rhizoctonia injury was noted on soils ranging from light sandy loams to muckv clay loams, but the greater reductions of stand and vigor were limited to a few fields of sandy loams, Carrington silt loam and Miami silt loam Seedling and Root Injury Caused by Species of Pythium When pea plants suffering from rootrot are examined in the laboratory, the species of Pythium 'ong known as a destructive seedling parasite will often be found present in the diseased tissue of many or all of the plants. Inoculation with some of the cultures of Pythium obtained in this way has shown that the fungus is capable of preventing germination of pea seed or of destroying many of the seedlings before they emerge from the ground, and occasionally some degiee of stem and rootrot is produced. S">me preliminary work with this fungus earlier led the senior author (6) to express the opinion that it was the most important cause of pea rootrot — an opinion which was not substantiated by further work, and which has since been corrected (10). However, more recently Stone (16) in Ontario has called attention to the association of Pythium with disease, ascribing to it a rotting of pea plants near the surface of the soil. Some attention has been given to Pythium species in relation to pea dis- ease during several years and though the study of the relation of species of this genus to root injury is far from complete, a few notes on the progress of the work may be presented. Although it will be shown in the following tables that species of Pythium capable of causing severe seedling injury under favoring conditions are present in adundance in some agricultural soils, yet the survey records no instance of important injury from these species. The most obvious explanation iCr the failure of this group of fungi to produce injury is found in the comparatively high soil temperature required for their activity. An incomplete study of the more actively parasitic species indicates that a soil temperature of 16 °C. is necessarj' before much seedling injury occurs. Most peas have passed the stage at which seedling injury is possible before the mean soil teinperature has reached this point. WlSKlXSlX RksEAKCH iiULLETIX 64 The study of the relation of Pythium to root and stemrot has been greatly retarded by the fact that the cultures obtained from peas in the field have been found to belong to several species differing somewhat in pathogenicity and frequency of occurrence ; and furthermore it is not at all certain that present methods used in making isolations from plants secure cultures of all species present. Thus a vast deal of work will be required before the relation of Pythium species to stem and root injury of mature plants will be fully known. It may be stated, however, that inoculations made under controlled conditions have shown very slight ability in any species studied thus far to produce either stem or rootrot under usual field conditions. During the summer, isolations were frequently made to secure cultures of any species of Pythium that might be present. Since previous experience had shewn that seme species arc almost as frequently found in association with '"oots apparently heaJthy as with those di.st-.,.sf d, c-/ltures wcie ma.k, from both. In making cultures, no sterilizing agents were applied to the surface of decaying tissue because they penetrate rapidly and destroy all Phycomycetous mycelium quickly. Thus mycelium adhering to the outside of roots may give rise to a culture, a difficulty which can hardly be avoided. The results of Td isolations are summarized in Table VI. This table cor- roborates previous experience that cultures of Pythium are obtained with approximately the same frequency from healthy as from diseased plants. It also shows that in about one-third of the isolations two species were obtained, though in such cases no two species seem to be found more frequently associated than others. The cultures obtained in this way have been partially classified into species groups which are designated by letter in Table VII. Species A, B, D, and perhaps E, seem to have been included by pathologists under the name Pythium dcbaryanum ; and this group contains the more aggressive parasites. From this table it appears that the several species were obtained with approximately the same frequency from healthy plants as from those showing disease. In fact a small number of isolations from clover roots made at the same time with those from peas have given equal success in obtaining cultures of Pythium, the frequency of occurrence of the several species being somewhat different. From the observations made thus far it appears that some of these species of Pythium occurring abundantly in the soil may at times be responsible for the death of root ends, and for some rootlet injury; but only rarely for root and stemrot as it is usually known. It is possible that almost universal invasion of the root cortex of peas and clover by a mycorrhizal fungus (9) renders this tissue especially accessible to these species, and that it is from this superficial invasion that the fungus is most frequently obtained in culture. T.\BLE VI. — Frequency of Occurrence of O.ne or More Species of Pythium in Isolations fro.m He.xlthy and from Decaying Pea Roots and Stems. No culture of Pythium obtained 1 species of Pythium obtained 2 species of Pythium obtained Healthy pea roots 10 0 16 14 Deraving pea roots and t 8 An Undescribed Wilt Disease During the survey a disease was observed which in its effects upon the vines superficially resembles rootrot injury, but which seems otiologically distinct from any of the known pea diseases in Wisconsin. It was character- I'i'.A DiSKASK SlK\K\- IN WISCONSIN' 2.^ izcd generally by rapid and cuijipktc withering oi the vine without conspicuous rotting or discoloration of the cortex of roots and basal stem such as are typical of the better known diseases. Root tip injury was frequently found associated with it but not to a sufficient extent to account for the death of the plants. Fifty fields were encountered in which what appeared to be this disease was present. Infestation varied from small patches to 100 per cent of the field ; in the latter case crop destruction, especially of the sweet varieties of peas, was almost complete. In several factory districts this Tahi,!-; VII. --Fukql'kncv cji" OccLaHENCK ()!•' SiiVEivvi- Spi:cies ok Pypiiium IX IIkalthy and Diseased Pe.\ Hoots as F-Ikpresented BY Isolation. Spe- cies A Spe- cies B Spe- cies C Species D and K Unclas- sified Healthy pea roots Decaying pea roots and stems. . . 3 20 9 4 2 14 11 6 7 disease caused greater losses than did rootrot, and in the total area surveyed it ranked second in destructiveness only to the disease caused by Aphanomyces. Slightly over half of the infested fields lay in Fond du Lac County, and nearly all of the remainder were in adjoining counties. Three- fourth of all such fields were on black soils of which Carrington silt loam was dominant. This disease appears to be correlated with the previous growth of peas, much the same as is rootrot. Leaf and Podspot or "Blight'* Caused by Ascochyta The most widely known of the foliage diseases of peas is the leaf and podspot caused by Ascochyta pisi Lib., the conidial stage of MycosphacrcUa pinodes (Berk. & Blox.) Stone. So abundant and important was this disease in Wisconsin in 1911 and following years that it was regarded as the chief cause of "blight" of peas. Recommendations made for its control assisted perhaps by climatic conditions during the past few years have brought about its almost complete disappearance. This fungus attacks all varieties of canning peas, and occurs on vetches. On pods, the lesions are rounded, somewhat sunken, light brown at first becoming darker with a light brown border, and with brown pycnidia in the center. On leaves the lesions are irregularly rounded with yellowish brown or ashy centers and dark borders. On stems the lesions are elongate. If lesions are abundant they may become confluent killing a large part of the foliage, or even the entire plant. Lesions near the surface of the ground may spread over a considerable portion of the under- ground stem, which in the past has given rise to the impression that this fungus was the cause of much of the injury to roots that has later been found due to Aphanomyces. It has been demonstrated by Stone (15) and Vaugham (17) that this fungus lives over winter on diseased vines producing the ascigerous stage in the spring from the spores of which plants may be infected. The fungus has long been known to be carried in the seed. Several years ago when this fungus was very abundant the senior writer found nearly 10 per cent of peas in one sample of commercial seed carrying this fungus, though among many samples examined, only a few w-ere found to contain infected peas. The fungus usually enters the seed coat and 24 W ISLdXMN JvESEARCH BULLETIN 64 cotyledons close to the plumule. When the seed germinates, the ends of the embryonic leaves in the plumule become invaded by the fungus. When the leaves are carried above ground and expand, the fungus carried in them produces lesions promptly, usually at the margins of these leaves. Pycnidia are soon formed, the spores are scattered over the foliage of other young plants by splashing rains, and the disease spreads from such centers of infection. Since the fungus penetrates deeply in the seed, no method of seed treatment thus far tried has been able to destroy it without damaging the seed. Leaf and pod spot caused by Ascochyta was of no importance in Wis- consin in 1924. Only 18 fields situated in 12 different districts were found to contain traces of the disease, and none of these were appreciably in- jured. The disease was first observed as early as June 2 where it was developing apparently from infected seed on plants only five inches tall. The disease was not found again until July 15. Its frequency of occurrence seemed to increase up to the end of the season, but no factory district was observed to contain more than three fields which showed traces of infection. Leafblotch Caused by Septoria Leafblotch of peas caused by Septoria pisi West is frequently associated with and confused with the leaf and podspot caused by Ascochyta. This disease was formerly considered an important factor in pea "blight." Melhus has shown (12) that it is rarely difficult to distinguish these two diseases except perhaps in their later stages. The disease caused by Septoria is typically a blotch rather than a spot. Its margins are irregular, sometimes angular when restricted by the larger veins of the leaf, and without any distinct marginal band. They are yellowish green at first, turning brown upon the death of the tissue. Such blotches increase in size indefinitely, sometimes extending down the petiole and infecting the stem. Infected tissues produce numerous pycnidia, yellowish brown at first, becoming darker with maturity. The overwintering of the fungus has never been traced in a satisfactory manner. Melhus found the pycnospore short-lived and found no seed infection. The senior writer, however, found one collection of this fungus, icept in a sheltered location out of doors, that maintained viable pycnospores for at least a year. Seed infection is not infrequent. Infection usually occurs at the hilum which exhibits a very characteristics pink discoloration, mvolving more or less surrounding area. Although the fungus can be isolated readily from such seeds, no infested seedings have yet been observed from their germination. In 1924, Septoria leafblotch was more abundant than leaf and podspot. but still was of very little importance in the state as a whole. It was observed first June 6, and was collected at intervals during the summer without showing any marked increase with late summer rains. It was observed in only 29 fields in twelve districts. Most of these fields contained no more than a trace, while a few were damaged to the extent of about 10 per cent. Septoria was most prevalent in the northern districts, all of the heavily infested fields being situated north of the latitude of the north end of Lake Winnebago. A Septoria Leafspot New to Wisconsin In one field in Dodge County was found an unfamiliar leafspot caused by a fungus identified by Dr. J. J. Davis as Septoria flagellifera E. & E. Both the spots and the pycnospores of the fungus are distinct from those of Septoria pisi. Ellis and Everhart (4, p. 57) have described the spots and the fungus from material gathered in South Dakota as follows : Pea Disease Survey in Wisconsin 25 "AmphiKcnoiis, spots suborbicular, 0.25-1 cm., diameter, subzonate, with a slightly raised border, rusty-brown at first, whitening out m the center: perithecia hemispheric —prominent or subconical, dark amber color, 7.T-1J4 microns in diameter, sporules filiform, hyaline, nucleolate, only slightly curved, 80-lJO x 3-2.5 niicvc>!i-,. "Differs from S. pisi West, in the different character of the spots and the much longer sporules." This fungus occurred sparsely in the one field where noted and showed no evidence of possible importance as a parasite. Anthracnose The anthracnose of peas caused by Colletotrichum pisi Pat. was recently described in Wisconsin by Jones and Vaughan (7). This disease closely resembles that caused by Ascochyta and may easily be confused with it in casual examination. Occurring on all aerial parts of the plant, it produces lesions that on pods are circular and sunken ; on leaves, irregular in out- line ; and on stems, elongate. These spots are generally brownish with some- what darker brown borders. Stetn lesions when covered with spores from the numerous acervuli are ashen when dry, and copper colored when wet. In later stages, small black sclerotial bodies, which are helpful in identify- ing the disease, develop in the lesions. The life history of the fungus has not been followed through the winter ; although the fungus attacks pods freely, it has not been observed on the seed. Pea anthracnose has been reported in the United States only from Wis- consin thus far. Although recognized as a destructive parasite, its distribu- tion has been considered so limited that its importance in the state as a whole was minor. In 1924, both in severity of the disease in individual fields and in the total injury observed during the survey, anthracnose was far more import- ant than any other foliage disease. It was found in 42 fields in 13 districts lying in 6 counties. The majority of these fields were infected very lightly or were infected so late in the development of the crop that injury was slight, but there were a number of fields in which losses were severe, amounting to as much as 50 per cent of the crop in one 28 acre field. A few fields showed severe defoliation relatively early, but in the main heavy infection did not occur until near the end of the season when heavy rains caused rapid increase of the disease where it occurred. In a number of fields anthracnose was associated with bacterial blight in causing important foliage destruction late in the summer. Downy Mildew Downy mildew of peas caused by Peronospora i-iciae (Berk.) DeBy. is found widely but sparsely distributed almost wherever peas are grown. Rarely does it become conspicuous. Frequently a few plants may be found which have been completely overrun by the inildew in an apparently systemmic infection. Such plants are dwarfed beyond recovery, but con- stitute at most a small fraction of one per cent. Typically the mildew occurs as irregular downy patches of violet gray color on the under side of leaves. Such leaves are yellowish above and are usually recurved. The occurrence of downy mildew in 1924 was limited to a very lew fields, in none of which did it cause important injury. Bacterial Blight Sackett (14) has described a bacterial disease of peas caused by Pseudonwnas pisi Sackett which caused much loss in Colorado in 1915 and 26 Wisconsin Research Bulletin 64 following years. A similar if not identical disease occurs sporadically in Wisconsin, rarely causing important crop reduction. The disease is characterized by the production of water-soaked lesions of olive green to olive brown color which may remain small spots, or, under favoring conditions, may spread rapidly to include large portions of leaves and stem. Such lesions become darker as they dry. In wet weather bacterial ooze may appear on lesions. Infection is through stomata and wounds. Lesions may develop on pods, and mature peas beneath lesions are sometimes found to bear flakes of what appears to be the dried bacterial slime. Although it has been assumed that the bacteria may be carried alive in this way and thus infect seedlings, such seed transmission of the disease has not actually been demonstrated. Early spring infection occurring while the seedlings are still young may result in important reductions of stand. Severe infection at any time later is able to weaken the plants seriously, at times destroying practically the entire leaf surface. Bacterial blight was seen in 1924 both early and late in 23 fields in various districts. Of the infested fields 74 per cent had never grown peas before, or at least not in recent years, a fact which seems to indicate no important correlation between the occurrence of this disease and the previous growth of peas. Likewise no correlation was found between bacterial blight and source of seed. Early attacks were observed to weaken the plants and to cause uneven development. Usually such early infection was outgrown and did not lead to subsequent increase of the disease. A few fields were observed to show signs of recent infection during late June and early July, but the most severe cases were observed after the middle of July. At this time a few fields were seen badly damaged, in one of which the disease had developed freely over leaves, stems, and pods. Mosaic Pea mosaic, observed in experimental plantings at Madison in 1923. was first recorded as occurring in commercial plantings in Wisconsin in 1924. when it was found widely but sparingly distributed. In spite of its early appearance in 1924 at Madison it was not conspicuous in commercial fields until July 12, but after this date at least traces of mosaic were found in practically every factory district visited. In all, it was recorded in 63 fields in 10 counties. Most infested fields contained a mere trace, while the heaviest estimated infestation was 20 per cent. A considerable number of fields showed from 5 to 15 per cent of the plants infected. Varieties found diseased with the number of fields of each are : Green Admiral, 20; Alaska, 14; Horsford. 13; Perfection, 10; Winner, 3; Advancer, 2. Not only did Admirals show the greatest numl)er of infested fields, but also the heaviest infestation and the most evidence of injury. Injury from mosaic appeared to be negligible except in a few fields where infestation was heavj'. In some of these fields, it appeared that mosaic plants were somewhat dwarfed and failed to fill as many pods as healthy plants. Judged by its behavior under conditions prevailing in Wis- consin in 1924 mosaic of peas can hardly be regarded as such a menance as mosaic diseases of some other crops have been. The origin of mosaic in these pea fields can be only conjectured at present. Dickson (1) has reported the appearance of the disease in several varieties of field peas from seed transmission. On the other hand, Doolittle (2), using seed from mosaic plants in experimental plantings at Madison, Wisconsin, and McMillian, Michigan, has planted nearly 1,000 seed from Alaska peas and smaller numbers from other varieties under controlled conditions without obtaining a single mosaic plant. The field occurrence of the disease in 1924 did not suggest seed trans- Pea Disease Survey in Wisconsin 27 mission, for different varieties and peas from seed from different sources appeared to develop the disease almost simultaneously in certain districts. In one, for example, mosaic was found in only four fields representing three varieties from different sources. On the other hand Doolittle (2) has produced mosaic in pea plants from mosaic red clover by transfer of aphids and by artificial inoculation. Inas- much as many pea aphids migrate to peas from red clover, on which they winter, it seems likely that mosaic clover plants which are abundant locally are the source of the disease in commercial plantings. FIG. 8.— LOCATION OF PEA CANNERIES IN WISCONSIN Circles indicate the location of the pea canning factories in Wisconsin. Black dots designate the factory districts surveyed for disease in 1924. 28 Wisconsin Research Bulletin 64 SUMMARY 1. — The Wisconsin pea crop of 1924 represented a total farm value of over $7,000,000. Pea diseases play a major part in de- termining the systems of pea culture employed and in reducing profits to both growers and canners. 2. — In 1924 a detailed survey was made of 688 fields compris- ing 5,416 acres representatively distributed in the pea growing sections of Wisconsin to determine the importance of the various pea diseases and especially to study the development of rootrot in relation to cropping practices, soil types, and other factors which a])peared to influence its occurrence and destructiveness. 3. — This bulletin is a summary of the findings of this survey, supplemented with notes from pea disease investigation conducted in this state by the U. S. Department of Agriculture in cooper- ation with the Wisconsin Experiment Station. 4. — The rootrot chiefly considered in this survey is that caused l)v the fungus Aphanomyces euteiches Drechsler. This fungus is assumed to be indigenous is Wisconsin soils, occurring especially in wet locations. It increases rapidly in the soil with culture of peas. 5. — The season of 1924 was so cool and favorable for the de- velopment of peas that fields infested with rootrot did not appear to suffer as great damage as in other years. 6. — The rootrot caused by Aphanomyces was more destructive in 1924 than all other fungous and bacterial diseases of peas com- bined, considering the state as a whole. In some localities a new- ly observed "wilt" disease was more destructive, and in the state it ranked second to the Aphanomyces rootrot. Anthracnose caused by Collefotrichum pisi Pat. was the most destructive of the foliage diseases, causing important losses in several districts. 7. — Rootrot was found in 32 per cent of all fields examined. Eleven per cent of all fields were severely infested. The total loss in inspected fields is estimated at 8 per cent of the total yield. Since diseased fields were especially sought in the survey it Is believed that the pea crop in the state as a whole did not suffer as great a loss as this. Even if the loss in the entire state amounted ta only half this amount or 4 per cent of the total yield it would represent a loss to the growers of about $300,000 in addition to losses incurred by the canning companies. 8. — Of the fields ins])ected 48 per cent were growing their first crop of peas. Fields which had been i)lanted more than once to peas were found to have on the average a rotation period of about two and one-half years. 9. — Rootrot increases both in frequency of occurrence and in severity with the number of crops grown. Rootrot occurred but rarely in fields growing the first crop of peas, while all fields growing the fifth crop were more or less infested. The occur- rence of severe infestation does not increase rapidly during the Pea Disease Survey in Wisconsin 29 first four croi)s ; l)iit it rose to 56 per cent of the fields j^rowing the fifth crop. 10. — Peas were found growing on 27 soil types and seven groups of incompletely classified soils, thus making the number of fields on most types too small for comparison. No soil type showed p)rospects of providing environment in which rootrot can- not develop. The two soil types which include nearly half of all the fields examined — Miami silt loam and Carrington silt loam — show little difference in behavior. In general, with similar crop- ping, clays and clay loams have a larger percentage of severely in- fested fields than loams and silt loams or lighter soils. In fields including more than one soil type, disease usually appears first in the soil with greater moisture holding capacity, or in poorly drain- ed spots. Greater precautions to avoid rootrot are needed on heavy or wet soils than on well drained, medium, or light soils. 11. — Rootrot was found to persist in some Wisconsin soils for 10 years after it had caused crop failure. After such failure no fields were found entirely free from rootrot in less than 10 years. 12. — The only commercial variety of pea that showed an im- portant degree of resistance was the Green Admiral ; and even this variety was greatly damaged when not planted early. 13. — A five or six year rotation is suggested as a method of control of this disease which should prevent its appearance on most Wisconsin soils not alreadv infested. When a shorter rota- tion seems advisable, careful insj^ection of fields can detect its development before it becomes destructive. Resistant varieties are being tested which may be of value in some situations. 14. — Other diseases discussed in this bulletin are as follows : Stem and rootrot caused primarily by species of Fusarium was not found in 1924. A new but apparenth- relatively unimportant footrot caused by a species of Phoma was noted. Seedling injury caused by Rhizoctonia solani Kuhn was noted in 35 fields, but for the most part was not important. The relation of species of Pythium to seedling and root in- jury is discussed briefly. A new wilt disease was found in 50 fields, in some localities causing greater losses than Aphanomyces rootrot. The cause has not yet been determined. Leaf and podspot caused by Ascochyta pisi Lib. was rare and unimportant. Leafblotch caused by Scptoria pisi West, was not abundant but was important in a few fields in northern districts. A leafspot new to Wisconsin caused by Septoria flagellifera E. and E. was noted. Anthracnose caused by Collet otricJnim pisi Pat. was the most important foliage disease encountered, causing considerable dam- age late in the season. 30 Wisconsin Research Bulletin 64 Downy mildew caused by Peronospora znciae Berk, was rare and unimportant. Bacterial blight caused by Pseudomonas pisi Sackett was en- countered occasionally both on early planted peas on wet soil, and on foliage of mature plants late in the season. Mosaic was encountered frequently late in the season, but rarely appeared to reduce yields. LITERATURE CITED ( 1 ) Dickson, B. T. 1922 Studies concerning mosaic diseases. MacDonald College Technical Bui. 2: 1-125. illus. (2) Doolitle, S. P. and Jones, Fred Reuel. The mosaic disease in the garden pea and other legumes. Phy- topath. (in press). (3) Drechsler, Charles 1925 Root-rot of peas in the middle Atlantic states in 1924. Phytopath. 15: 110-114. (4) Ellis, J. B., and Everhart. B. M. 1900 New species of fungi from various localities with notes on some published species. Bui. Torrey Bot. Club 27 : 49-64. (5) Haenseler, C. M. 1924 Pea root rot investigation. New Jersev State Agr. Exp. Sta. Rpt. 44: 366-375. (6) Jones, Fred Reuel 1920 Pythium as a causal factor in "pea blight." (Abstract) Phy- topath. 10: 67. (7) ■ and Vaughan, R. E. 1921 Anthracnose of the garden pea. Phytopath. 11: 500-503. illus. (8) (9) 1923 Stem and rootrot of peas in the United States caused by species of Fusarium. Agr. Res. 26 : 459-475. illus. 1924 A mycorrhizal fungus in the roots of legumes and some other plants. Jour. Agr. Res. 29: 459-470. illus. (10) — and Drechsler, Charles 1925 Rootrot of peas in the United States caused by Aphanomyces euteiches n.sp. Jour. Agr. Res. 30 : 293-325. illus. (11) Linford, M. B. and Vaughan. R. E. 1925 Rootrot of peas. Some wavs to avoid it. Wis. Agr. Extension Cir. 188: 1-10. illus. (12) Melhus, I. E. 1913 Scptoria pisi in relation to pea blight. Phytopath. 3: 51-58. illus. (13) Richards, B. L. 1923 Soil temperature as a factor affecting the pathogenicity of Corticium vagum on the pea and the bean. Jour. Agr. Res. 25 : 431-449. illus. (14) Sackett, Walter G. 1916 A bacterial stem blight of field and garden peas. Colo. Agr. Exp. Sta. Bui. 218: 1-43. illus. (15) Stone, R. E. 1912 The life history of Aschvta on some leguminous plants. Ann. Mycol. 10: 564-592. (16) 1924 Root rot and blight of canning peas. (Abstract) Phvtopath. 13: 348-349. ( 17) Vaughan, R. E. 1913 Mycospherella pinodes the ascigerous stage of Ascochyta pisi. (Abstract;) Phytopath. 3: 71-72. Research Bulletin 73 December, 1926 Studies of the Epidemiology and Control of Apple Scab G. W. KEITT AND LEON K. JONES Agricultural Experiment Station of the University of Wisconsin Madison CONTENTS Introduction 1 Epidemiology in relation to control 3 Field studies 3 Seasonal development records 3 Meteorological records 5 Discussion of results 6 Laborator}^ and greenhouse studies 6 Spore germination experiments : 19 Viability and longevity 19 Relations of moisture 22 Relations of temperature 23 Relations of light 24 Infection experiments 25 Apparatus and metliod for experiments under con- trolled conditions 25 Mode of penetration of the fungus 27 Relations of temperature 29 Relations of moisture 32 Relations of light 34 Relations of stage of development of host organs 35 Relations of certain fungicidal treatments 39 The development of epidemics 41 Modes of overwintering of the fungus 41 Production and dissemination of the ascospores 42 Apparatus, methods, and records of field studies 43 Time of earliest maturity of ascospores 44 Periods of ascospore discharge 45 Frequencies of ascospores in orchard air 45 Some factors which affect production of perithecia and discharge of ascospores 46 The occurrence of primary infection 47 The special significance of earh' infection of sepals 51 Production and dissemination of conidia 53 The occurrence of secondary infection 57 Critical periods for the development and control of epi- demics 57 Spraying and dusting experiments 58 Metho^ds 58 Experiments in 1919 63 Experiments in 1920 66 Experiments in 1921 69 Experiments in 1922 72 Experiments in 1923 T^ Experiments in 1924 83 Discussion of results 88 Other studies bearing on control measures... 95 Summary and conclusions 96 Literature cited 100 J \ \ \ ^ Studies of the Epidemiology and Control of Apple Scab ^ r (i. W. Ki:nT Axu Lkcjn K. Joxes APPLK Sl'AH. caused by I'cniuri.i inacqiuilis (I'kc.) W'int., is one of the most widespread and destructive fruit diseases. The range of its occurrence closely- parallels that of apple culture throughout the world, save in comparatively limited areas, as in certain arid sections and in climates which approach the higher temperature limits of apple production. In most of the chief apple producing sec- tions of the world, it occurs in sutTicient severity to make its control essential to commercial apple culture. The disease occurs at its worst where the climate is humid and cool in spring and earl_\' summer. It is, therefore, very destructive in the more humid sections of the Pacific Northwest of North .\merica and in the north-central and northeastern apple hidt of the United States and Canada. It is very severe in its outbreaks in Wisconsin, especially in those sections where tile cliniate is most influenced by the Great Lakes, as on the Door County Peninsula. The widespread occurrence and great economic importance of apple scab have contributed toward making it tlie subject of many investi- gations. Most of the voluminous literature that has resulted, except the more recent contributions, is readily accessible through the works of .\derhold (1896, 1000), Clinton (1901), Wallace (1913), and Morris (1014), which include extensive bibliographies and reviews. Conse- cpiently, a general review of literature in the present paper appears to be unwarranted. Such discussions of previous work as seem neces- sary appear in the appropriate connections in the bod\' ot the paper. Studies of apple scab at the Wisconsin Agricultural Experiment Station were begun by Trelease (1884), who gave one of the best early accounts of the nature and cause of the disease and a discussion of its occurrence and prevention in this state. Several years later, the work was taken up by (ioff,' who played a leading part in adapting recently developed spraying methods to apple scab control. .A.s a result of GofF's pioneer work, spraying with Bordeaux mixture came to be an accepted practice for scab control in Wisconsin orchards. Subsequently, the Department of Horticulture has conducted numerous 'Fi-T an annctated bibliography of Goff's wirk mi apple scab, see Clinton (1901, p. l.^n.l.^l). 2 Wisconsin Research Bulletin 73 tests of the more promising materials and programs which have come into use for the control of this disease. In 1916 and 1917, at the request of R. H. Roberts/ who was then in charge of this work, the senior author cooperated in planning the tests and recording results, and made such observations as were feasible upon the development of the causal fungus and of the disease in relation to control measures. This work, which was conducted in the orchard of the Northern State Hospital, at State Hospital, Wisconsin, included comparative trials of commercial liquid lime-sulphur in various concentrations, Bordeaux mixture of various concentrations and ratios of lime to copper sulphate, a mixed program of lime-sulphur and Bordeaux mixture, barium tetrasulphide, and sulphur-arsenate dust (90-10). Arsenate of lead (powder), at the rate of 1 pound in 50 gallons, was included in all spray treatments. Each program was applied according to a four- treatment schedule which, with minor modifications, was then generally accepted as standard for apple scab control throughout the north- central and northeastern apple belt of the United States. The treat- ments were made (1) just before the blossoms opened, and if possible when the blossom buds were well separated in the clusters, (2) when about three-fourths of the petals were ofif, (3) about 10 days later, and (4) at a time most advantageous for combating the second brood of codling moth. Because of limitations of space, the detailed data from these trials will not be given. Their most outstanding feature was the failure of any of these programs to control the disease con- sistently. In 1916, a year of moderate scab development, the disease was easily controlled, except in the case of a very susceptible unidenti- fied variety, which was severely scabbed. On this variety none of the programs tested controlled the disease satisfactorily. In 1917, a year of severe scab development, none of the programs tested gave satis- factory control even on moderately susceptible varieties. In view of these results, it seemed desirable to undertake a more critical study of the disease and its control. After an interruption occasioned by the world war, such studies were begun by the present writers in 1919 and are still in progress. Field laboratory' and field studies, the latter in cooperation with apple growers, have been conducted at Sturgeon Bay, Wisconsin. For the five seasons, 1920-1924, the junior author lived in quarters adjacent to the field laboratory, thereby gain- ing an excellent opportunity for making adequate observations and records. Field, laboratory, and greenhouse studies have been con- ducted at Madison. •Grnteful :icl;iio" Icd^; incuts ;iri' iiijidc: To Prof. li. TI KolxTts iiiul the Department of Horticulture of the University of Wisconsin for iliiif cordial and efficient cooperation in the work of 1910 and 1!M7. To >'e>sis. K. .\. Stokdyk (in 1919K R. H. Streets fin 1920 and 1921), and E. E. Wilson (in lOS.^ and 1921) for valued assistance in conducting the field work. To ISIessis. W. I. Lawrence, M. B. GofT, S. T. Learned, and B. W. Sackett for the cor V. ^ J f , '' ^ »_ vO ). ,i y^ t ^ \ ' •\ s ^^ < - — * ^^ ro -C ^^;n ts. 7 '^ *^s CM ' , > rO X ■«.^ •-- CNi s <*' a< ? t._ ^ «: 1 i. ■^^ s. i_ t^ \ ,.•' >• «. lO o a:_ w .0-. O ■z- ' o CD ° o u oc I (O a: tO- O (J i? / ; V, <^ > \ <^ ) ==:; " ^*-1 — 1 ■^ >^ — - DC LJ 5< a. cn ii- o lij o -I CVi ,< Si z < on (^ 't c--^ o *^ / ^ V^ ^ 0^ (.\i \r> w y ''^ V ->^ cv ^ (s ^> ^ PO a> 1- / V, lO 5 3 \ '•-. 1 3 / Ns '•. I A ^ 3 3' ,, '■ ' ^ Z? <,^l ; -.- ;_ y- ? - J /< V / : \ n J- \ / 1 • • Si Si /^ 1 !' ^ x"^ <:. s [ -o ^■-~^ s k,^ ^ 1 in <0 O '. Q ° UJ <1 j^ ' r ^ oc '.5 '^9 ■^ & ^ 1 r J . * ■-. 1 ^o^^^ ^ \ |3_ CO l"^' o , en ^ \ 1 p t /' en: « t Si <^^ T L iz i2 ac -> 1 u -to- Q- il * 1 3^ ^ ■? \ n s i \ fr^' *( ^ ^> \ -:? -5 < 2 >g ■& ^ r I — X^ . ■ <^ 1 — 4 ^ is '! , £i ,-> ; — -w- ^ ^ ' o 1 , *^ * ~~. 2 / ' 1 1 ^ to ^ * — ^0 |z < > .-' ? M_ I w ? ^ 7i • o , ; > m '' — : C3 f ; 0> 1 - * I. \ ^ \ '• <^ / V ; > '. "^ s S 8 s i 2 S ? S SS § S °2 o^ «, M — O ■. !£! O lf> OK? 2 >0 ~"| ^ ? z o S te jZ 5 CC « 1- o lu 2 q: ' 1 <= a^ 5 10 < a: 5 1- 2^ 1? iji < SS5 Z in !i H 5?i -J UJ is 5<5 at® o ofjuj g 5°c >UI >: ^!i? V 1 o « Epidemiology and Contkoi- of Apple Scab SiggR2X%S S^Soa'alJa'fio ino!SS>0 , s \. • -4-^.^1 ► ' - < "^ p»^ §1 — ^ »* -^-0^ 1 k y N. evj ■^ ^^ 1 *, -^ :f> 1 r V.N Ti ^ / , ^ tj - ^ v -*= Ti N. <^ •' ^7 ,^ •v^ ^1 } If. 2 J ,^-- •' 1 1 tO ^ • — _] O) ^ -T^ — "1^ — 3» 5■ 3 \ . \ j H 3 3' > — ■ " t '•^ 1 ' 3 r — Q^ 1 — ci i I -<■ X -o ^2 -z ^ '••- I <^4 /•' I - ^ I , « li- 1 -o -< W N 1 [ 1 / \ 1 ; UJ >^ / /' \I ; p N- O \ V I ^ : ID ■n (O } \ o Q - $Q " f '^< CO 1 '} 1 •o — — ' ) i;;' 1 —J C51 t^ ~-^ *. i < 1 ( ^^ '*-._ jl Joe. •Ck. if) ..• \ ^^ « •J'* 1 .\*' 1 ^ ,.^^ •* 1 s ■^Y ' 1 ^ 03 1 ^ 1 c c r - <\^ s N ^j : • > ■ 1 •o ^ 0 u Q ) 2> J 1 '. s; ' Cvj ^«. ., 1 c = < -i CO 2 - -^ ^ 5 ;' < — 1 "7 * ', 1 a 3 3 i: lO * •^ ^ / J r-" ^ ^ A- "S, s --"* 1 c. ' ^ c sCiCbcvCiCsCi^ O Q O OQ C o o vn o . lO o>r> o IT) m 3 uj uj (_ kj -p ct: J"^ s ^ ^ 1^1 O |i < z /) LU < Q. Q Z O 1 o (0 z o o Q. Q- < a. ;n li- o in UJ 2 O o _i a. < '.O u. o lO UJ 5 JO q: in r a. O CO z _ilO _1LJ ^ 1" < ? u. fe 5^ ^>< Wisconsin Research Bulletin 7Z •Q ^ «o *n Ti If ^0\ CO t~- -o ■« ■< ~ < ' '••]., ; toO > ; J , > ^ If; ^" ^ .-- ■' ; 1 o « /• "* > ^ v^ .•■ I < — < CM s ^ *; . "^ <. I \ — ^ «•; \ — i s^ ' 2; i < s **., 't a S I /' ,„-- \ — < o * S; "H > .; N s, > ::^ _^ Si S c ->■ / I ^ _^ "V "^ "-. ' — o O ^ .J ■ / ■<: c. V s, '•• ..._ UJ ' ^ i ^ ^ ^ _,- ^ q: I •■ J < / SI -% ^ <:£* -. ', X ~ o 4 z. < 1 c^ •t,. < Q '^ s -^ -^ ,'■ ^ -a: ^ - n ^ ^ a > ; C!^ 1 i^ ^ •' 1 — — o •^ ^ r - z UJ . h- -lo: iD -iQ. 4 ' . \ ; 1 % \ ^ ^ ^ '■• \ CM — ■ ! ' t ^ lO ) !. z to ^ ^ t _^ ! 5 — % t. ~, I' 1 ; ^ ■^ o y, ^ !-* ^ \ — ' h. ^? £ \ in s z O - CO = ^ > •^ < _ V -' ^ -N in -CO^ - £ W ^', s ... - , < ^ " 1 to § o -1 0- < >- < Q- m o i in r o a. a. < 1- lO Z) Ci in 1 \ i ininininininio C7> o " «1 » - §_: s 3- s- CM s 2- 2 « - a; Sl- R 2 - « t- 21 o _ ' ■ ! "^ 1 J 0^ — »^ \s — COjj ^ < -Si L_* «. , 1 S ; J / — > »•' 1 \ ^ /" •^ .•' 1 s. -\ _ o -1 S «^ " ' - - K 5 - oe L- t ' V" 1 -- _^ (0 ^ ' ; — i - _< 7 < ' *• V "' -H — < ■"cc •^ -^ '•.._ 'k " > ' , . < t. * ~ =cr =^ * 5 L, •. \ -1 ^ ^_ .• \ J i H ' ■•• \ 1 l/ * ,,-- 1 ^ ,•■ I 1 rO < C; 1 • \ i 1 ^ -S 5 o ^v /' . a., in. <- z y' - ^ ) ••-., n|^ ■v^ <^ 1 <. - ^ \ ^ 1 3: . f' r ' \ ^ _^ ?*- ^ _L I — — — — — — — 9 IS 0 - Zm _ a- ~ — ~ < ^'^ •^ ,-' • X £ __ r.< ■' , .^^ ~i / o 9? I I " -- ^" ■ *"■ > (^ •CC " -, >v c* - , 1 ^ \ > ' / ' ,' / ,••' A y I vn ^- ^ ^ 3 "■ \ — < ^ / s > [ _^. ■ — - — ^ o ( •-., ! ; ^ - s - « - 2 ^- m - ■V ,' • if -ils^- 1 r- /- 1 ■ £ "Z " Si 8 -^ \ '■• h. I 1 '~- __ V^ — ^ , 1 " 1 - Ul 1 q: J -•' \ ^ o -+ > i^ »'.'. ! V , 1 ,...-/■ ■z - / -j - a o - o ^ - h \ *^ v\U j ^ . 1 1 r — V -i ■^ r~> ^, \ 1 — '5 I 1 ' 1 /' or: / V 1 \j -«- «i S S ^ s s s < >>- 1 y r ^ -a^ — ^-'' \ a X a v>^ : \ 1 — -| ^ \ 1 < o ' 1 I 1 '^ s, '•. 1 1 .„- § o < •■■ R - ] L. « / x' •j — < pL. S; "- < 5: :l d — ^ 1 0. ■ 1 !P 2 • ^ — i 13 ^ __ .:^ \ N ~-~ '•, • < - -- r^ It ••■ - ! ,,-- U^ < '•-. ~^ Z^<^ — -* '; < s _._ ^i ■ o 1 - 1 --V •■ — o_ -- 01 1 - . . — ^ ' — , I — ' — > < in " » o o i o z 1 3 1 1 § 1 in !n 1- itj r q: o ^ a Q ct 1- "- <) Tf M O n i! oca O O Q o o § £ a ■" t ■ ■ 2 0^ < SH r '^ - m a z O UJ ?iO o< z < El'IDKMIOLOGY AND CoNTROL OF AfPLE ScAB 13 Table I. I'he'jvkncies of Ascospohes oi- V. 1nae(juai-is in Orchard Aik in Hki.ation to Curtain Factors of the Environment, Sturgeon Bay, Wis., 1921' Date ami sub- ject of record ^ Averaged records for tw o-hour periods beginning A. M. P. M. 12 2 4 6 8 10 12 2 4 6 8 10 May Rainfall .01 .20 .14 _.. .05 .22 .33 5 37 ""5 .07 41 "I 41 ""3 .01 43 ""5 41 "5 39 "8 38 ""9 40 ""9 40 "i2 41 'is 40 'is 39 Sunshine '18 Rainfall . . 6 i 39 "ll .01 39 "ii 38 'i7 -0"6- 42 'i5 .13 -2.2- 39 '02 -0.2- 36 "is .01 —28- 41 "3 41 + 3 38 "16 .01 40 'i9 42 "17 39 "is ^.06 42 'lb 39 "17 -i"5- 39 "2i .07 42 'is 43 'i7 42 'i7 .01 41 "ie .07 40 Sunshine is f Rainfall .24 7 41 "is .14 42 "is .11 41 "i4 .09 39 '15 .08 -ii 39 "i9 39 '22 .04 38 '24 17 38 Sunshine Wind Veloc 24 Rainfall 08 8 39 "03 39 ?i3 39 'oi >■ 36 "is 38 '08 -<-l.l 37 "i4 38 Tie 41 '22 .10 40 "22 .04 39 'i9 .03 37 "21 .02 '■7 Sunshine Wind Veloc Rainfall ... 22 9 36 '22 40 5 44 ""4 36 "2i 40 ""4 40 ""3 38 'i2 39 ii 39 "9 40 ""6 40 "3 40 ""5 1,40 4 Sunshine Rainfall - 10 43 ""3 49 + 3 55 + 5 57 + 6 56 + 7 55 + 7 47 + 8 49 ""6 47 Sunshine "s Rainfall 11 53 + 3 55 + 4 59 + 5 — -6- 64 + 7 .09 61 + 8 59 + 7 59 + 8 53 + 6 48 '"6 .01 48 Sunshine '5 Rainfall .01 12 \ 50 5 49 ""3 .01 50 "1 .01 -2.2- 43 ""3 —16- 45 ""2 48 "0 .03 >► 42 "'2 55 + 1 .03 -<-289 47 5 64 + 8 .01 65 + 10 .01 -2.2- 47 "5 .03 63 + 7 .01 57 5 55 "3 .01 51 Sunshine 4 Rainfall .02 13 \ 46 ""4 .06 45 5 .05 46 47 S -i"7- 53 + 4 62 + 8 46 "6 .01 47 "6 45 43 Sunshine ■> Rainfall 14 45 "2 45 ""3 44 "3 46 5 46 + 4 49 "e 51 "8 51 "5 47 5 47 Sunshine "9 Rainfall In 1 45 ""8 46 ""6 44 + 7 49 + 11 56 + U 62 + 6 61 + 11 61 + 8 56 + 6 52 "6 51 Sunshine Wind Veloc Rainfall 16 \ 50 "'6 51 ""s S3 + 7 59 + 11 65 + 16 70 + 17 .02 54 ""e 71 + 19 58 + 4 71 + 17 -3 'fl- og + 4 66 + 15 55 "7 56 'io 58 Sunshine Wind Veloc. _ 9 Rainfall Spore Freq Temperature Sunshine Wind Veloc 17 • 62 "is 59 'is 58 '"9 58 "9 55 5 60 "7 59 ""8 52 'io 47 "io 14 Wisconsin Research Bulletin 7Z Table I. — Frequencies of Ascospores of V. Inaequalis in Orchard Air in Relation to Certain Factors of the Environment, Sturgeon Bay, Wis., 1924 — Continued. Date and sub- ject of record ^ Averaged records for two -hour periods beginning: A. M. P. M. 12 2 4 6 8 10 12 2 4 6 8 10 Rainfall .., .,. "0 46 + 9 18 44 "6 42 "4 44 + 5 48 "16 52 + 19 52 + 24 51 + 23 48 + 17 43 + 12 40 "i5 36 Sunshine "8 Rainfall --- 19 • Temperature 33 32 "2 32 "2 34 ""5 36 "7 39 "9 45 + 10 45 + 15 45 + 13 41 '"8 37 "4 .03 34 Sunshine Wind Vecloc Rainfall "6 .09 20 ] 32 ""4 32 '"5 -0"2- 40 "16 34 + 3 43 + 4 50 + 6 54 + 6 54 + 16 51 + 20 49 + 23 47 'ii 46 "8 41 Sunshine "6 Rainfall — 21 i 41 5 37 + 9 39 + 12 42 + 13 46 + 11 50 + 11 52 + 11 52 + 11 47 "7 ■'>'■< 43 "5 42 Sunshine "8 Rainfall ... 22 ] 41 "6 41 "6 44 46 'ii 46 'i2 -^85- 45 '16 45 '16 .08 45 + 16 .18 -<47>- 43 "i2 48 + 12 .08 -<31>- 42 "9 48 + 10 - 61 + 6 .03 50 >-<- > > > 49 + 9 71 P. M. 12 — 0- 62 + 7 60 67 + 10 57 56 61 50 8 2 57 62 67 16 Wisconsin Research Bulletin 73 Table I. — Frequencies of Ascospores of V. Inaequalis in Orchard Air in Relation to Certain Factors of the Environment, Sturgeon Bay, Wis., 1924 — Continued. Date and sub- ject of record* Averaged records for two -hour periods beginning: A. M. F . M. 12 2 4 6 8 10 12 2 4 6 8 10 I Rainfall 08 "7 -3"9- 59 "7 -s'.l- 45 ""5 59 "7 5".?- 45 ""3 .06 -1.5- 54 "9 0 .23 .05 ... 60 "6 61 "3 — y 63 ""3 62 + 7 69 + 11 66 + i 4 ! ! i 6; 4- S 73 + 59 + 7 67 + 13 73 76 + 9 75 + 10 67 + 11 67 + 10 73 + 6 64 + 8 68 "9 77 + 6 79 + 11 74 + 11 78 + 8 70 + 3 78 + 10 80 + 8 80 + 8 .05 1.5 58 "12 66 + 13 66 + 11 69 + 5 65 + 9 68 ""9 77 + 10 80 + 10 73 + 11 79 + 71 + 4 78 + 10 77 + 11 79 + 9 62 + 14 64 + 9 .44 <-- 64 5 63 + 8 66 "8 73 + 10 80 + 9 68 + 9 73 4- 9 72 + 4 75 + 7 76 + 10 77 + 7 .04 57 + 8 61 + 7 60 7 52 6 57 5 54 6 47 Sunshine "4 Rainfall 01 45 .04 48 + 5 57 + 6 . y 57 Sunshine "'6 '12 50 60 + 9 64 + 11 53 Sunshine "7 I Rainfall ... 18 1 46 + 7 53 + 9 58 "9 63 "4 68 + 7 74 + 7 62 + 6 66 ""8 69 63 + 2 70 + 6 70 ""6 .07 54 "8 62 "2 63 "7 69 "h 55 "7 64 54 3 8 < 59 19 56 "'9 56 '12 57 "9 62 + 14 70 + 7 71 + 7 76 + 9 Sunshine "2 Rainfall <"" 20 • 58 "i 60 '4 "0 63 "4 67 + 5 ■ ■ ->■ 62 Sunsiiine ■"§ Rainfall ^--- 21 \ 60 59 ""5 JHo- 64 "6 57 5 -^0- 58 "2 -— 6- 50 "2 "0 62 5 "6 61 "6 58 + 3 65 + 4 68 .. 8 66 "7 10 58 "i 62 ""•i 55 ""3 62 "4 7 Rai nf all ... 64 + 3 71 + 4 22 56 Sunshine "5 Rainfall ._ - ■< ■ 62 5 68 + 10 69 + 10 66 + 5 72 + 7 72 + ,8 74 + 5 .01 74 4- 67 + 5 75 + 8 76 + 7 78 + 7 .07 23 64 S'in.'hine ■5" Rainfall ... 57 + 2 62 + 4 64 2 59 67 5 "<- 63 "'9 .09 24 62 Sunshine "6 Rainfall 7 - 58 + 2 66 + 4 68 + 7 67 + 6 .01 25 63 Sunshine _.-- "i Rainfall y 62 + 2 26 63 Sunsh i ne "3 Rainfall- ... 97 62 ""6 61 + 5 a9 Sunshine '"8 Rainfall Spore Freq Temperature Sunshine Wind Veloc .47 28 57 54 "7 55 5 55 "16 56 "ii 57 "io 58 + 13 56 I2 55 "14 56 "16 53 "15 l"JMI)i:.MI(M.()(;V AND CONTKOL OK AlM'LE SCAB Taiu.i: I. I'"ui;(jii;nc,ii:s oi- Ascospokics of V. Inakqvai.is in Ohciiakd Am in Hi:i.A'ri2+ — Continued. Date and sub- ject of record - Averaged recoriis for two-hour periods beginning: A . M, P. M 12 2 4 6 8 10 12 2 4 6 8 10 1 Rainfall .13 50 "is .01 -6.5- 50 "io .02 52 "6 — ^0- 49 "6 ... ... ... ... -e's- 70 + 9 .02 60 + 5 65 7 Q ... Spore Frofj 20 \ 52 + 10 50 + 7 57 + 10 -^' 54 + 9 63 + 12 59 + 7 57 + 9 68 + 11 60 + 7 63 + 7 71 + 5 57 + 5 .24 63 "4 65 + 2 .01 56 + 6 "<- 61 + 5 59 5 53 "8 56 "3 52 '6 55 1 Sunsliinp "3 Rainfall -- o4 "(S 50 "4 ^ 30 ] 52 Suiisliine 5 July Rainfall 1 Temperature - .. - 48 + 6 52 + 9 59 + 4 51 "3 49 i ( Sun.shine "3 Rainfall 2 46 ""i 49 ""i 51 "4 45 ""i "0 49 "2 49 "2 50 + 3 56 + 4 . .. .y 60 + 5 66 + 4 71 + 6 .04 68 + 5 73 + 5 67 + 4 .25 65 + 6 68 + 6 74 + 7 74 + 5 75 + 6 .14 68 + 7 69 + 7 73 + 9 69 + 6 69 7 69 + 8 66 + 7 69 "6 65 5 68 + 6 65 + 6 .14 61 "2 63 ""4 62 "3 63 + 4 61 "3 .30 57 ""3 53 ""3 .08 52 Sunshine "2 Rainfall 3 \ 52 + 2 52 + 3 .03 60 + 2 63 + 4 .03 51 Sunshine ""3 Rainfall 60 "'6 08 "3 59 "3 .11 ■i 59 Sunshi.ie "2 Rainfall 56 55 "3 ""0 53 ""5 56 ""2 .01 56 "2 60 "3 73 + 7 73 + 7 .18 72 + 7 73 + 6 .05 55 Wind Veloc Rainfall 4 54 ""3 58 ""4 2 61 + 2 59 "i .09 69 + 4 .01 66 "2 .19 6 59 Sunshine "5 Rainfall .06 63 "7 .09 60 "4 59 "4 59 "3 58 ~ 6 59 "9 .28 7 59 Sunshine "7 [Rainfall 8 Temperature Sunshine 59 ""6 59 ""4 59 "3 .01 58 "7 .09 58 ""s .02 59 "s 63 -1- 9 .07 66 + 9 65 "7 64 -1- 6 62 '6 59 "6 'Rainfall 9 59 "7 60 "7 "5 60 "7 60 "9 61 "lb 69 + 12 '77 '70 "63 65 64 + 7 73 + 13 'so '73 '66 '67 66 "9 76 + 11 '82 '76 '68 '68 68 + 15 78 + 10 'si '79 '65 '69 70 + 14 78 + 9 "78 "78 "81 "69 69 + 9 72 + 7 "76 '72 '57 "63 64 "4 67 5 '65 "67 "52 '58 64 Sunshine io Rainfall --- 10 61 "'9 "62 60 "ii "63 .02 65 '61 '46 61 + 11 '65 .02 65 "59 "50 63 -1- 12 '73 "66 "60 "60 64 Sunshine Wind Veloc "3 11 Rainfall '64 12 [Rainfall 66 '62 "48 64 13 [Rainfall '49 14 /Rainfall 1 Temperature '57 18 Wisconsin Research Bulletin 7Z Table I. — Frequencies of Ascospores of V. Inaequalis in Orchard Air in Relation to Certain Factors of the Environment, Sturgeon Bay, Wis., 1924 — Continued. Date and sub- ject of record ^ Averaged records for twc )-hour periods beginning: A. I VI. P. M 12 2 '4 6 8 10 12 2 4 6 8 10 15 ' Rainfall '58 "60 "53 '43 '53 '57 "59 "71 "69 "68 "55 "51 "60 "67 "65 "69 "55 "48 "49 "61 "69 "64 "66 .01 59 "64 .03 57 "49 "58 "58 "52 "58 "59 "52 '43 '54 '58 '58 "70 "68 '68 "55 '56 '61 "65 '66 '68 "55 "47 "51 "61 "68 "64 "66 "58 "66 "56 "■48 '59 "55 ■49 .01 45 '60 "ei "51 '59 '60 '51 '49 55 '59 "60 "67 '68 '67 '56 "57 "65 "64 "65 "67 '54 "48 "56 "61 '67 "61 '66 '57 '65 '56 '50 '57 '54 '48 '46 .20 59 '58 '51 '65 '67 '54 57 '60 '65 '67 '69 '74 .01 66 '61 "65 '74 '70 '73 '65 "57 "57 "59 .11 61 "68 '65 '67 '63 '67 '57 '65 '59 '59 '53 '57 .01 59 '58 '54 '72 '64 "57 "58 "63 '68 '75 '75 '79 '64 '70 '73 '78 '75 '79 '67 '59 '62 '62 .36 60 '71 '67 '68 '69 '68 '62 "67 '62 '63 '58 '63 .01 63 '62 '57 '77 '67 '61 'ei '66 '73 '76 '80 'sh '63 "n "76 "80 "77 .54 74 '68 '63 "64 '6.5 .01 62 '76 '70 '69 '72 .11 66 '68 '67 '63 "66 "60 'c.5 '67 '64 '61 '82 '76 '64 '64 '69 '75 '77 '83 '87 '68 '73 '76 '79 '78 . 55 69 '68 '65 '65 '67 '70 '77 '73 '69 "74 .01 68 '69 '68 '66 '70 '62 '(')7 .03 66 '67 '64 .01 80 '72 '65 '65 '71 '74 '78 '84 '83 '69 '73 '73 '79 '78 '70 '66 '65 '66 '68 '68 '80 '72 .01 72 '76 '71 '68 '65 '67 '65 '65 '66 .07 64 '68 '65 'so '67 'ei "65 '65 '70 "75 '83 .03 75 .01 64 '76 '67 'si '75 .02 72 .46 63 '65 "65 '68 "71 '79 "71 '70 '75 '70 '65 '63 '69 '62 '65 '62 .21 64 '65 '60 .01 74 'ei '55 '59 '57 '63 .01 72 '75 '72 '62 '64 '64 '74 '70 '7i '59 '60 ".59 "6.5 '71 '70 .02 70 '67 '70 "63 "ei 'ei '62 '60 '57 '60 .51 64 '57 '.53 '68 '55 '53 '52 '56 'ei '73 '70 '73 '59 '60 '62 '69 "67 "70 "56 "53 ".52 "63 "70 '64 '69 '65 "63 "62 "58 "59 "57 ■57 '51 '58 .17 63 '55 '50 .72 61 l(i 1 Rainfall '55 17 18 Rainfall '48 Rainfall 51 19 Rainfall Temperature 56 20 Rainfall : 'ei 21 Rainfall '73 22 Rainfall '69 23 Rainfall '73 24 Rainfall 55 25 Rainfall '54 26 • Rainfall '62 27 • Rainfall '68 28 ■ Rainfall '65 29 ^ Rainfall '69 30 31 Rainfall '56 Rainfall '50 Aug 1 Rainfall '.50 2 Rainfall '62 3 Rainfall '69 4 Rainfall '63 5 Rainfall '67 6 Rainfall... '62 Rainfall 7 63 8 Rainfall Temperature . .02 60 Rainfall 53 10 Hainfall '57 11 Rainfall '57 12 Rainfall 55 13 Rainfall Temperature . '48 Rainfall .06 14 46 '60 .01 60 "52 59 15 Rainfall .01 62 Rainfall 10 54 17 f Rainfall Temperature . _ '49' Epidemiology and Control of Apple Scab 19 Table I. — Ehi:ql'i;ncii;s of Ascosporms of V. Inakqiams i.n Ohchahu Aih in Hi.i.ation to Certain Factors of the Environment, Sturgeon Bay, Wis., 1921 — Continued. Date and sub- ject of record ^ Averaged records for two-hour periods beginning: A. M. p. M. 12 2 4 6 8 10 12 2 4 6 8 10 ,^ /Rainfall ^° \ Temperature "47 .01 54 'ei '58 .19 64 62 "62 "53 "46 .15 55 "61 '58 .06 63 '58 "61 "52 .16 62 "69 '53 '60 '65 "76 '52 .04 56 "60 "59 '64 '56 "61 "54 .04 61 "66 "52 '60 '64 "72 "60 .02 57 "61 "59 "69 "65 "65 "63 '62 "65 '63 '70 "64 "68 '64 '57 "63 "60 "73 "74 "68 "70 "74 "68 "n '77 "69 "67 "67 "57 "65 'ei "76 '78 '72 "75 "si "71 '74 "si "75 "70 '66 'ei "66 "62 "80 '§5 "77 "77 . 85 "74 "77 "si "76 "74 "65 "66 '66 "67 'si 'si 'so '77 '87 '75 "75 "so "79 "75 '64 "64 '63 "69 "so "76 "76 '73 '86 '74 '70 '75 "78 "73 "85 .10 62 "ei '69 '72 '72 "64 "65 '79 "67 "64 '76 "74 "67 '55 .06 61 "57 .58 69 '65 "65 '59 "62 "78 "ei "64 "68 "7i '63 54 /Rainfall 1 Temperature .. 61 .-,„ /Rainfall ~ ITemperature 56 .,. JRainfaU \ Temperature. _ . 1.3 67 64 ^o .Rainfall 1 Temperature "63 .,. /Rainfall 58 .,- /Rainfall \ Temperature -. '62 63 '73 "54 "60 "67 75 77 .yj J Rainfall "57 2g JRainfaU '62 on [Rainfall ^Temperature. . "67 2Q , Rainfall '74 3j [Rainfall \ Temperature "64 ' .See pages 4.3-44. ' Rainfall in inches. Spore frequency in average number of ascospores per cubic foot of orchard air filtered during periods indicated by arrows. Temperature in °F. Sunshine in hours of bright sun- light. Wind velocity in miles per hour. + = Sunshine recorded. — = No sunshine recorded, or no rainfall. Spore Germination Experiments Viability and Longevity Ascospores. Frequent tests made from year to j^ear throughout the periods of natural discharge have shown that recently matured, fresh- ly ejected ascospores possess a high degree of viability, their germina- tion in sterile distilled water usually approximating 100 per cent. Under ordinary conditions, ascospores appear commonly to be discharged during the first sufficient rain period following their maturity, pro- vided the ostiole is open. Consequently, if rain periods occur with ordinary frequency, most of the spores available for discharge will have been comparatively recently matured. Numerous inoculation ex- periments have confirmed germination tests in showing that naturally discharged ascospores constitute a very uniform and highly efficient inoculum. The longevity of ascospores both in the asci and after discharge is of potential significance in relation to epidemiology. Its practical importance is minimized, however, by the following considerations : 20 W'iscoNsix Research Bulletin 7i (1) ascospores germinate readily when wet for a sufficient period at ordinar_v orchard temperatures, and (2) a new supply is ordinarily discharged during each infection period throughout the time of their production, thus furnishing a fresh inoculum. Some practical import- ance attaches to their ability (1) to withstand droughts wdiile still enclosed in the asci, and (2) to survive periods of exposure after having lodged upon a susceptible host part in a moist period too short to per- mit infection. Miss Curtis (l')22) reports liaving ol)served apprecial)le discharge of ascospores from perithecia in leaves which liad l)een exposed to an un- broken drought of three montlis' duration, but makes no statement regarding the vial)ility of these spores. The present writers, however, have observed no case in which there was ejection of ascospores after a general loss of their viability. On May 25, 1923, freshly collected air-dry apple leaves bearing perithecia which contained abundant mature ascospores were stored in a refrigerator at a temperature of 10° C. and a relative humidity of approximately 60 per cent. On Novcml)er 16, a nn'croscopic exam- ination showed the perithecia to be in ai)i)arently healtliy condition. When fragments of these leaves were moistened and placed above agar (2% agar dissolved in water) in P'etri dishes, ascospores were dis- charged ill abundance upon the surface of tlie medium. .Vjiproximately 90 per cent of these spores germinated, and cultures were readily isolated from this source. On March 12, 1925, sub-samples of a single collection of air-dry apple leaves bearing abundant perithecia were ])laced (1) in diffuse light in a chamber in which the air temperature was lield at 16° C. and the relative humidity at 78 pt'r cent, (2) in a greenliouse held at approximately' 16 (". without control of hnmidily. and (3) in a dark chaml)er (.\Umann apparatus) in which the temperature was 16° C. and the humidity was not controlled. Tests of this material on March 12 showi-d tliat ascospores were discliarged al)undantl\' wlien repre- sentative' leaf fragments were moistened, and tliat aiiproximately 100 per cent fif the ejected s]iores gernn'nated vigorousl\- in sterile dis- tilled water. On May 21, similar tests of leaf samples from each of the three sources revealed abundant discharge of ascospores, which, howi'ver, germinated InU siiarselx' in sterile distilled water. On May 22. in a iiarallel germination ex]H-riment, \\\v spores were shot ui)on the surfaces of plates f)f agar (2% agar in water). After 24 hours microscopic examinations showed flie following re-ults, listed l)y jilaces where the leaves had been stored: 16° C- -78 per cent relative humidity chamber, 95 per cent germinated, with an average germ tube length of 85 microns ; 16° C. greenhouse, 80 per cent germinated, with an average tube length of 65 microns ; 16° C. Altmann compartment, 90 per cent germinated, with an average tube length of 75 microns. Many tests and observations have shown that the development of ascospores is sharply checked when the moisture supply is reduced to the anidunts employed in Kl'lUKMIOLOGY AND COXTKOL OK Al'I'LE SCAB 21 this and the i)rtci-(hiig ixpcriiiR-nt. It i^ evident, therefore, that most, if not ah, ot the ascosjjorcs which were (hscharged at the end of these experiments had remained via!)U' in a mature condition in the asci during the periods covered l)y the^e tests. In seven years of field experience in Wisconsin, tlie writers have not encountered a drought of sufficient duration to limit the via- l)ility of mature spores contained in the asci. Severe droughts in early spring, however, have heen ohserved materially to retard and reduce the production of ascospores. ( )n July 6, 1920, ascospores from leaves which had heen held for al)out six weeks in a refrigerator at 10° C. were discharged so as to fall upon clean slides which were then lield dry in diffuse light at room temperature (connnonly 20-27 C.) and humidity (variable). One slide from this series was at once covered with droplets of sterile dis- tilled water and placed in a moist chamber at 20" C. for germination. Appro.ximately 100 per cent of the spores germinated. On August 6, in a similar germination experiment, 10 per cent of the spores on an- other slide of this series germinated. Other similar tests showed that the longevity of ascospores after discharge varied considerably with tiuir vitalitx at the time of discharge and the conditions to which tluy were subsequently exposed. The longevity of spores on glass under these experimental conditions is, of course, no index of their survival in the field. Further studies relating to the probabilities of the survival of discharged ascospores of V. incieqKalis during intervals l)etween rain periods are reported in other connections. Conidia. Numerous germination experiments have shown that con- idia freshly produced in nature possess a high degree of viability. In sterile distilled water on glass, however, the germination of con- idia is frequently much less consistent than that of ascospores. On the surfaces of agar plates or on susceptil)le host organs conidia fre- quently germinate in greater percentage and with more vigor than in water on glass. (hi July 1, 1920. severely scabbed apple leaves were collected from an unspra\ed seedling tree and placed in a standard shelter for hous- ing meteorological instruments. .At intervals conidia from these leaves were placed for germination in drops of sterile distilled water on glass slides in moist chambers at temperatures near 20° C. The results, which are summarized in Table II. show that imder these con- ditions a high percentage of the conidia survived tlie winter. It is noteworthy that the lowest temperature recorded during this period by the Madison Station of the United States Weather Bureau was — 10° F. on Dec. 28. During this winter, temperatures belo\v 0° F. occurred only four times. It should be noted that the conidia which survived the winter were protected from wetting. The writers have not succeeded in finding viable conidia of F. iiiarqualis which had sur- vived the winter under natural conditions. 22 Wisconsin Research Bulletin T^i Table II. -ReSI'LTS of TliSTS OF THE LONGEVITY OF CONIDIA OF V. InAEQUALIS, MaDISOX, Wis., 1920 Dates of tests Spores gprminated Average length of germ tubes Duration of tests 1920 July 2 % 99 3,^) 60 10 25 50 70 50 70 50 80 80 Microns 91 78 78 26 10 26 1.30 52 104 104 104 78 Hours 36 July 13 --- -- . 24 July 23 _ - 24 24 Aug. 18 -- --- - - 24 Sept. 8 - -- 24 Oct. 6 _.-. -. .. -. 24 Nov. 5 24 Dec. 3 - 24 1921 48 Feb 21 -. - -- 24 Apr. 1 --- 48 Relations of Moisture The relations of moisture to spore germination are of primary im- portance to an adequate understanding of conditions which limit in- fection. Continuous wetting. Data on the rapidity of germination of asco- spores when continuously wet at various constant temperatures appear in Figure 7. A consideration of Figure 7 and Table III suggest that the minimal periods of continuous wetting necessary to permit leaf infection at the various temperatures studied permit the development of germ tubes of fairly constant average length in the hanging drop cultures. Discontinuous wetting. Aderhold (1900, p. 562, 576-577) states that the spores of the apple and pear scab fungi germinate readily when exposed to brief periods of alternate wetting and drying and that brief periods of drying stimulate the formation of appressoria by bring- ing the germ tubes in close contact with a firm substratum. Results by Wiltshire (1915) and the present writers are in general conformity with those of Aderhold. Exposure to water vapor. On the basis of observations and of limited experiments in highly humid atmospheres, Aderhold (1900, p. 561) opines that liquid water is necessary for the germination of the spores of the apple and pear scab Fusicladia. He states, however, that complete immersion in water is not necessary, since germination occurs readily when spores float on the surface of a drop. Wiltshire (1915, p. 2>n) states that the conidia of these fungi will not germinate in an atmosphere saturated with water vapor. The results of numer- ous observations and experiments by the present writers suggest that in nature the sitores of these Fusicladia ordinarily germinate only when in contact with liquid water. The difficulties of controlling the conditions of germination tests at high humidities with such accuracy as to preclude the possibility of the presence of minute amounts of li(|ui(l water are so great, however, that fully convincing experiments on the Epidemiology and Coxtrol of Apple Scab 23 relation of liigh concentrations of water vapor to the germination of these spores are lacking. Further trials are in progress. Relations of Temperature Aderhold (1900, p. 557-559) states that conidia of the pear scab fungus germinate in open drops of rain water on glass slides through a range of temperatures extending from 2° to 30° C, the optimum appearing to be near 22°. From 11° to 22° C, the germination was vigorous, while above 22° the vigor waned rapidly. From similar experiments with ascospores of V. inacqualis. Frey (1924, p. 333) states that opti- mal germination and growth occurred at 10-18° C, and that vigor waned rapidly at higher temperatures. Ascospores from perithecia in moistened apple leaves were ejected so as to fall into sterile redistilled water. Droplets of the spore sus- pension thus obtained were transferred by means of a platinum loop to clean cover glasses and suspended as unsealed hanging drops over glass rings in moist Petri dishes. The procedure was standard- ized with the aim of having the size of the drops and the number of spores in each as uniform as feasible. These cultures were placed in duplicate for germination at a series of constant temperatures (vari- able 1° C). At stated intervals records were made from 50 spores taken at random in the cultures at each temperature, showing the per cent germinated and the length of the germ tubes. In earlier ex- periments the spores were placed in drops of water on glass slides in moist Petri dishes. In such cultures, however, there were two dis- tinct types of germ tube development, depending upon whether or not the tubes were in close contact with the surface of the glass. If they attained such contact, they were usually comparatively short and thick : if not, they were long and slender. This variable detracted somewhat from the value of comparative data on the length of germ tubes in such cultures. Furthermore, spores near the edge of drops some- times appeared to germinate more vigorously than those which rested on the surface of the slide near the- middle of the drops. Since other experiments have indicated that spores of V. inarqiialis have a com- paratively high oxygen requirement for germination, particularly as their viability wanes, it seems probable that these variations were due to inqualities in oxygen tension in different parts of the drops. The hanging drop technique was used, therefore, to prevent the con- tact of the germ tubes with a hard surface and to provide as uniform aeration as feasible, the spores accumulating in the low^er part of the drop. It was found desirable to use a comparatively small number of spores in each drop (about 100 to 200), since germination sometimes appeared to be inhibited if large numbers were closely aggregated at the bottom of the drop. The average length of germ tubes recorded in five series of tests appears in Figure 7. The data on the per cent of spores germinated are not given in detail, since they add little to the results shown by length of germ tube. Germination occurred at 24 Wisconsin Research Bulletin IZ temperatures ranging from less than 1° (after 24-36 hours at i^° C.) to Z2° C, but not at 35^ C. The optimal temperature for the elonga- tion of term tul)es under the conditions of these experiments appeared to be in thc' range df 16 to 22- C. The data of Figure 7 suggest a minor bi-modality within this optimal range. Germination and germ tube growth were quite vigorous from 11° to 22° C. Below 11 C. 0.5 '^ 6 9 II 14 16 16 22 24 26 28 30 32 35 Temperatures in Degrees Centigrade ITC. 7. -THE R]:LATI0\ ()!• IIOIPI^HATUHE TO THE GERMINATION OF AS(.<)sp()iu<:s ()i- V. iNAi':orALis in hanging drops of sterile dis- TiLLi'.i) \vati:r the devclopmient of germ tubes was normal hut retarded. At 28° C. and above, the germ tubes were much retarded in development and fre- quently assumed aberrant forms. Numerous germination tests of coiiidia of f. iiiacqiialis in open drops of sterile redistilled water on glass slides in moist chambers have shown thermal relations very similar to those just reported for asco- spores. Relations of Light No differences in development were observed when ascospores and conidia of V. hwcqualis were germinated comparatively in diffuse light and darkness. Ascospores and conidia wliicli had been placed on the upper surfaces of leaves of potted apple plants which were then ex- posed to full sunlight for periods up tO' six days germinated well. It is evident that these spores are able to tolerate considerable exposure to sunlight without losing their viability. EPIDKMIOr.OGY AND COXTROL OF Al'PLE SCAIJ 25 Infection Experiments Throughout thf coursf of this iiivostij^atiou, lumK-rous infection ex- periments have been performed, both in the field and upon potted plants in the greenhouse or out of doors. In the field experiments, suitable moisture conditions were provided by the method described on pafj;e 5 or by one of the methods devised by the senior author (1917, p. 22; 1918, p. 547) for the study of other diseases, in tlie earlier experiments in the Hrt'enhouse the jjotted plants were inoculated in a moist chamber previously described by the senior author (1918, p. 548). Conidia from various sources or naturally discharged ascospores from overwintered apple leaves were used as the inoculum. As the work on epidemiology progressed, it seemed increasingly desirable to study the details of infection and of its variability, particularly in re- lation to the play of certain factors of the natural environment. The technique decribed below was developed for this purpose. This technique proved so nuich more reliable and satisfactory than any hitherto employed that tlie results of the earlier experiments will not be given in detail, 'i'luy constitute a check on the results obtained by the later method, wit'i which they conform in all important parti- culars. Apparatus and Method for Experiments under Controlled Conditions Through the kindness of Drs. J. Cj. Dickson and James Johnson,' their apparatus for the control of air temperature and humidity was made available for our work. This equipment was very useful for studies after the fungus was sutTficiently well established in the host tissues to be independent of an external water supply. For studies during the initial stages of spore germination and infection, however, it was necessary to devise a chaml)er in which a saturated atmosphere could be maintained at a wide range of constant temperatures. This was accomplished in the apparatus shown in Figure 8. Potted apple plants of suitable size were placed in this apparatus, and inoculated by abundant natural discharge of ascospores of J'. iiiac(iiia!is from moist overwintered apple leaves which were suspended on moveable trays of wire netting in the upper part of the inner chamber. The abund- ance of the inoculum was observed b.v microscopic examination of plates exposed in representative situations in the chamber. Condensation water on the susceptible parts of the host constituted an adec|uatc sup- ply of moisture for spore germination and infection. This method pro- vides an inoculum which is remarkably uniform and free from ex- traneous material, and which in an unusual degree parallels natural conditions. The period during which an adecpiate suppl.\- of ascospores is availa- ble may be much protracted by suitable care of material. Ascospores mav be forced to maturity l)v nn'dwinter if the leaves in which they 'For an account of this apparatus sec Dickson ili)2()i ami .lolinson (1921). 26 Wisconsin Research Bulletin JZ are developing are given suitable conditions of moisture and tem- perature, or their development may be much retarded by cold storage. In certain experiments conidia or ejected ascospores, applied in sus- pension in sterile distilled water by means of atomizers, were used as the inoculum. Two potted trees (two or three years old) were used for each test. The dates at which the lesions first showed olivaceous color were recorded. After an adequate incubation period records were made of the average numiber of infected leaves per twig, and the average number of lesions on the most severely infected leaf of each twig. Only the infection on the upper surfaces of leaves was considered. The total number of twigs on the two trees of each test averaged about eight. These records were found to constitute a satisfactory basis for comparison of results. riG. 8.— INOCULATION (.nAMlii:il Tlic (Icsiicd Iciiipcratuif' and saturated hiiinidily arc iiiaintaiiic i^rcriiliouse- S -7.S' ,, chiimber.. 26°-80 % chamber_ 20°-90''f chamber _ L'O -SO' , rliMTiiliiT I'd -.Ml' , i-lKirnliir 2(I-J."i t;ri(iih(ju.si'_ L'll-'.MI' , cliainber- . lid -S(l' , chamber. 20 -.50', cliamber- Period Days 11 10 9 11 14 13 10 9 9 9 9 Final results — averaged per twig No. leave infected 3.1 3.2 3.3 2.8 1 0 0 2.2 2.5 2.9 3.1 0 2 2.9 3.1 2.2 2.7 3 3.2 0 1.7 2.6 3.3 0.8 2.6 0.7 1.3 0 3.5 0 2.3 3.1 2.8 0 0 0.9 1 0 0 2.5 3 0 3.1 3.4 2.6 0 0.7 2.3 2.8 3.8 0 3.4 3.2 2.4 2.9 3 2.9 3.1 Max. No. lesions on one leaf iTemperatures in degrees C: relative humidities in per cent. In the greenhouse, no effort was made to control humidity. The chambers used in the experiments reported in sections IX and .X are discussed on page 25. Immokmiolocy and Control of Apple Scab 31 During initial stages in moist chajnber. The experimental plants were inoculated and held in the moist chamber at various constant temperatures ranging from 6° to 28'^ C. (Table III). Infection re- sulted at temperatures ranging from 6° to 26° C. (Sect. I). It is probable that further studies will somewhat extend this range, es- pecially toward the lower limit. The data on the number of leaves infected and the maximal number of lesions per leaf agree in sug- gesting that the optimal temperature for rapid development of the initial stages of leaf infection is near 20° C. The incubation periods appear to suggest a slightly lower optimum. It should be noted, however, that the factor of error in recording incubation periods is comparatively large, due to the difficulty of determining when lesions should be classed as macroscopic. A further criterion of the relation of temperature to the initial stages of infection is the time required for the fungus to establish itself in the host sufificiently to become inde- pendent of an external water supply. The results of experiments on this aspect of infection appear in sections II-VII of Table III. In- fection occurred freely at 6° C. the lowest temperature tried, al- though the process was much retarded. The periods necessary for in- fection were progressively shortened at temperatures of 9°, 15° and 20° C. At 24°, as at 20° C, infection occurred when plants were sub- jected to a six-hour moist period, but not when the period was cut to four hours. At 26° C. the period was considerably lengthened. It is noteworthy that, in all of these tests, an increased amount of in- fection attended the longer periods in moist chamber. The data now available suggest that the optimal temperature for rapid development of the initial stages of leaf infection is near 20° C. During incubation. From limited data (Table III. Sect. IX) it ao- pears that the effect of temperature during the incubation period par- allels its influence upon the initial stages of infection. At 8° C. the minimal period of incubation was prolonged to 17 days, as com- pared with periods commonly ranging from 8 to 12 days at temper- atures of 20-25° C. When the temperature during incubation was raised to 26° C, the disease developed sparsely after 13 days in one trial and failed to develop in two. It is evident, however, that in nature the organism frequently tolerates periods of summer heat in which temperatures materially exceed 26° C. The effect of intermit- tent temperatures is, therefore, of interest. Four successive daily exposures of 8 hours each at 31° C, beginning the second day after inoculation, did not preclude the development of the disease, though they appeared to inhibit it slightly. A single exposure of 24 hours at 31-32° C. did not preclude scab development. Similar exposures of 48 hours or more, however, prevented macroscopic development of lesions. These exposures to high temperatures were made in the chambers discussed on p. 25 at relative humidities near 80 per cent. The experimental plants were incubated in 20-25 °C. greenhouse. To parasite and host. The experiments reported on page 24 show 32 Wisconsin Research Bulletin 7?i that in sterile distilled water ascospores and conidia cif V. inacqualis germinate through a range of temperatures beginning at less tlian 1° and extending to 32° C, with an optimal rate of germ tube elongation in the range of 16° to 22- C. Limited observations on the rate of growth of the fungus on various agar infusions suggest that the temperature relations for mycelial growth on these media closely paral- lel those for germination of spores and elongation of germ tubes. While no extensive experiments have been made relative to the thermal rela- tions of the host plant, observations of its behavior in the inoculation experiments in which temperature was controlled, supplemented by field experience, indicate that it has a distinctly higher temperature range than has the scab fungus. It is worthy of note that, within the temperatures at which the disease will develop, its thermal relations agree very closely with those for germination and growth of the parasite. Relations of Moisture The results of experiments on the relations of moisture to in- fection appear in Tables III and IV. The minimal periods of continuous wetting necessary for infection. The results thus far available (Table III, Sect. II-VTI) sugge^-t that the minimal periods of continuous wetting necessary for leaf infection by ascospores fall within the following limits for the temperatures listed: 6°, 13 to 18 hours; 9'', 9 to 11 lidurs; 15°, not more than 8.5 hours; 20°. 4 to 6 hom-s; 24 \ 4 to 6 hours; 26° C, 8 to 10 hours. It it notewortli\- that an increased anumnt of infection accomjianii'd pro- longation of the moist periods to the linn'ts reached in these tests. The longer periods of wetting undoubtedl\- increase the chances for inl\Tti(in ])y llie less vigorous and less favorably- placed spores, as well as those which happen to jml out tlu'ir germ tubes in directions which do not lead to an early contact with the cuticle of the host. Frcini these data and from fudd records, il is a]iparent that most iiatur;il infections re(|uire soniewbat longer periods of welting than the niinimrd jieriods indicated in Table III. Discontinuous wetting in relation to infection. .\derliold ( I'HIO. p. 562. 576-577) sbowt'd that sjiores of the ai)i)le and jiear scab fungi germinate well when exposed to brief jieriods of alternate wetting and drying, and ojjined that suitable di-eontinuous wt'tting favors infec- tion by bringing germ tubes in contact with the cuticle and stinndat- ing the formation of appressoria. The results recorded in Table W show that abundant infection re- sulted from inoculations made under conditions of discontinuf)Us wet- ting. At 20° C. two successive wet periods of four hours each were sufficient to permit infection (Inoc. 3). In one experiment three suc- cessive wet periods of four hours each appeared to be as efifective as ElMDKMlOLOGY AND CONTROL OF ApPLE ScAB 33 1 1 « 1 c 1 o_ 'w 5! i — IN QO !0 -; 4i O O 1- C-l •* ff. ^1 o ococo CO *?• 'X *X O 3 •"■ 0 2 C'l CO C-IM , .ai z o \ 'f • = (o« 5 5 " t. OJ " is. ^-^ Sl'l Final averagi leaves fected CO O M CO CO lO o ■r ■M ■>i M CO I-: t- CO -i ■M O ?l M — ^ CO = S CI 3 d = Z 1 c~ 3 O.O 0} as J; & 2 .-t -r -t -f'^' cj i-rco o o 10 .j: r- -c -n ►-ja a 1 , , _. — - 1 ( , Ih ' . 1 ;;*!::; ; 1 !•* 1 1 1 1 1 IJ5 1 ' '-S i '.r^ i i io i i i 1 ! ;5 i i iP i io 1 ' iin 1 ' lo ' ^ ; 1 in ! 1 c ! ;o^> ; . . , I 1 >o 1 "o ;! 'VW'So-o'o- ,"^0 = 0 = ; - -r ^ ^1 ^) , >. >> ?o >>>> >> ?0 >! >i°0 >. >i >v M 'ii d S3 3i M ea cS IS C^l 3i ri M cj S ci 6 'XX -3-3T3 ^■o^t: c - - d — "^ c^ t -4< ^ Tt> u- t^ (M rt ,-. —1 (M - — M - - -- 'E .s t4 <^ Tf ^^ XT ?" ?■ M M ' ' —'"•*■'*' » cu ■€ 1 3 1 ^ 1 ^ 1 c 2 E" ""^ -^ 'y ^' ?i ?i ?i M c-1 ^^ :-) :-! C-) M ?i ?i 01 ?i ?i "i S 1 ^ ci "s 2 X 3C ^ s — C^ i.- ^^ ro Tf ■- -o r~ >.xc-. c- ^ •-, ^- _ ,- .^ t- >: ^ 6 1 ~ ~- rr. ^ ! ~Z 1 11 34 Wisconsin Research Bulletin 7Z one 24-hour ptriod (Inocs. 4 and 1). In 1''24 infection occurred on the youngest leaves of a potted apple plant which had been exposed to ascospore infection for two hours in moist chamber at 22° C, in- cubated out of doors in the day and in the greenhouse at night for six days without being wet, and placed in the moist chamljer at ll"^ C. for 24 hours. Humidity in relation to infection. In considering tlie relations of huniidit\- to infection, at least three aspects should be taken into ac- count: (1) its effect upon the host plant i)rior to infection, i)articu- larly in relation to the development of the cuticle. (2) its possible in- fluence ui>on spore germination and the initial stages of infection, and (3) its effect during the period of incubation. The data now available (Table III, Sect. X) ai^'ly only to the last of these phases. These limited experimental variations in humiditv do not apjiear to have exerted an\' significant infiuence upon the development of the disease. Relations of Light In relation to the possibilities of natural infection during discontin- uous ptM'idds of wetting, it is desirable to know whetlur exposure to full sunlight during the dry intervals will check infection. In the ex])erinient recorded above it was shuun that exposure nf the experimental plant to full sunlight for six days, following a two-hour imiculation period in the moist chamber, did nut preclude infection on those leaves which j-ad Udt i)eciime resistant wbe'i the ])lant was returned to the mi)ist chamber. Similar exixisures of shorter duration (Table W , Inocs. iS-lS) appear to have e.xerted no inhibiting influence on infection. Inasmuch as most of the exin'rimenlal ])lants used in the inoculation work were grown and incubated under glass, which in the later spring and sunnuer must be shaded by suitable spraxs to avoid excessive tem- peratures, some simple experiments were made to test thi' eltecls of different degrees of shading during the incubation period. The re- sults ap|)ear in Table 1\', Inocs. 1. .^-1. 577-579; Cliiitf)n, 1901, p. 114; W'iltsliirv, 1915, p. 339) liavc observed that organs of apple and ])ear are iiiiuh more susceptible to scab infection when young than wlieii mature. In tlie infection studits of the present writers, hundreds of inoculation exiierinients have slujwn a striking relation between the stage of developnien.t of a])i)le leaves and tluir suscepti- bility to scab. Tile bulk of tlie detailed data i)rechides jjublishing the results of more tlian tlie few experiments included in Table \'. This work has shown that young growing apple leaves of all the varieties studied pass through a stage of maximal susceiitibility into a period of increasing resistance. It is of potential significance in relation to the nature of this resistance that it develoi)s at different rates and in different degrees in certain parts of tlie leaf. The ventral surface of the lamina (usually uppermost when the leaf is mature) leads in the rapidity and degree of development of resistance. Even on this sur- face, however, the development of resistance is not uniform, the larger veins lagging behind the remainder of the lamina. Not infretiuently the disease develops rather generally over the midrib and larger veins while the adjacent surface of the lamina remains clean. Similarly, lesions are often observed to extend farther along the larger veins than over the adjacent surface. The dorsal surface lags behind the ventral both in rate and degree of ac(iuisition of resistance. On this surface, in- creased resistance is manifested through a much prolonged incubation period and through a sparse and often diffuse and inconspicuous de- velopn;ent of the fungus. This is well illustrated in Table V. Inoc. 2, wliere the disease developed al)undaiitly on the dorsal surface even of the oldest leaf, but only after an incubation period of between 39 and 55 days. The data in this talile show a rather consistent increase in the period of incul)ation for dorsal surface infection with the age of the leaf. While data from inoculation studies on fruit and twigs are frag- mentary, it is apparent from field observations and from the inocula- tion results available that all susceptible parts of the apple plant in- crease their resistance to I', iiiacqiialis with age. The fruit lesions which result from late-season infections are smaller than those which occur wlien tlie fruit is _\ounger. Since the external environment is frequently very favoralde for scab infection in the fall, these differences can scarcely be attributed wholly to external factors. Bagging experi- ments on Shields (crab) in 1919, which will not be reported in detail, showed that pedicels which were protected from infection until ten days after petal-fall resisted natural infection throughout the season, while those which were continuoiisl>' exposed became abundantly' dis- eased. Scab infection periods occurred after the bags were removed. In all of the inoculation studies by the writers, no twig infection has 36 Wisconsin Research Bulletin 71) "2" , ooocooooocco oooooooooooo OOOOOOOOOOO I '-'.-( |. oooooooooooo I -^ Ot^'^'^'^OX'OcCCDOOCO •lO \ oooooooooooo T'-lto K COCOMPOCOrOCCCCCCmMCOfO OMiNC^OOXOOGCOOCCXaO 9 I Q O'^C-OOlOOOiClOOlOlCi OtOCOOCDiOCOCDtOdcDtD I'-ir^ 1.^ O^ > ^ >C<5C 12 Q OOOOO0)iMC005OO4_ I OOOOOOOO'O C B C-)- ,0 I , OOOOOOOOOOOO ©OOOOOOOOOOOO .OOOOOOOOOOCD OOOOOOOOOOO c c OOOOOOOOOOOO I "'o 1 ^ OOOOOOOCOOOOO I '- 1 -^ OOOOOOO-hOO C- £ I ^ oooooooooooo I _ooooooooocccc oocoooooooooo £2 Q 00000000:0 CO c '-OCO C-)- 2 Q OOOOOOOOOO'OO c oooooooooooo ooooooooooooo ,00000000 'OiO S- .,_ oooooooooooo I ooooooooooooo ooooooooooooo OOOOOOOO'J'Tj" C-)- (N 1^ OOOOOOOOOOOO OOOOOOOOOOOOO OOOOOOOOOOOOO t-d6o'EOOoo6o>. ^Ti'C'^ (2,13 "3 T) T! -S T3 J3 fc^ 6 6 6 6'C 6 o o o o >i o ^-3-5-3-3 Q,-3-3'3-3-3^-3 O -S I ^ El'IDEMIOLO(iY AND CoNTKOL OF xA-I'l'LE SCAB Z7 MO00««0000000000 I ,„ _ CMMOOOOOOO I ?>■ ^ « ,-< CC CO CO CO CO CC CO lO i°°®S?S+++++ oocoooxt^r^t^t^t^ I (N C^ CO CO CO CO CO I OOOOO— 1 25++++ •i-i X 3 ,0000000 a ++ OOOOOOOOOOO 7--H X OOOOOOO c ++ OOOOOOOOOO I — X , OOOOOOOOiC 2 + _0000000000 I t-iQ .OOOOOOOO"^. OOOOOOOOOO I —!( .OOOOOOO OOC I OOOOOOOOOO I ooooooooo OOOOOOOOOO _ X .OOOOOOOOOO I OOOOOOOOOO 1 .OOOOOOOOOO OOOOOOOOOO I OOOOOOOOOO OOOOOOOOOO I " O O O ■■-: >r: I.- .C ■o I ,^ X 1 '-'® , OOOOOOOO >+++++ 1 f. . OOOSCSO-t- ,ooo=ss++ OOCOCOCOCOiOiO , oo ooo a c- OOOOOOOO I -H, . OOOOO C C- OOOOOOOO ^oooooo a-)- oooooooo .oooooo a- OOOOOOOO I ^OOOOOOO-j- oooooooo .OOOOOOO 2 + OOOOOOOO .OOOOOOO a OOOOOOOO I OOOOOOOO I OOOOOOOO o-' — cococococo . O O O O 'C lO "O lO ; — I ^ CI c>) M "-■; CJ M -x: --O -^ -i » OO-*OO0-. «30 J N C^ C>1 -f i.O ^oooooo — o OOOOO OOl^ .OOOOOOOO OOOOOOOO . OOOOOOOO OOOOOOOO . OOOOOOOO OOOOOOOO OOOOOOOO OOO o.o ooo ^OOOOOOOO OOOOOOOO .OOOOOOOO OOOOOOOO ^ OOOOOOOO OOOOOOOO OOOOOOOO OOOOOOOO C, o o t>-03'--*t~C0-H«D -2 ^ ^ ^ X — -; 4/ 4) 2 s ~ cs o so > a a To M a 2 o - S 3 ~ 1^ a> S j2 O—* r-^ « a- o c^ ■~ (1> O s o a to T) 1^ at; M 5 1) ^ a CJ 3 M o M ■3 = u :s 3 O is j^ -^ -*j a C3 c Oj s C<1 0 2 03 _>. ■s ?) _o 5' 5 > 1- c c - 0 3 o u ~l c t. C3 ^ s. S 4. - C :! s? ^ ■;^ ^ .- t. = # ^ 3 t=. = ■^^ = t. a ■" >> t-n-^ ? a s . > S ° a c = S.2 53 t;~ ^ ^ b a £ — o « aa: — 5 a-? ■2.3 % «:=i:s g „ '"o 5 ^ a II a <^^E^TJH+-S S3 38 Wisconsin Research Bulletin IZ 3 te I . ^ 1^1 J° 'X o o •* o o fc > s fe o ;2 z ~ (N O O C^ O O -H -^ t^ -X C-. r~ -^ -r C-. -> CC O O CO O O CO CC C^l C^l C>l C^I CO C-I C] cc c^i cc M O S — C O 3 O o - rt o 1 'N M M CM M n C-l M M 3 -So J ^o J 3 i- O 3 +^ St. -, „ _ J; ^ _L -f M M i'K^-afflctcc P(-!-cfflMW cc M T3 P3 a; 02 ►-; -c P5 tc CO Phn-oBtBtB (33 o_o o o 'X,2. °.S-£3h^IE^ °^^K^^^^ ;-c — "a -3 -v-Y-r-r CO CO M CO CO CO CO c^ CO CO CO CO CO CO CO CO CO CO ^ -^ ■* ■* Tf t)< MM CI CI cocococococococococococococococococo (M(NNMl M CM J'jmi)i;m i(jlo(;v and Control of Apple Scab 39 been observed. Only in rare instances has twig infection Ijeen ob- served in Wisconsin, except on certain unsprayed crab trees. It is evident that this type of infection is limited by a very narrow range of conditions. These data offer strong indication, thougli not conclusive evidence, that the cuticle ])]ays a very important part in determining the sus- ceptil)ilit.\- or rt-sistance of host organs to scab infection. It is pro- jected tliat a lurtlier discussion of this phase of the subject will be gi\en in a later jiaiier. Relations of Certain Fungicidal Treatments The literature of spraying and dusting contains many apparently conflicting reports concerning the effectiveness of fungicides. The many factors which contril)ute to tliese discrepancies will not be dis- cussed here. One potentially important consideration, however, is the influence of environmental factors, particularly moisture and tem- perature, upon the effectiveness of fungicidal treatments. While many field observations have I)een made concerning these relationsliips, little work has been flone under conditions in which tiie fungicidal appli- cations could be related at will to the experimental production of the disease under controlled conditions. Such studies, which were made possible by tiie apparatus and metliod descril)ed above, have been started on aiijile scab in tlie hope of gaining a clearer understanding of the efficiency and adaptations of fungicides in combating this dis- ease under the range of conditions commonly met in practice. Onh- a small beginning has been made on this work. The results now avail- able appear in Table \'I. The most striking feature of these results is the inefficiency of Bordeaux mixture, as compared with tlie sulphur fungicides, under the conditions of these tests. The Bordeaux sprayed plants, in all cases, developed essentially as severe scab infection as did the con- trols, wiiereas all tlie suli)hur treatments tested commonly gave ex- cillent control of the disease if applied prior to the beginning of the infection period. These results with Bordeaux are at variance with most reports of field experience, and with the connnonly prevalent idea that it is a more effective fungicide than lime-svilphur for apple scab control. The results of spraying experiments reported herein con- firm very extensive data of other investigators in showing that in practice Bordeaux mixture can effectively control scab on the fruit. The present writers' field results do not, however, bear out the idea that Bordeaux mixture is more effective than lime-sulphur for this purpose, except as each application of Bordeaux is effective through- out a longer period of adverse weather conditions than is a treat- n-:ent with lime-sulphur. In certain programs, therefore, critical per- iods for scab control may come at a time when the lime-sulphur ap- plications arc more in need of renewal than the Bordeaux, and lead to 40 Wisconsin Research Bulletin TS results indicative of superior efficacy of the latter spray. This super- iority' is very real under the conditions stated, l)ut seems to disappear when the programs provide the appropriate time and number of appli- cations of lime-sulphur. The following explanation of the occurrence of abundant scab infection on leaves recently sprayed with Bordeaux, as reported in Table VI, is offered tentatively, pending further study. As pointed out by Barker and Giminghami (1911-12), the more import- ant possil)le modes of rendering the "insoluble" copper of Bordeaux mixture residues soluble, and therefore available for fungicidal ac- tivity, may be grouped under three headings: (1) the solvent action of inorganic compounds in the air or in meteoric water, notably car- bon dioxide; (2) the solvent action of compounds emanating from the host plant; and (3) the solvent action of compounds emanating from the fungus. In the infection experiments under consideration (Table VI), it seems unlikely that the Bordeaux residues had weathered long enough for any considerable solvent action to have been exerted upon the copper compounds by inorganic chemical compounds from the air or the water which condensed uptMi tlie ])lants. It seenis unlikely that any considerable solvent action had occurred from compounds from witliin tin- leaves, since the experimental plants had been grown in the greenhouse under conditions which would minimize wounding, and the work was confined to the upper (ventral) leaf surfaces, upon which the cuticle is highly resistant to exosmosis and few if any stom- ata occur. It would appear probable, therefore, that the solvent ac- tion responsible for fungicidal activity undiT these conditions resulted chiefly fmm materials which exosmosed from the germ tubes of the germinating spores in contact with or close proximity to particles of the Bordeaux residues (see Barker and Gimingham, lQll-12, p. 82-90, and Aderhold, 1899, p. 262-271). However, the detailed studies of penetration of the cuticle by the fungus under llu- conditidus of these experiments, as rejjnrted above ( ]). 29), revealed the liitlierto un- observed fact that the infection hyi)liae very connnonly grew directly froni tile ascospores, which themselves functioned as appressoria, witli- u\\\ tile development of distinctly differentiated germ tubes. In this way infection was accomplished within >i.\ hours at 20'^ C. and with a minimal exposure of delicate-walled fungal liyidiae to tlie toxic action of the fungicide. Tile extent of tiiis type of infection in nature, and whether the infection hyphae can penetrate thin films of the Bordeaux mixture or must chance to strike unprotected leaf surface, are sub- jects for further investigation. .'\ ])oint of sjiecial practical and theoretical interest is the apparent effectiveness of the various sul])luir treatments, including sulphur dust used alone, at 6° C. These data are in accord with the results from field experiments reported herein and similar results of other workers, in which satisfactory control of scab was repeatedly effected by linie-sulidiur and other sulphur fungicides at comparatively low temperatures in early spring. They are, therefore, at variance with EriUEMlUl.U(iY AND CONTROL OF ApPLE SCAB 41 the view of those investigators who have maintained that finely di- vided sulphur is not effective as a fungicide, save at comparatively high temperatures (see Table VI, Inocs. 25-27, 29-30). The experiments in which the fungicides were applied after infec- tion occurred are too limited to be conclusive. In certain instances, they suggest some inhibition by the lime-sulphur preparations. In preliminary experiments in 1924, which will not be reported in detail, distinct iiihil)ition of scab development on apple leaves attended spray- ing with lime-sulphur, 1-40, plus arsenate of lead, 1-50, 24 hours after the infection period began. Studies of the relation of fungicides to infection under controlled conditions are being continued. The Development of Epidemics In the light of the results recorded in the foregoing sections, it is sought in the following pages to trace the development of epidemics and to define critical periods in their development and control. Modes of Overwintering of the Fungus It is now generally accepted that the most important mode of over- wintering of the apple scab parasite is by the formation of the as- cigerous stage in the dead leaves on the ground, as first scientifically demonstrated by Aderhold (1894, 1896) and later confirmed by many other investigators. In the studies of the present writers, the as- cigerous stage has been found to occur in such abundance and to mature at such periods as to furnish a source of inoculum which has seemed suflficient to account for all the primary infection which they have observed. It has long been known that the scab fungus may overwinter on infected apple twigs. The amount of twig infection varies greatly with varieties and with environmental conditions. The literature bear- ing on this subject has been reviewed by Wallace (1913, p. 577-578) and Morse and Darrow (1913). This type of overwintering appears to be of minor importance in relation to epidemiology, save on certain very susceptible varieties in environments which unusually favor twig infection. In Wisconsin, the observations of the present writers during a period of ten years have revealed no instance in which overwintering of the fungus in twig lesions was of practical significance. It has been suggested that the scab fungus may overwinter in the conidial stage, or by means of appressoria or of superficial mycelial growth developed from spores which germinate upon the surfaces of buas, twigs, or other host parts (see discussion by Wallace 1913. p. 576-578). In extensive studies of field material, the present writers have 42 Wisconsin Research Bulletin 7}, found no evidence to indicate that any of tliese types of overwintering play a significant role in the life history of the fungus under Wisconsin conditions. The information now available appears to justify the conclusion that, under present conditions of commercial apple culture in Wisconsin, the scab fungus overwinters to a significant extent only through forma- tion of the perithecial stage in dead leaves. Production and Dissemination of Ascospores Since ascospores aj^pear to be by far the most important inoculum for primar\' infection, and the onl\- primary inoculum of economic sig- nificance under Wisconsin condition, an adequate understanding of their production, dissemination, and role in epidemiology is of first importance in relation to control measures. Aderhold (1900, p. 583) observed that perithecia of the ai)i)le and pear scab fungi matured be- fore the blossoms of their respective host plants opened, and states that the first application of spray would be too late if made after petal-fall. Clinton (1901. i). 121) notes that perithecia containing ma- ture ascospores were found in Illinois in April and May. Lawrence (1904, p. 6-7), having worked in Washington, writes: "About the time the flower buds connnence to open the spores of the winter stage are matured and set free." As a result of his observations in New York, Wallace (191,?, p. 559-560) states: "In nature the ascos])ores usually begin to mature at or about the time when the apple blossoms are ready to oiien . . . ;in(l the rii)ening i)rocess may continue for about one month." 'i"he idea of a synchronism (_)f the maturity of |)erithccia of the scab fungus and the (jpening of the aiijile blossoms has had wide acceptance, and was for nviny years a leading consideration in timing the first fungicidal application for seal) control just before the blos- som buds opened. Working under comparatively mild and moist cli- matic conditions at Hood River, (Oregon, however, (liilds (1917) found that ascosjjore discharge began in 1916 before the first ajiple buds opened and continued at intervals for more than three months. In 1917, in consecpience of the unsatisfactory control ol scab in the experiments reported above, Vxcy and Keitt (1''25) beg;in a stud\' of spore dissemination of V. iiiacqiialis in relation to the sea.sonal develop- ment of apple scab. By means of a filtration apparatus devised for the ])ur]K)se, frc(|nencies of ascospores in orchard air wt're determini-d throughout llie season. The first natural discharge, which was much retarded by drought, was recorded on May 19 (wlu'u most of the blossoms had opened) and the last on Jvdy 18. The maximal con- centration of ascospores, 71 i)er cubic foot of orchard air filtered, was registered on .\la\ 21. Significant discharges occurred only when the leaves were wet by rain. If abundant asci were in condition to eject their spores when wet. heavy discharge started soon after the rain began, and continued with continuous wetting for periods varying from Epidemiology and Control of Apple Scab 43 3 to 15 hours. Following a continuation and extension of these studies in 1919, the senior author of the present paper (1920, p. 58) records heavy discharges of ascospores at Madison and Sturgeon Bay, Wis- consin, well in advance of the apple blossoming period, and reports failure of the four-spray programs of lime-sulphur or Bordeaux mix- ture then in common use in many sections to control scab adequately except when an additional application was made soon after the blossom buds became exposed in the clusters. Further preliminary reports up- on the continuation of this work by the same author (1921, 1922) and the present writers (1924 a and b) have recorded additional evidence of the occurrence and importance of early discharges of ascospores. Miss Curtis (1921. 1922), Bennett (1923), Krout (1923), Schneiderhan and Fromme (1924), Williams (1924), Adams (1925), Schneiderhan (1926), and others have contributed to the development of a valuable body of data concerning the time of ascospore discharges. This work, in general, points to the importance of early discharges, but indicates much variation with seasonal and regional climatic conditions. Apparatus, Methods, and Records of Field Studies The work of the present writers on the production and discharge of ascospores in nature covers a period of seven years. The simpler studies, which were made by a slight modification of the method used by Wallace (1913) and Childs (1917), are described on page 4 and 1-IG. 9.— INTAKE OF THE FILT1\ATI0N DEVICE USED FOR DETERMINING THE FREQUENCIES OF SPORES OF V. INAEQUALIS IN ORCHARD AIR reported graphically in Figures 1-6. The more intensive studies were made at Sturgeon Bay in 1924 through the use of the following appara- tus and technique (Keitt and Jones, 1925 a). A nitrocellulose filter was placed in a suitably fashioned glass tube (Fig. 9) which was connected by thick-walled rubber tubing to the in- take of a motor-driven blower with suction capacity of approximatel}" one cubic foot per minute (PI. \'I, A). A gas meter was con- nected in series between the filter and the blower to measure the vol- ume of air filtered. At the end of each run, the nitrocellulose filter, bearing the spores caught, was placed in a straight-walled flat-bottomed 44 Wisconsin Research Bulletin 73 glass container of suitable size and dissolved in a liquid composed of three parts bj' volume of ether to one of absolute alcohol. This pre- paration was agitated in a manner designed to facilitate uniform distribution of settling spores, and allowed to evaporate to a thin layer of transparent gel, in which the spores were distributed at but little variance from an optical plane. Counts were then made of the number of ascospores of V. iuacqiiaUs in 50 microscopic fields. The use of a mechanical stage permitted random selection from represen- tative parts of the film without duplication. The olivaceous color and characteristic form and size of the ascospores of the scab fungus, and the scarcity' of other spores in the orchard air during rains, made iden- tification easy. From these data the average number of ascospores per cubic foot of orchard air filtered in each run was computed. This method is a modification of one employed by Pasteur (1SC)2) in his early investigations of the microbial content of air. The resistance offered by the filter occasioned a factor of error because of the resulting slight attenuation of the air when it was measured. The difference in pressure between the air as it entered the meter and that outside was determined experimentally lor ten filters through the use of a suitably ccmnected U-tube of mercury. The re- sults, which showed only minor variations, were averaged, and by the application of Boyle's law the factor of error was computed to be 8 per cent of the volume measured. Correctif)ns were made according- ly, and care was exercised to make the filters as nearly uniform as feasible in size and resistance. The apparatus just described was set uj) in the Dudley block of the Sackett orchard on the afternoon of Alay 6, V)Z4. and run as con- tinuously as feasible until natural discharge of ascospores ceased. The Dudley block of this orchard was abundantly strewn with leaves bear- ing the perithecia of J\ iiiacqiialis. Although tiie orchard was disked shortly before the- blooming period and at frequent intervals there- after, no special eiYort was made to cultivate close to the bases of the trees or to accomplish a maximun of sanitation b_\- turning under the old leaves. The result of these (.•x])erinunts, wliich are summarized in Table I, are discussed in the following ])aragraphs in relation to data from other sources. Time of Earliest Maturity of Ascospores Data on time of earliest maturit\- of ascosjiores in relation to the blooming period of the apple appear in TabK> XTI. Representative il- lustrations of stages of advancement of the host plant in relation to these data appear in Plates I- 1 II. From these records it is evi- dent that under Wisconsin conditions ascospores of V. maequaUs are commonly available for discharge by the time of first exposure of sus- ceptible host tissue to infection. However, there may be considerable Epidemiology and Control of Api-le Scab 45 "variation in tlif rilativc ratrs of tlcvflopniunt of host and parasite, since all environnirntal factor.s do nut exert an e(iual influence on botli. Periods of Ascospore Discharge The periods over which natural discharges of ascosporcs were ob- served to occur at intervals in the seasons covered by these studies, as shown in Figures 1-6 and Table I, varied from about five to nine weeks. The number and duration of the individual discharge periods and the i|uantit\ of ascosporcs ejected were influenced greatly by the amount and distribution of rainfall. In the event that there was sufTicient rainfall in tiie early spring, the major discharges occurred well in advance of the blooming period, as at Sturgeon Bay in 1919 and 1924 (Figs. 1 and 6). If the early spring was very dry, however, as at Sturgeon Bay in 1920, 1921, and 1923 (Figs. 2, 3, and 5), dis- charge was delayed or in extreme cases much reduced. Frequencies of Ascospores in Orchard Air In the filtration experiments of 1924 (Table I), the maximal fre- quency of ascospores in orchard air was recorded on Alay 13, 17 days before the beginning of the blooming period of Wealthy. Dur- ing a five-hour period ascospores were caught at the average rate of 289 per cubic foot of air filtered. Catches at the average rate of from 13 to 85 ascospores per cubic foot were made on May 10, 14, and 23, and June 6 and 9. A number of minor discharges were recorded. These frequencies are even higher than those reported by Frey and Keitt, in which the maximum was 71 per cubic foot. Table VII. — Time of Maturity of Ascospores of V. In'aequ.\lis in Relation to the Blooming Period of the Apple Year Date of first observation of: Place Ascospores mature in nature .■\scospore discharge from freshly col- lected leaves Open blossoms on Wealthy Wis. Madison do 1916 1917 1919 1920 1921 1922 1923 1924 1919 1920 1921 1922 1923 1924 April 26 April 19 April 17 April 21 April 14 April 13 April 23 March 22 May 3 Mav 17> Mav 12' May S' Mav 5 May 5> .Vp"rir26 April 17 April 21 April 14 April 14 March 22 Mav 3 May 17' May 12' May S' May" 5' Mav 18 do May 15 Mav 19 do do Mav 3 do .- - . - Mav 8 do May 15 Mav 20 do Sturgeon Bay Mav 28 do Mav 31 do Mav 16 do Mav 21 do May 26 Mav 30 do — -. --- 'The field laboratory was not opened in time to permit a record of the earliest maturity and dis- charge of ascospores. 46 Wisconsin Research Bulletin 7?i Some Factors Which Affect Production of Perithecia and Discharge of Ascospores An increasing appreciation of the importance of the time and amount ascospore infection in relation to epidemiology and control led the writers, with the assistance of Mr. E. E. Wilson, to begin a detailed study of major factors which affect the development of perithecia and the discharge of ascospores. In view of the fact that full responsibility for these phases of tiie work has now been undertaken by Mr. Wilson, the present discussion is limited to a brief preliminary re])ort of the earlier results (see Keitt and Wilson, 1926). A study of infected leaves which were removed from tlie tree at in- tervals from August to February and placed on the ground in the or- chard under suitable conditions for the natural development of peri- thecia showed a marked relation between the time of leaf-fall and the time of maturity of ascocarps. In confirmation of the work of Killian (1017) and others, it was found that the fungus was conunonly confined to its tyi)ical sub-cuticular position so long as the leaves remained alive on the trees, but soon after leaf-fall began to ramify the lamina with a vigorous mycelial development. Perithecia de- veloped abundantly l)oth in the older and the younger leaves, and in those which showed only a sparse diffuse infection on the under (dor- sal) surface as well as those which bore large, well-defined lesions. Temperature and moisture appeared to be the cardinal environmental factors governing the rate of development of ascocarps. Leaves in which the most advanced perithecia were in an early stage of ascus formation were subjected to controlled conditions of temperature and moisttne. In moi^t chambers in dark compartments (Altmann appar- atus), the perithecia matured comparatively slowly at 4 and 7" l". The rapidity of development was successively increased at 12 , U) , and 20° C. At 24" C, only occasional asci reached maturitx. The optimal temperature for rapidity of develo])ment in the later matura- tion stages of the ascocarps appears, therefore, to be near 20° C. At 16''' (". and relative humidity of 78 per cent, in one of the chambers used in the inii'Ctioii stndieN ( ]). IS), the niJituration ol jieritbecia was checked until the leaves were thoroughly wet. Discontinuous wetting led to niort' rai)id m.'ituration than continuous wetting. These results agree with fielil observ;itions at Madison in l')17. 101'), and 1024, when tlu- development of perithecia was definitely checked by dry weather in earl\- spring. As pointed out by Miss Curtis (1022) and Frey and Keitt (1925) and further confirmed 1)\- the data in Table I, an adequate supply of water is the chief requisite for the ijectioii of ascospores from asci which are in condition for discharge. In the present writers' experi- ments wetting by dew vvas never observed to be sul'licient to oc- casion any discharge of significance. When abundant ;tsci were in condition for discharge, lu'avx' t'ji'ction connnenced at the begiiming of rain periods and continued with continuous moisture as long as the Ei'ii)kmi()I.O(;y and Control ok Aim'LE Scab 47 supply of ripe asci lasU-d. Tlu' data in 'l"al)k- I indicate tliat on May 13 the heavy discharge lasted approximately five hours and on May 23 about ei^lit or ten hours. These results accord well with those of Frey and Kcitt (1925). In lal)oratory e.xperinients the present writers have observed ascospore discharge to occur freely at temperatures ranging from Y2 to i2° C. No trials were made at temperatures out- side of this range. The Occurrence of Primary Infection Under Wisconsin conditions, where ascospores constitute tlie only important jjiiniary in(jculuni known, the occurrence and amount ol primary infection are governed largely by (1) the abundance and timeliness of maturity of ascospores, (2) conditions of moisture and temperature in relation to spore dissemination and infection at critical periods in host development, and (3) the rapidity with which the host passes through its more critical periods for infection. It has already been shown that, under Wisconsin conditions, mature ascospores may conmioniy l)e found by the time the first susceptible host tissue is exposed in the sj^-ing, and that primary infection may occur at that time at ordinary orchard temperatures during rain periods of sufficient length (PI. III). The relations of host development to priiuary infection and the extent to which the latter occurs merit further consideration. The order in which tlie susceptible host parts are exposed bears an important relation to early infection. It is a matter of conunon knowledge that, in the case of the apple, the "fruit" or "cluster" buds, which contain the "blossom" buds, are the first to open, and that those borne on fruit spurs are generally more advanced in unfolding than those on ternu'nal shoots (PI. I, C, D). Tiie apical portions of sepals and of leaves are the first susceptible parts exposed as the "fruit" buds open. The tips of sepals are found to be exposed to in- fection at the early stage of bud unfolding shown in Plate I, A, B. At Sturgeon Bay in 1924, abundant sepal infection occurred on l)uds at the stages shown in Plate III, A, B. The "leaf" 1)uds are ordinarily some days behind the "fruit" buds in opening. In the early stages of leaf developiufnt the under (dorsal) surface is more exposed to wetting and, consequently, to infection, whereas, in the intermediate and later stages, the upper surface is m,ore exposed. With the exception of the sepals, the "blossom" buds are very well protected from infection througli a considerable period after the "fruit" buds begin to open, partly l)y the adjacent leaves and partly by hairs (Pis. I, II, A. B). It has been pointed out (p. 35) that susceptible host parts pass through a period of maximal susceptibility when quite young into stages of increased resistance. The rapidity with which the more susceptible stages are passed may greatly affect the amount of primary infection. At Madison in 1922, in a season of comparatively rapid growth, the stages of bud development of Wealthy shown in Plates 48 Wisconsin Research Bulletin 7Z C ^•■■y I I'l.ATI. L Sr.\(il.S IN IM OLDlNd Ol "I lU II • lUDS OF WICALIIl V AI'PI.K A. — Very early f^rccii lip slagc, nIionv iiig apical parts of sepals exposed at a (X 7). B. — liiul shown in A, viewed Ironi above. ('., I). — Later green tip stage, Madison, Wis., ,\pr. 28. Iit22 I .\ it'lOi: C, borne on terminal shoots; ]), b.:rne on liiiit spurs. (See PI. II I'or later stages.) Epidemiology and Control of Apple Scab 49 plate II. -stages in unfolding of "FRUIT" buds OF APPLE, MADISON WIS., 1922 A. — Early closed cluster, Apr. 30: a. borne on fruit spurs: b. borne on terminal shoots. B. — Middle closed cluster. May 2. C— Late closed cluster. May 4. D.— Open cluster. May 7. (All flgures Wealthy X f,/ 10. Sec. PI. I for earlier stage.) 50 Wisconsin Research Bulletin I'h PLATIC ML THI-: l)i:Vi:LOPMKNT Ol' APPLE "FRUIT" BIDS IN" RELATION TO scAii ini"i:<;tk)n and control, sackett orchard, SrCRCEON RAY, WIS., 1!>2I A.— Wealthy, May 10. R.— DiuUcy. May 10. C.— Wealthy, May l.'.. D. Dudley, May 15. Severe sepal infection oieurred on both varieties prior to May 10. Epidkmiolo(;y and Control of Aimm.k Scah 51 I, r, D ,111(1 11 vveTf i)ass(.-(l in eleven days. In seasons of retarded growth a like development has been observed to re(|uire as much as four weeks. The amount of priniar\- infection varies greatly with the factors just considered. In seasons of rapid host development and unfavoralde conditicns for infection it may be sharply restricted (Figs. 2, 3, 5). In seasons of slow host development and suitable cor.ditions for in- fection, particularly in the pre-blossom period, i)rimary infection may i)e very al)uiidaiit (Figs. 1, 4. 6). .Kt Sturgeon Ba\- in 1924, a year of severe primary infection, counts made on imspra\ed plots before se- condary infection developed revealed sepal infection on 69 per cent of Wealthy blossoms and 87 per cent of Dudley. Not infre(|ueiitly, several sejjals on a single blossom were infected. Further data on the abundance of sepal infection appear in Tables XV and X\'II. In considering the amount of primary infection on leaves shown in Figures 2-6, it should be remembered that the averages given there include the leaves which had become resistant as well as those which were susceptible. In the case of Lubsk's Queen, which was in a comparatively low state of vegetative vigor, the sparseness of fol- iage infection (Figs. 2-5) is largely attributable to the fact that leaf production was completed and resistance developed so early that oppor- tunities for primary infection of foliage were minimized. To develop the greatest number of infections the individual leaf must be inocu- lated after it has expanded the maximal surface consistent with reten- tion of the necessary degree of susceptibility. Only one or two leaves of a shoot ordinarily approximate this stage at a given time. The present writers have often found from 2S to 50 primary infections on individual leaves which were exposed to severe natural infection at about this stage. The Special Significance of Early Infection of Sepals The development of scab on sepals or other parts of "blossom" buds or blossoms has often been reported (Fairchild, 1894, p. 43; Beach et al.. 1899, p. 386; Scott and Quaintance. 1907, p. 21; Jackson, 1913. p. 238; Jackson and Winston, 1915, p. 13-14; Darnell-Smith and Mc- Kinnon, 1915, p. 26). The relations of earl_\' sepal infection to epi- demiology and control appear, however, to merit further consideration. The present studies of primary infection have shown that in years of severe epidemics abundant sepal infection occurred at a surprisingly early stage of bud development (PI. Ill, A. B). In view of the fact that conidia of V. inacqualis are disseminated chiefly by meteoric water (p. 56), it was at once apparent that early sepal infection establishes the fungus in ideal position for secondary infection of the affected fruit during its period of greatest susceptibility. It seemed probable, therefore, that lesions on sepals might play a role in scab infection of apple fruit similar to that of twig 52 Wisconsin Research Bulletin 7Z lesions in fruit infection of the peach l)y CladosporiuDi carpo^Jiiluiii ( Keitt. 1917, J). 36-40). Consequently, the following studies were made. In 1923, alter abundant sepal infection had appeared but before secondary fruit infection was evident, lOU unsprayed Dudley fruits which showed sepal infection and 22 which did not were tagged for subsequent study. Records of the amount of infection were made on July 3, after the appearance of the first wave of fruit infection. The results, which appear in Table VIII. show that scab development on the sepal-infected fruit much exceeded that on the sepal-clean fruit, both in the percentage of fruit infected and the number of lesions per infected fruit. Furthermore, the percentage of drop was materially greater in the case of the sepal-infected fruits. The data shown in Tai)le IX were taken from the fruits (5t) of each variety) which were used for seasonal development studies (p. 3-4). The records taken on June 24 show a strikingly greater development of scab on sepal-infected than on sepal-clean fruits. On July 30. after a further period of disease development, this difference was very little changed in the case of Lubsk's Queen, on which there was practi- cally no leaf infection during the season (Fig. 5). On Dudley, how- ever, upon which both leaf and fruit infection were very abundant, this difference was materially lessened, because of the abundant general inoculum from the leaves. Nevertheless, even under these conditions, the average number of lesions on the sepal-infected fruits exceeded that for the sepal-clean fruits by more than 75 per cent. Furthermore, though not shown by the tabulated data, the lesions on sepal-infected fruits were generally larger and more injurious than those on sepal- clean fruits, doubtless because of an earlier average date of infection. In the material studied, most of the badly scabbed, misshapen fruit developed from blossoms which sustained early sci)al infection or else' chanced very early to develoi) suitably situated lesions. The fact that a considerable percentage of sepal-infected fruits on the unsprayed trees did not develop secondary infection appears to be due primarily TAni.E VIIL — Tin: Hi;i.,\ti()N ok Skp.\i. Ini-kction to thi; Dkviu.opmknt of Sr,.\B on Young DuDi.KY Apples, Sturgeon B.^y, Wis., 1923 Classes of fruits observed Results from fruits on U ee, July 3 Fruits off, July 3 Infe cted> Clean' Per cent Per cent kve. No. lesions 100 fruits which showed SEPAL INFECTION 85 45 5.9 2.9 15 55 24 22 fruits which showed no infection on June 10 9 'Exclusive of sepal infection. 'NTo fruit infection other than tnat on the sepals was evident on Juno 10. Epidemiology and Control of Apple Scab 53 Taiii.e L\. — Till-; liKi.ATKiN ')i- Siu'.vr. Im-iiction to tin; I)i;\ i;i,<)1'.mi;.\t oi- Scab on Apple FhVIT, STl'HCiEON Bay, Wis., 1923 Variety iiiul class of fruit' Scabbed- fruits, June 24 Scabbed^ fruits, July 30 Per cent Ave. No. lesions Per pent Ave. No. lesions Lubsk's (jucon 8EPAL-I.\FKCTED on June 3 (48%) CLEAN on June 3 (.52'f) 54 L5.4 54 38 70 43 8.2 1 7 3 2.8 6.1 1.4 59 29 100 100 8.4 1.6 Dudley SEPAL-INFECTED on June 3 (56%) CLEAN on June 3 (44 'j) 34 19 Wealthy SEPAL-INFECTED on June 3 (56%) CLEAN on June 3 (44';)... 'The numbers in parentheses refer to the percentage of fruits of each variety which showed sepal infection or were free from scab when observed on June 3 (see text). *E.\clusive of sepal infection. to unfavorable exposure of the infected sepaLs and the adjacent .sur- faces of the fruits to wetting l:)y meteoric water. Tliis situation ap- pears to be closely similar to that described by the senior writer (1917, p. 36-40, Fig. 1 of PI. Ill) for peach scab. Typical stages of scab development following sepal infection appear in Plates IV and V. The relation of sepal infection to secondary infection is shown very emphatically in Plate IV, B, in which a sepal-infected fruit shows heavy secondary infection, while a sepal-clean fruit of the same cluster (a) remains free from infection. Similar studies in 1924, another season in which sepal infection was abundant, gave results which confirmed those of the previous year. It is evident, therefore, that early sepal infection is of much im- portance in relation to epidemiology and control through establishing the fungus in a remarkably favorable situation for secondary infection of the affected fruit during the period of its greatest susceptibility and at a stage when thorough protection by fungicidal applications is difficult. Production and Dissemination of Conidia Under field conditions the production of conidia ordinarily begins before scab lesions become macroscopically visible, and on the indi- vidual lesion may continue throughout the season. Extensive ob- servations have shown that sporulation occurs freely on scab lesions under ordinar],' field or greenhouse conditions without rain or arti- ficial watering. The following record illustrates the rapidity and consistency of sporulation. On a potted Wealthy plant in the 20-25° C. greenhouse, mi which humidity was not controlled, a leaf bearing abundant lesions which had developed as the result of a single inoculation was atomized thoroughly with water by a standardized procedure. The water was recovered and examined for conidia, which were found in abundance. 54 Wisconsin Research Bulletin 7Z PLATE IV. SIvl'AL IM i:( IION IN 1?1:L.\TI0N TO SKCONDAHY lM"i:(.riON A.- Sovcic sepal iiirci'lioii on VirKiiiiii, •Iiiiic (i, 1!»2L B. S('i)al iiirci'tion followed l)y :il)iiii(hinl secondary iiilicl ion, excej)! a, which is sepal-cleaTi and scab-lree Ihont^h in contact with a hadly scabbed Iriiit. Snniiner I'ear, .Inne 1. ]!)21. ('.. — Secondai'y infection followins sepal infection, Snininer I'ear, Jinie 1.'), 1!)2L All material from Madison, Wis. Epidemiology and Control of Apple Scab :)D 56 Wisconsin Research Bulletin 7Z The leaf was then thoroughly washed with the aid of a camel's hair brush, after which examination of wash water showed only a trace of conidia. Repetitions of this test were made on the same leaf on the third, fourth, ninth, and thirteenth days following. In each atomizing test conidia were found in essentially undiminished numbers, while only traces of conidia were found after each thorough wash- ing. Field observations showed that an abundant source of secondary inoculum was never lacking on unsprayed trees after infection was once well established. The records of seasonal development of the disease (Figs. 2-6) may, therefore, be regarded as a rough index of production of conidia. Frey and Keitt (1925, p. 537) showed that the conidia of V. macquaWs are very resistant to separation from their conidophores when dry, but promptly become detached in the presence of water. They state * * + "These results, in conjunction with those from the air filtration experiments, indicate that no important dissemination of conidia is to be expected in the absence of water, though undoubtedly some spores are dislodged by wind- whipping of leaves, fruit, or branches, by contact with wind-blown particles, and in other minor ways. It appears, therefore, that the important agency for dissemination of these conidia is meteoric water moving under the in- fluence of wind and gravitation." The present writers, using a slightly diiiferent techni(iue, have re- peated and somewhat extended the experiments of Frey and Keitt, with confirmatory results. Currents of air from an aspirator were driven against the surfaces of abundantly sporulating scab lesions on leaves and fruits in the laboratory. Glycerined slides were placed in a favorable position to catch samples of any conidia that might be dislodged. The catches of spores were ml or sparse, except in cases of excessive air velocity (estimated to be higher than velocities commonly attained in the orchards), when they were somewhat larger. Similar tests were conducted (a) in a saturated atmosphere in the moist chamber shown in Figure 8, using the same specimens after they had been held in a moist chamber for three hours, and (b) in the laI)orat()ry, using the same specimens after they had been atomized witli water sufficiently to detach the conidia from their conidiophdres, l)ut not to scatter them widely or wash them off. and dried rapidly. Microscopic examination showed myriads of conidia in the droplets of water which stood on the lesions after atomizing. The results in the saturated atmosphere and after washing were not essentially different from those obtained from air-dry lesions in the laboratory. In the filtration studies of the frequencies of ascospores in orchard air in 1924 (Tai)le 1), watch was kept for conidia. They were found only four times, appearing in comparatively small numbers and only during rainy or windy periods. EriDKMIOLOGY AND CONTROL OF Al'PLE SCAB 57 The Occurrence of Secondary Infection The occurrence of secondary infection depends chiefly upon the following factors, each of which has been discussed in other connec- tions : (1) a sufficient inoculum, (2) sufficient moisture for spore dissemination and infection, (3) suital)le temperatures, and (4) the presence of unprotected susceptible host parts. Records of secondary infection appear in Figures 2-6 and Tables VIII, IX, XV, XVII. Critical Periods for the Development and Control of Epidemics It is generally accepted tiiat critical periods for scab infection and control occur in spring and fall. The foregoing studies seem to make possible a somewhat clearer understanding and definition to these per- iods. Under Wisconsin conditions the most critical period for the develop- ment of epidemics extends from the time the apical parts of the sepals are first exposed in tlie opening "fruit" buds, (PI. I, A, B) to an indefinite time some two to four weeks after petal-fall. The early part of this period is the more critical in relation to control for the following reasons : 1. Early infections provide an early secondary inoculum, which in the case of sepal infection is situated in a peculiarly favorable posi- tion for severe infection of fruit. An early and abundant secondary inoculum is of special importance in the case of apple scab, because (a) the fungus has a comparatively long incubation period, (b) the host passes rather rapidly through its period of maximal susceptibility, and (c) environmental conditions are more likely to be favorable for early than later secondary infection. 2. The rapid expansion of host parts in the pre-blossom period (Pis. I, II) makes it exceedingly difficult and expensive to keep them adecjuately covered by a suitable fungicide. In cool summer climates, as at Sturgeon Bay, Wisconsin, if there has been sufficient primary infection to lead to the development of an abundant secondary inoculum, important infection periods may occur at any time prior to harvest. Hot dry weather, however, sharply checks scab development. Consequently, in warmer climates control during the summer is comparatively easw A second critical period for seal) development occurs in the fall, when cooler weather and sufficient moisture may lead to important late infection of fruit and to the more abundant establishment of the fungus on the foliage, particularly on the younger leaves of late terminal growth and on the lower (dorsal) surfaces. If primary infection has been adequately prevented, however, the disease is easilv controlled in its later stages. 58 Wisconsin Research Bulletin 73 SPRAYING AND DUSTING EXPERIMENTS The spraying and (lusting work, which was conducted in four conin;ercial orchards at Sturgeon Bay during the six seasons, 1919- 1924, inchided 249 plot experiments on the following prohlems : 1. The cfiniparative ctTectiveness and desirahility of li(|uid lime- sulphur, dry lime-sulphur and Bordeaux mixture. 2. The desirability of early applications of fungicides. 3. The effects of variations in the number and timing of applica- tions. 4. The effectiveness and desirability of a mixed program of lime- suli)hur and Bordeaux mixture in comparison with programs of lime- sulphur and Bordeaux, respectively. 5. The effects of certain variatidus in the lime-sulphur program in relation to fruit injury. 6. The comparative effectiveness of spraying with rod and gun. 7. The effectiveness of adding certain spreaders to sprays. 8. The effectiveness of certain dust programs. In accumulating such a large amomit of data, special efforts were made to systematize methods, with the aim of enhancing the com- parative vahu' of the experimi-jnts and facilitating the presentation of results. The tdlbiwing account of methods is, therefore, applicable to all the spraxing and dusting work, except as otherwise noted. Methods Choice of Plots The experimental jilots were located with the aim of including: (1) varietit-s loeall\- most subject to scab ;ind those of leading com- mercial impoiMance; (2) trees in good bearing and as uniform as feasible in size, condition, and surroundings; and (.1) accessibility from the field laboratory. Care was taken to make plots of such Hze and location as I0 nn'nimizc any possible error occasioned by drifting of spray or dust. A brief description is given of the plots used in each experiment. Spray Materials Lime-sulphur (liquid). ,\ connni'rcial product obtained each year from the same companw .\nalyses have shown onlv slight variations fnnn year to \ear. The specific gravity ha^ varied little from 1.295 (33" B.). Dry lime-sulphur. .\ proprietary product donated each year by the Sherwin-Williams (omjiany, Cleveland, Ohio. Copper sulphate. Technical cojijier sulphate crystals. I'J'IDK.M lOt.OCY AND CONTROL OF ApPLE SCAB 59 Lime. In 191U and l';2(), local coiniULTciai stone linie. In 1921, 1922, and 1923 hydratcd liiiic. In 1924 a IiIkIi grade stone lime which con- tained 98 per cent of calcium oxide. Arsenate of lead. The commercial i)owdered acid arsenate of lead. Yearly analyses showed a satisfactory uniformity of product, the arsenic oxide content varying but slightly from 32.5 per cent by weight, and tile arsenic in water solul)le form,s (expressed as metallic arsenic) never exceeding 0.5 per cent. Glue. .\ very high grade finely ground glue obtained each year from the same coni])an\'. Gelatin, (iold label "Wli Xo. 1866" obtainefl from the .\rthur H. Thomas Company, I'liiladilphia, Pa. Casein-lime. "Kayso," a proprietary product donated by The Cali- fornia Central Creameries, Inc., San Francisco, California. Sulphur-arsenate dust. .\ connnercial preparation (909^ by weight of sulphur and lU'/c powdered lead arsenate) donated by The Niagara Sprayer Company, Middleport, N. Y. Sulphur-arsenate dust A. This material was pre])ared l)y thoroughly mixing 10 jiarts by weight of ]iowdered arsenate of lead with 90 parts of flowers of sulphur of which 99 per cent was of a fineness to pass through a 200 mesh sieve. This sulphur was 65 per cent soluble in carbon bisulphide. Sulphur-dry lime-sulphur-arsenate dust. A mixture coniposed of 75 per cent by weight of finely ground dusting sulphur (flour), 15 per cent of finely ground dry lime-sulphur, and 10 per cent of pow- dered arsenate of lead (Giddings, 1921). Copper-lime-arsenate dust. .\ mixture composed of 10 i)er cent I)y weight of finely ground anhydrous cupric sulphate, 80 per cent of hydrated lime, and 10 per cent of powdered arsenate of lead. In 1924, the following formula was substituted: 12 per cent of mono- hydrated cupric sulphate, 10 per cent of powdered arsenate of lead, and 78 per cent of hydrated lime. Preparation of Sprays Bordeaux mixture. Pound-to-gallon "stock solutions" of cupric sulphate and lime were prepared. About three-fourths of the desired volume of water was run into the spray tank. With the agitator and the tank filler running, the necessary amount of stock solution of copper sulphate was added, followed I)\- the recpiisite amount of the stock preparation of lime, which was diluted with water as it passed through the strainer. The arsenate of lead was added in water sus- pension, and water was run in to make up the necessary volume. .All materials were passed through a strainer. In 1919, 1920, and 1924 stone lime was used: in 1921, 1922. and 1923. hydrated lime. Xo correction was made for the decreased amount of calcium oxide in 60 Wisconsin Research Bulletin IZ the hydratcd lime, as compared with the stone lime. Tests made with the materials used each season showed that a large, though varying, excess of calcium hydroxide occurred in all the Bordeaux mixtures used. The 1 to 1 ratio by weight of cui)ric suljihate to lime was used unless otherwise stated. Lime-sulphur and arsenate of lead. The necessary amount of liquid linie-sulpinir concentrate was added to about three-fourths of the recjuired volume of water in the spray tank. The arsenate of lead was then added, as in the case of Bordeaux mixture, and water run in to make up the necessary volume. The agitator was run continuously during the mixing process. Dry lime-sulphur and arsenate of lead. The necessary amount of dry lime-sulphur was vigorously stirred in a bucket of water and poured through a strainer into the partly filled spray tank, as in the case of liquid lime-sulphur. Arsenate of lead and water to make the required volume were added as in making Bordeaux mixture. The Addition of Spreaders and Adhesives to Sprays Gelatin. The gelatin was dissolved in a suitable volume of water by heating and stirring. This solution was added to the agitated spray mixture just before the latter was made up to volume. Glue, (jlue was added in same way as gelatin. Casein-lime. The necessary amount of "Kayso" was placed in susi)ension in water and added to the spray as were gelatin and glue. Technique of Application of Sprays All sprays were applied with high grade power sprayers of 200- gallon tank capacity. When rods were used, each was equipped with two angled disc nozzles, with nu'dimn apertures, and pressures of from 225 to ITS pounds i)er s(|uare inch were maintained. The trees were thoroiigliiy and evenly covered. Care was taken to avoid "drenching." One man sprayed from the top of tiie rig and another from the ground, covering one row at a time. Wiien the spray gun was used, pressures of 250 to 300 pounds were niaiiilained, and when feasible the sprayer was operated along rows running jiarallel to the wind direction. i'"acli row was treated from two sides by a single gun operated from tlie lop of the rig. In view of the fact that most of the apple trees in Door t'onnt\- ari' conijiaratively young, the spray gun tests were conducted on twelve-year-old trees. Technique of Application of Dusts .\11 (lust applications were made with a standard jiower duster equipi)e(l willi a live horse-power engine. The duster was operated along rows as nearly i)arallel to the wind direction as feasible, and 1'J'i1)i;m loi.odv AND Control of Apple Scab 61 each row was treated from two sick-s. (are was taken to make tliese treatments as thoronyli as possible witliout nndue waste of materials. Treatments Unless otherwise stated, all spray treatments were made under con- ditions which permitted satisfactory ai)plication and thoroupfh drying before rain fell. Tiie conditions which attended eacli ai)plication of dust are noted in tlie appropriate connections. Seasonal Development and Meteorological Records Records relating to the seasonal development of host, parasite, and disease and meteorological data are reported above (p. 3-6). A special telegraphic weather forecasting service was furnished through the coopera- tion of the U. S. Weather Bureau.' This proved to be of much value in facilitating advantageous timing of applications. Results on Fruit at Harvest Count trees were chosen at random as regarded the condition of fruit. They were viewed from a distance and selected as representa- tive of the plots in regard to the quantity of fruit borne. They were located as near the middle of the plots as was consistent with other requirements. Results were ordinarily taken from four trees in each plot. In certain minor experiments and in cases where the yields were unusually large, the count work was limited to the crop from two or three trees. The number of count trees used in each plot is recorded in the appropriate Table. .At harvest time, every fruit from each count tree, including fruits wdiich lay upon the ground, was critically examined and its condition recorded according to the classification given below (see PI. VI, B). Scal)bed fruit was divided into three classes, slight, connnercial, and bad. It was aimed to class as slightly scabbed those fruits of which the market value was not seriously impaired except for their being thrown out of grade A (Wisconsin standard). The commercial scab class was designed to include fruits on which scab injury was greater than "slight," but not enough to preclude sale of fruit. The bad scab class was planned to contain tliose fruits which were so injured by scab that they were rendered nearly or wholly unsalable. While, of necessity, the judgment of the sorter played some part in the classification of fruit, the following standards were adhered to as closelv as feasible. Hiiiitol'iil lukncw Icdf-tUfnts ;np nsKie to Prof. H. J. Cox and his associates of the Chicago Station of the U. S. Weather Bureau for their kind cooperation in fuiuishing lliis v:ihinl)!e serv'ce. 62 Wisconsin Research Bulletin 71) I'l.ATi: \L ii.i.rsriiA'noNs oi- appahati s am> mi/iuods A.- A|)|);ii;iliis ii'-cd |ii|- dclci mi niiit; I ii'(|iiciuics (if -ijorcs in orcluiid ;iil- < sec p. i:! 111. l'>. Tnk'iij; r(sulls in liuit ;;| liinvcsl. Epidemiology and Controf. of Aitle Scab 63 Slight scab. Maxiiiniui of seal) \)vv fruit: oiu- s\)(>t 5 iniu. in (lianicter and sliulitl.N' cracked, two sixits 5 nnii. without cracking, three to four spots 2-4 nun. without crackinj^, or not more than five spots, 1-2 nun\. Commercial scab. Maxinnun scab i)er fruit: one spot 15 mm. in diameter witliout erackini^, one spot 1(1 nnn. with slight crackiu)^. three to five spots 5 nnn. without cracking, or 10-25 sjjots ufjt larjijer than 2 nnn. Bad scab. Any seal) injur\- more serious than ■"connuercial," as just descrihid. Riisseted fruit was classified as slight, connuercial, or had. The class of slight russet was designed to include those cases in which the grade of the fruit was not afifected and its market value was not reduced except as might result from its inferior finish. Commercial russet was planned to include fruits sufficiently injin"ed to throw them out of grade A ( W'isccjusin standard) and distinctly' reduce their market value. The had russet class was designed to include all more seriousl\ russtted fruits, man\- of which were cracked and rendered unsalable. The jitdgment of the sorter was based as nearly as feas- ible on the following standards: Slight russet. iMa.xinimn injur\' per fruit : light net russet covering 20 ])er cent of the surface of the fruit or solid russet of 1 stpiare cm. without cracking. Commercial russet. Maximum injurs': heavy net russet without cracking, or solid russet covering 25 per cent of the surface without cracking. Bad russet. .Ml russet accompanied by cracking and all solid russet covering more than 25 pvr cent of the surface. The time and labor recpiired for taking residts and the injury to fruit by handling were minimized by taking all data on each fruit from a single examination and recording the results upon a series of hand tall;' registers (PI. \'I, B). The data from each count tree \\'ere then computed on a percentage basis for each classification outlined above. The figures shown in the Tables X-XI\' and X\'I are the n;eans of these tree percentages. The standard deviation of the percentage of scab was determined i)y the fornuila, .S. D.=-vl?^, in which ^ represents summation ; d, the indiviilual tree difference from the rriean ; and n, the number of trees. Experiments in 1919 Condition of Plots Learned orchard. Well developed l.ubsk's Queen trees aboiU .iO years old. Orcharrl had been in sod for several years, and was not cultivated in 1919. Trees in rather low state of vegeta'.ive vigor. 64 Wisconsin Research Bulletin 7Z Orchard on a western slope witli excellent air and surface drainage. Soil sandy to light silt loam. Burned over about May 1. Many patches escaped burning, only about one-fourth of old leaves being destroyed. Scab had occurred in great severity in this orchard in previous years, and an abundant supply of ascospores escaped the fire. Lawrence orchard. Well-grown, vigorous Fameuse and Mcintosh trees, most of which were about nine years old. Orchard well ele- vated, with a slight slope to the east. Good air and surface drainage. Soil light to heavy silt loam. Orchard disked in early spring and cultivated until July 1, after which weeds were allowed to grow until they were mowed in early September. Seasonal Conditions Seasonal development and meteorological data appear in Figure 1. No ascospore discharge from the leaves selected for the seasonal development experiment was observed until May 15, when the "blos- som" buds were in the closed cluster stage. That ascospores were matured earlier, however, was shown by the fact that a sparse dis- charge was obtained on May 3 from freshly collected leaves which were moistened in the laboratory. Furthermore, sparse leaf infection was observed on May 24. This allows an incubation period of 18 days from the rain of May 6. which accords well with early-season incubation inriods of other years. The more critical periods for scab control were May 19 to 21 and June 5 to 6. The rains recorded on May 15 and 17 were of short duration and apparently led to little if any infection. The season of 1919 was one of rather severe scab infection, the iiercentage of scabbed fruit on untreated plots varying from 69 to 97, of which a large jiart was severely infected. Treatments and Results A sununar\- of treatments and of the residts taken at harvest appears in Table X. Li(|uid and dr\- lime-sulpliur eoiitn illi'd scab as etficieiitly as did Bordeaux mixture, and occasioned much less russeting (jdots 77, 71, 85, 87, 81, 94, 96, 90). There was little difference between the lime- sulphurs either in scab control or russeting. The nn'xed program of Bordeaux mixttn-e and li(|nid lime-sulphur (plots 8,v85, 87, 92-94, 96) gave about tlu' same scab control as did the lime-sulphurs, but nujch more russeting. However, the mixed i)rogram occasioned nuich less russet than did the full jirogram of ]^>ordeaux mixture. In no case was satisfactory scab control obtained on i)lots where a])plication 0 was omittid, while very salisfactor\- control resulted on plots where it was apidied in addition to the four-treatment program (plots 71, 72, 77, 78, SI. 82, 85-88, 90. 91. ^-1<0000 ■!«'0-HOC^'OT).« i-hiOxcO"Ccocon -H c-i ?: M -r o c>i iM o > iM o — i oi -< M o cm t^O—iWO—c-iOOC^Ii t~c-. — 'i.'^-^cco'oo otoooeocM-^r^Tfc^) t>.C>lc«5TfiM-<}'CO OJW C! CO Ol C^ CO CO C) ) X CO X "*" C'l 05 c^i 1.0 — X 1^ CO CO CO c. CO c^j c; -ti lo -H CO CO CO CM 1.0 CO o c; i^ -* o CO — C-. Tf -1 1^ lO a; T)« coa;xx-ix-fco CM-rx^oco-^TPt-. ^ -f Tjl .^ T^ Tf Tfl 'T ^ -^ Tt^ rfl It 1^ Tf -^ T}1 .r^l "^ *<** •-( l^ 1^ Tfl Ttl Ttl Tf CO 't o^ C3 c ^^-'^ COCO ca ^scoco^ CM •£ J; a; CM CO CM CM CM .' -.".t^. Oi — C^ — t . -c -h\^ O r-c 0) — Tt< - -c O-.^ TJ- rt ^^ ^-^ ,— T-J" C r ci3 ^-^ ^-^ r^ ^ ^ -. 7 1 . Si: - r - r^XxSi; - r r r°X COTf S cm'co" > ■ ■ • • ^ . - . o a • ■ • ■ - - , I 5 ft " ' o o Oi-iOlP0-*i0t0t^X05 O^CMCO'^i.ocOt^X C:0-H01SO'*iCcOt^ t~ t^ t~ r- 1>. t>. t^ t^ t--. t^ xxooxxxxxx xoososooiosoio t: ;;5 .. bO n, &s ■73 3 o ►2 ^ X - °x 13 CD 3 ►J-o c c 00 to O O 03 t-> A ft ii ^co E o 1 s .^ •- ;=: CM c &«« II ^Q 66 Wisconsin Research Bulletin 7Z treatnuiit to the four-spray program resulted in an increase of scab- free fruit amounting to from 11 to 25 per cent of the crop. The . reason for these results is readily apparent ujjon referring to Figure 1, where it is shown that the most important ascospore discharges of the season occurred under favorable conditions for infection well in advance of the open cluster ("pink"j spray. Results from variations in the tinije and number of applications are valuable primarily as the}- are interpreted in relation to seasonal conditions. The decreased scab control when application 1 of Bor- deaux was made three days early on Lubsk's Queen probably resulted chiefly fmni imperfect covering of the expanding "blossom" buds, which had not yet separated in the clusters. It is not surprising that an application three days later than usual gave increased efificiency in scab control, as no rain occurred between May 21 and May 30, and the later application had the advantage of covering a greater area of the unfolding parts (plots 75, 72, 74). Considerable increase in scab infection resulted when applications 2 and 3 were delayed seven days each (plots 76, 12, 79, 78). This appears to have been due primarily to lack of adequate protection during the important infection period of June 5-6. The increase in scab infection attendant upon the omission of treatment 2 is explained in the same way (plots li, 72). Experiments in 1920 Condition of Plots Learned orchard. .Same Lul)sk's Oueen trees as in 1919. Also, 48 nine-year-old Wealtiiy trees (plots 113-llC>l in same orchard. Wealthy block fairly well cultivated. Lubsk's Oueen block, along with re- mainder of orchard, lightly disked in 1920, with the aim of bringing it graduajlv into clean culture. Lawrence orchard. Same trees as in l'H9, and same cultural methods. Increased attention, however, given to clean cultivation in advance of first ascospore discharge period. Seasonal Conditions Seasonal develojiment and meteorological data appear in Figure 2. The first discharge of ascospores from tlu' leaves used in the sea- sonal development series was recorded on M a.\' 2(1, wlieii the "blos- som" buds were in tlu' larly closed cluster stage, 'i'lie earliest ob- servation at Sturgeon i!ay was made on May 18, when abundant ascospores were found to ])e mature. Tlie first iiiteetion ot the sea- son, which was rather sjjarse, ajipears to Iiave occurred on May 17 and 18. 'i"he more critical periods for seal) development occurred during the rains of June 7 to 11 and 15 to 16 and July 6 to 7. Less important infection periods occurred from May 20 to 2?>, July 17 to Epidemiology and Control of Apple Scab 67 18 and 22 to 23, and August 11 to 13. Prc--l)lo.-,soin infection was sharply liniitid by scarcity of rainfall of sufficient duration to cause infection. The season was one of moderately severe scab infection, the percentage of scabbed fruit on unsprayed plots varying from 72 to 91. The occurrence of abundant fruit infection on Lubsk's Queen in the absence of a significant amount of leaf infection is noteworthy (see p. 51). Treatments and Results A summary of treatments and of the re>ults at harvest appears in Table XI. Under the conditions of this season hfjuid lime-sulphur and Bordeaux showed little difference in effectiveness of scab control (plots 99, 101, 118, 122, 127, 131). Serious fruit russeting, however, made Bordeau.x unsatisfactory commercially. The mixed program of Bordeaux and lime-sulphur controlled scab as well as did the full programs of Bordeau.x or lime-sulphur. It occasioned less russeting than did Bordeaux, but more than lime-sulphur (plots 120, 118, 122, 129, 127, 131). In full programs which included treatment 0, dry lime- sulphur gave essentially as good results as did liquid lime-sulphur (plots 112, 111, 101, 124. 122, 133. 131). Where application 0 was omitted, however, tlie control of scalj l)y dry lime-sulphur was slightly less efficient than by li(|nid lime-sulphur or Bordeaux (plots 125, 123, 119, 134, 132, 128). On Mclntosli and Fameuse, the disease was satisfactorily con- trolled without the addition of treatment 0, except in the case of dry lime-sulphur noted above. On Lubsk's Queen, however, the con- trol was not fully satisfactory without this early treatment (plots 102, 101, 100, 99). The relatively small value of the pre-pink treatment this year is due to the dry period wliich followed its application. Due to the dry weather, little dift'erence in results attended the changes in timing f)f treatments 0. 1. and 2 (plots 106-108, 101). The omission of treatments 1 and 2, respectively, let to a relatively small increase in the percentage of fruit scabbed (plots 105, 104, 101). It appears that either of these applications furnished sufficient pro- tection to prevent severe infection during the period of June 7-16. No increase in scab development atteiukd the omission of treatment 3 (plots 103, 101). There was no evidence that the addition of gelatin to lime-sulphur was beneficial (plots 109, 101, 110, 103). There was no significant difference in results from experiments in which the spra\- gun and rod were tested comparatively (plots 116, 114). 68 Wisconsin Research Bulletin 7Z ° 3 Hi o o 52; ■*coo5coooo5'MO'.o-Ht^t^^cceo 6^ d^iocooo^-i^'Mo-f-^o — CO X t^ M 0000 r^ c-i o o re 'O « >o uo -(■*OlXtOOCOC^rt OCCMOOOOOO t^ cjTt*ooc<30t^ooot^-Hr^(Z)-^< :cot^ O5t^oiooc3'coc-irt x^ot^O'OTjto'Oco "Oa>00'- 10 o: X' C5 -js CO 1^ ID o CO 05 -^ X .-cMcocOiOCOXO>tDX"C-*ait>-COO^ coxoto COC^iOlNCO'-<'*C-c* r-iMiCTjXtO'}< (MCO(NCOiOCO-*T)H(Nf-i,-i(NiOT}Xi-iT»XiOCO!CCOX>-iI^C0CO ■^ OlXiOtO-^r-iC^COCOCOi-iiMOlt^tO tOtOXOOSX'OCOCO OC^— "COXIC>JXtJ< tO'HO"^00^)COt~ t'J ^_J "J ^tJ l^ »— ' ^'J ^») UJ C. 'C"0.tOC^t^ iMOgcOOJ'MOl'NC^iMC^CO-f-f'MCO CO-f-^-f J rf ■^ -t 1^ -t -t -f ' -r -f -)• -f -t -f -<■ l5« CO cico ^ ^ ^ > >i >iCO T) T3 T3 M c^i (NCO to - ^Noc:- c^cocooj'Nci'^'-S .\ ri(M"co"cvi •«_; MM./ .J >> - - -"^S• ■Jim pa jJ] COtC Co'rJ-'coS :(^j„- „ . _ _ ,drt" 0000 00++'''^ X _ _.. - - rrrrrr. .Jj>.S'rr : S S-SS 'Ji^-j^'Xi'r.-mi-n , , .^ g cc cc cc ^ g . _ CO-* im'co" CO I icJco j-co'-^f.'J:'^ MX"-' ►-3J3 m QCi 2 'X CO-* Vh'x'x xa>o-^£)co-fic:Dr»xo:o •* ic to r^ X 01 o -^ c^ CO -f'O oooodoodoooo^--^^ ^^rn^ ^—i^cgc^c^cqc^c^i "3 m >. - ""So o c 3 c^^ 0 |J 01 IN 3 5 0! ffl M >> Q & d It fiCO ^ >iO m ■ri 0 S3 Q c J3 0 0 0 0) rf ~ ■^ c « M ii; 0 a is m 0 01 IN 0 ~ I-] -* «J M 3 ot a a P dob . s. m "0 3 -* 03 0. Oj T3 c« c C 3 = > c 3 0 T1 01 <5 a 0 -0 2 ^ i 0 'C a 0! ■§ "c. 1 a. cS u C3 = C3 03 T^ & >- n. 0 t^ crt ririi P9 a ,_, /J II •0 t CO II "o 1 0! 3 < a 0-* Epidemiology and Control of Apple Scab 69 Experiments in 1921 Condition of Plots Learned orchard. Same Lubsk's Queen and Wealthy trees as in 1920. For dusting work an additional block of Wealthy and Lubsk's Queen used. These trees were part of same planting as Lubsk's Queen in previous experiments, and were of like vigor and general condition. Young Wealthy orchard (plots 153-156) continued in clean culture. Old orchard disked three times, but considerable blocks of sod were left about bases of trees. Lawrence orchard. Same trees and cultural plan as in 1920. Special attention given to thorough cultivation prior to first period of ascospore discharge. Goff orchard. Vigorous, well-grown, twelve-year-old Wealthy trees. Orchard well elevated, nearly level, with good air drainage. Soil light clay loaiii to silt loam. Well cultivated seven or eight times a year with disk harrow, l)eginning l)efore buds opened. Seasonal Conditions Seasonal development and meteorological data appear in Figure 3. The first ascospore discharge from the seasonal development ex- perin^ent was recorded on May 13, when the "blossom" buds were in the open cluster stage. It should be noted, however, that no rain fell in the period between April 28 and May 13. The fact that sparse infection appeared on May 18 indicates that ascospores were mature in the rain period prior to and including April 28, when the initial infections evidently occurred. The dry weather throughout the spring and early summer so checked the development of the fungus that there was little scab, even on unsprayed trees. Treatments and Results A sunnnary of treatments and of the results taken at harvest appears in Table XIL There was not sufficient scab infection to constitute a satisfactory test of the various programs used. Conse- quently, the detailed data on conditions at the time of dust applica- tions and a detailed discussion of the results at harvest are unneces- sary. All the spray and dust materials used controlled the disease satis- factorily. As in previous years, however, Bordeaux mixture gave unsatisfactory results because of fruit russeting. An unu.^ual feature of the season's results was the occurrence on the lime-sulphur plots of a variable amount of fruit "burning" of a type which has not infrequently been reported by other investigators (Young and Walton, 1925). This injury was first observed on June 70 Wisconsin Research Bulletin 73 L° OOO'0T)iOiCr^C0-*'-<'-HCSOOOOO OCDTO'^OOOO OOOC-lOt^ OMCO c^o>!»c^'N c^ c^ojcoc^ro^rt t^ccojoD 01 n cq cj m ^ © c^ -h cvjc^ccm-h-^ ■^' — I O CO '-C -H rt 0'Mro>0>Ct^TOM<—it^M05'0 'CiO Xl t~ C'X •-^ '-' C^ •<}<(^iO30C^M t~MiO 6^ or-t^ooddooodooiMoo-HO ooooO'-'OO omc-iooo doo .-HXtD'Oc^t^comcot^cMt^ooocioco'N C-. '^M"— 't^c-icooj ci X '-c X c^ I--: c-fc^ '^■^ rt-i)S0rHrH.-4.-iCS01'-i(MOO'*'HrHO^ O^-hC^CJ-^^O ©""tCOOO ccoo tl; OOr~-*O-^'HTtiO0i--i«0Tt<-#rtC0'0O OiOrtC^TOrtC-lcO -fC^iOXC^)-* -^o-* H .-ioooooooooo-HOcodd(NO -f^c^irc'-'ooo oidoooo --lOO rjj r-4r-l'J<-Ht-(t-iO!MOM^'0tN^CDt^C0'X Coiotci0 ■Oi-iTi<.-i^m i-nooo ^^ COOC^OOOOi-iO— iO(M— itOMiCMCO O1NP0tJHt><(N^m XOOOOO r-O^ Ol C^ C^J C^ 01 eg C^) ?q OJ C^ 7-J -N C^l C^l •>» M C) M C^l -^ -1^ -t C-H(NO:"3CVlTjnu5-HC<:«ic^co . r~c<5>o«-< — i-Hccn Tt<0)-.+XX >- j««. - ^» x-J^" :''^^5g32xaJa!>-3CJ< cDf^X Oi O -< C^ M tf lO to r~ X t~t~f~ t^xxxxxxxxx 4) ,J 01 11 _.- s^^ Sd J<^ c o So.;." -2 5 '^" » ~ a ^. >, i iJ o -r' - ° c'S =3 OT S fc. ^7 ^ S £ o ^^ . « a ^ I t fc* C- CO ^^ -*- 5 to , -^ 3 X 3 0.2 ^x = O i ..= Md;1 XX d ," 3 t. t. a .. " „-i; ^x N 3 r O 3 >.< (d - .X CO a =s t.1 " S J= . ^ d-^s 3-37 oj - i4 i> 3 a .2c ga .ii o a^ |S 5 5 3 a ■3 *< CO 5 a-i S*"" ^ 3 S 1 i fi M "3 1 s ^ Iljf ""S ^ S ^ J °d < o". a II i2 "S S^E Tt< s^" t-, O a 3 -3 >> CCS a 0; 1— 1 T fe K — X CJ ^ cS . 72 Wisconsin Research Bulletin 7Z 18. It occurred mainly on fruit borne on the south side of trees and was confined to fruit parts which received direct sunhght during the hottest part of the day. The injury was greatest on fruit borne on the lower branches. It appeared as more or less circular, somewhat sunken, chocolate brown areas, often covering about one-half of the fruit. In cases of less severe injury, healing sometimes occurred and the dead tissue was later sloughed ofif. A study of Table XII will show that this injury was associated with the application of lime- sulphur in treatment 3 (plots 138, 140, 164, 167). The injury was slight on the experimental plots which received treatment 3 at the regular time (ten days after the petals fell), June 8-9. An increase m the amount of injury occurred on plots on which treatment 3 was delayed until June 15 (plots 175, 173, 166, 164). A similar relation of fruit burning to delay of treatment 3 was noted in various com- mercial orchards. The most severe injury noted occurred on a Dudley orchard which was sprayed on June 17. Seventy-five per cent of the fruit was badly injured. Reference to Figure 3 shows that. for the period of June 8-17, the maximum temperature was 93° F. on June 17, which was clear, hot, and calm. At no other time during this period was the temperature higher than 82°. June 18 was also a clear, hot day with a maximum temperature of 90°, followed by two days of cooler weather. Since no increase in the amount ■ injury was observed after June 20, it appears that the major portion of the injury occurred on June 17, with possibh' some increase on June 18. The results of plots 173 and 176 indicate that the addition of hydrated linu- to linie->uli>hur at tiie rate of five pounds to 50 gallons reduced the amount of l)urning. These data, iiowever, seem too limited to be conclusive. Experiments in 1922 Condition of Plots Learned orchard. Same trees as in 1921. Clean cultivation was practiced in the young Wealthy block f])lots 203-212), wliile the old orchard was disked three times early in tiie season. Goff orchard. .Same trees and cultural practice as in 1921. Sackett orchard. Well-grown, vigorous, Dudley, Wealtii_\-, and Duchess trees n>ostly twelve years old. Orchard on a gentle western slope, well elevated, with good air and surface drainage. Well culti- vated, beginning before first discharge of ascospores and continuing until the nn'ddle of Julw Seasonal Conditions Seasonal development and meteorological data aijpear in b'igure 4. The first ascospore discharge from the seasonal development ex- periment was recorded on May 12, when tlie "lilossom" buds were in the closed cluster stage. The first observations on ascospore maturity l^I'lDK.MIor.OGY AND CoNTROL OF ApPLE ScAB 73 at Sturgeon Bay were made on May 8. Abundant mature ascospores were found at this time. The first infection of the season was ob- served on sepals and leaves on May 22. This indicates that ascospores were mature at some time during the rain periods of May 3 to 9. A moderately severe infection period occurred May 18 to 19 and a lighter one May 30 to 31. The most severe infection period of the season occurred June 8 to 10. Abundant infection also occurred in the periods of June 15 to 17 and July 9 to 16. The season was one of severe scab infection, the percentage of scabbed fruits on untreated plots varying from 85 to 100. Treatments and Results A summary of treatments and of the results taken at harvest appears in Table XIII. Supplementary data relative to dust applications follow : Treatment 1, Learner's. Dusted from 8:00 to 10:00 A. M. Light rain prior to and during the first hour of treatment. Moderate breeze (about 5 miles per hour). Treatment 2, Learned's. Dusted from 10:00 to 12:00 A. M. Light rain up to 8:30 A. M. Leaves fairly dry when dust was applied. Light breeze. Treatment 3, Learned's. Dusted from 9:30 to 10:00 A. M. Light rain just before and during dusting. Light breeze. Treatment 1, Goff's. Dusted from 4:00 to 5:00 P. M. Trees wet from rain just prior to dusting. Light breeze. Treatment 2, Goff's. Dusted from 2:00 to 4:00 P. AL Trees dry. Moderate breeze. Treatment 3, Goff's. Dusted from 9:00 to 10:30 A. U. Trees dry. Moderate breeze. Treatment 4, Goff's. Dusted from 9:30 to 11:30 A. M. Trees dry. Strong breeze. The control of scab was not consistently satisfactory with any fungicide used, due to the fact that abundant infection occurred before treatments 0 were made. In general there was little difference in scab control between Bordeaux mixture and liquid lime-sulphur (plots 190, 192, 204, 209, 217, 219, 223. 225, 227, 229, 191, 193, 205, 211, 218, 220, 228, 230). As in previous years fruit russeting made the results from Bordeaux un- satisfactory. The results from dry and liquid lime-sulphur were not essentially different (plots 197, 192, 198, 193, 212, 210, 221, 219, 222, 220, 231, 229, 232, 230). The most striking feature of the results from treatment 0 is their variability in different orchards and on different varieties. Unfortun- ately, much of the value of this treatment was lost because the spray 74 Wisconsin Research Bulletin 73 o ? Hi lO TO -^ ■* C^ O CX) X ■* CO 30 IN t~ t~ CO .-H t^ t^ r-(l^iO^^ O) OO— "-iMtN-H £"^ OC^l-HOC^CO— 'OO— '—. — C-) OfOINCOOMOOOOOOO '^ "■ OOOCO'H.-i.-HrtC^rtOO'-i.-ir-i OCD05C<3COCOOC — OO Tt)'-H-HCOiOcO CltK'COXi-iC^OOOOCOOO i-H rt CC C-l M CO ^- -ji— .cDCiO--iMOOi-';xXT)(o> o)cococ^roM-Hioocroc^'H(N CD "O lO Tt< o rt •!< •* O C^ O CC •>!< C^ O O O O —i C O O — ■-• W "-i rH ooooco-4iNO-1*'.OCOt^l>OCD—l &"" O^C>)i-iTtioOOCOCOOiC3-t"^ t^CO'OTlf^cDiOCO-HOCCCOCOW O'-ii-iO^O'^OOOCMOO "nOt^O5CD-*a0COQ0 tD(NIN-HO .I c^ Tl< 00 -H "O CO tC CO O CO 0> r-( &~- O -* t^ CO t^ CJ O IN — I O) 00 CD rt< CO CO (N CO O -h —c —i O ^ IN CO ■-i IN T-l rt rt rt IN c^coooqcDoooiococoici^^oo O -# O 'C lO -^ lO O -1* lO 00 IC -C lO 00 5- -HTf(iOC3C0O:Oi-1O OOiOt^.-^CT. C^JO'Nt^OOit^iN 05CO>0 00'-iiO-HTf>cDCO'HcD ■O CD CO CO CO CO C^ -^ C^ CO O) M M (NC^INCOOqiNCOOlC^iNOlMMCO C^-»-^ - ' co-^ J," -^'dScO^Q ■ I'M CO OO'-i'NCO-fi-OCOI^CCOO— ICS CO-fi.OCOt^XOO— iiNCO^O '00 o; o C5 a: c oj c: r". C3 CI c o o <— ^ '^ .^ .— -^ ^ — -^ !:_"■,__•: l'jMi)i:.\i if)i.o(;\- AM) C'oxTRor, of Ai'IM.k Scab 75 O 10 O M i— -< 0> 00 » >0 O CO O 'O CO C^ lO O C'l O O O ^ CI O —I o>nci---<-^!D oo-*M rtOXM-^-HF-i ©■-"Oo-hocsoo o'ffl'*ddd-H d^d d'-o o d d o) 01005 O-.O'-( CO O'*OC0'-ii0C0(N— I drtoddd^dd c^I'TC^sj-^Oti" cooo co:o:o-i --o o; '-1 CO -H CO M XO-^rHO-HTJHCOt^ >*< ® OS CO O 'H 00 COOtO o N CO X N o — 1 COXOOJiOOJiOcOOi ej th lo Th iM ■* o CO-* CO O -H O CO 35 CO O X o o 1-1 o ■* m X c>i OOOOINCO-H lOOCO O lO l^ lO O M -H X -H X O X M- CO O Tf 0<0-*00t- 00 CO coo O Tf M X lO 05 --0 C^) C^ r-c C^J O) 01 Ol -H O ^ .-1 t^ -H 00 -H lOOltNOOXO Oi-OO) 0-0 0 05r^Xt~ O — J^XXCOOCOO) Moxocicoo — >ci-* r-'^jcoooc-i'-i t^cococox^t>.-f'-i tD-r-JiCiX-^O i-icCH ^-i-— iX-HiOO] -tco-,o M -)f rt C^l — I CO ■* Tl r-( rH Cq rt OJ CS CO CO 04 O) M rH rH CVJ rH O) Tj<-*CO-*'*T}> ►-5 C3 CO S o ■^^ ^ ^ "S.^ 11 X'S u ^M 9 "--^ » C "? >'rk c a 4 fe^ 3 ° a o.J. 3rH m c3 ~ Pw 30 II O S . A"-a M _ 2 a t. 't oj S - 3 ° M_2 O 35' ri O f cD-rr 3 O 3::3 CO 3 O - 1 ' IT 3 S'-5 M. . 3 CO , - C3 3 -^ a g> ^ c II o"'«^ i '^ . o o - 1) ai o >>oj X -r. "S » 4T!^J^rHrH- ; Tf-O _g ^ ^"'~- -■3 Cj 5S SS r^ -S , « =! =_, M 3 : - m a. l^ a >v so . ■^ ^ 2 m 3^ = - x-3^ '^ 9'--:o ■J °S o - o l[-^ .^^"°5|| Ill2 <. 7^ Wisconsin Research Bulletin 71^ was not applied until well after the only important pre-blossom in- fection period of the season (May 3-9). In spite of this fact it greatly reduced the percentage of scabbed fruit on the more severely affected plots (maximum reduction of 45%, plots 229, 230). On the less severely scabbed plots the effectiveness of this treatment was relatively slight. The following factors were prol)ably important in relation to these variations: 1. The greater abundance of ascospore material on the more severely scabbed plots (Sackett orchard), leading to more severe sepal infec- tion. Counts made on May 22 on 50 fruits selected at random on each variety showed sepal infection on the following percentages of fruits : Dudley (Sackett's), 66; Wealthy (Sackett's), 36; and I^ubsk's Queen (Learned's), 10. 2. Variations in the stages of advancement of the opening "fruit" buds at the time of the pre-blossom infection. This may in part ex- plain the greater severity of the early outbreak of scab on Dudley than on Wealthy in the Sackett orchard, since yearly observations have shown that the Dudley buds open somewhat in advance of those of the Wealthy. 3. Variations in the timing of applications. Treatment 0 was ap- plied one day later on the Sackett than on the Learned orchard, and slight differences occurred in the timing of the later treatments (Table XIII). The addition of gelatin and glue, respectively, to liquid lime-sulphur appeared to detract from its effectiveness in scab control (plots 194, 195, 192). The apparent slight increase in the effectiveness of this fungicide by the addition of casein-lime is proI)ably not significant (plots 196, 192). There was no significant difference in results from experiments in which the spray rod and gun were tested comparatively (plots 209, 210). Bordeaux mixture, 4-10-50, gave essentially the same seal) control as Bordeaux, 4-4-50. The apparent decrease in russet which attended the use of the 4-10-50 formula is of doubtful significance (plots 206, 2(W). Both Bordeaux mixtures, however, occa-ioned too much russet to be satisfactory commercially. The substitution of lime-sulphur in treatments 1 and 2 of the full Bordeaux program (plots 207, 204) re- -•'.ilted in no significant difference in seal) cimtrol and (il)viated most of the fruit russeting. The results from this program, however, were less satisfactory than from the full linie--uliiliur program (jilot 210), because of inferior finish of the fruit. The substitution of Bordeaux mixture for lime-sulphur in the third treatment of the full lime-sulphnr ]ir(igrani was planned in relation to fruit burning. I{owever, no fruit hurniuL; developed on any of the experinu-ntal plots in 1922. This substitution did not significantly affect scab control or russet. However, the fruit which received this program was somewhat inferior in appearance to that from the full Hi'IDKM lOI.or.Y AKD CoNTROL OK Al'I'l.K ScAB 17 limc-sulphur profirain (i)lots 235, 234). The treatment of plot 236 was parallel to that of plot 235, except that treatment 3 (Bordeaux) was delayed 19 days. This was done to make it come at what ap- peared to be the most advantageous time for codling moth control. Under the conditions of this season this delay had no significant effect on scab control or fruit russet. In the Learned orchard each of the three dust programs used controlled scab witli apjiroximately the sanxe effectiveness as did the full program of liquid lime-sulphur (plots 200, 201, 202, 192, 213, 214, 215, 209). No significant fruit or foliage injury was caused by the dust treatments in cither the Learned or the Goff orchard. In the Goff orchard, however, the dust treatments did not control scab satis- factorily (plots 239, 240, 241, 234). These sharp variations in the effectiveness of dust programs appear to be due primarily to the differences in the timing of treatments and in conditions under which the various applications were made. In the Learned orchard each dust application was made when the trees were wet, and, with the ex- ception of tile first treatment, during or immediately preceding a criti- cal infection period. In the Goff orchard, all applications except the first were made on dry foliage, and usually in stronger wind than in the Learned orchard. Furthermore, the first application in the Goff orchard was made a week later than in the Learned orchard. No in- fection ])Lri(i(!. however, occurred during this interval. In the Goff orchard a mixed program consisting of li(|ui(l lime-sulphur spray in treatments 0, 1, and 2 and sulphur-dry linie-sulphur-arsenatc dust in 3 and 4 gave essentially as good seal) control as did the full lime- sulphur program (plots 237, 234). Experiments in 1923 Condition of Plots Learned orchard. Experiments confined to I.ubsk's Queen trees used in previous years. Orebard disked four times early in season. However, sod remained practically unbroken in radius of about five feet from base of each tree. Sackett orchard. Same orchard of Dudley. Wealthy and Duchess trees as in previous season. Disked four times prior to June 20, first cultivation being made May 8. Thorough cultivation continued until July 15. Seasonal Conditions Seasonal development and meteorological data appear in Figure 5. The first ascospore discharge from the seasonal development ex- periment was recorded on May 20, when the "blossom" buds were in the late closed cluster stage. It is probable that ascospores were ma- ture considerably earlier, since an examination of field material on 78 Wisconsin Research Bulletin 71) May 19 revealed many perithecia in which 50 per cent of the asci were mature. The rain periods of May 9 and 15 to 16 appear to have been of insufficient duration to permit infection. The first in- fection of the year appears to have occurred May 19 to 20. The second infection period, which was the most severe of the season, occurred during the rains of June 3 to 7. Other infection periods occurred July 6 to 7 and 2Z to 24 and August 6 to 8. The season was one of moderate scab infection, the percentage of scabbed fruits on unsprayed trees varying from 31 on Lubsk's Queen to 96 on Dudley. The scarcity of rain periods of sufficient duration to permit infection (hiring the critical period of spring and early sum- mer accounts for the lack of a more severe development of the disease. Treatments and Results A summary of treatments and of the results taken at harvest appears in Table XIV. Data regarding early infection on fruit are found in Table XV'. Supplementary records pertaining to dust applications follow : Treatment 1. Dusted from 9:00 to 10:00 .\. M. A light rain fell prior to and during dusting and continued until 4:30 P. M. Light breeze. Treatment 2. Dusted from 6 :00 to 7 :00 A. M. Very heavy dew. Trees wet. Very light breeze. Treatment 3. Dusted from 8:00 to 9:00 A. M. Trees very wet from a light mist which continued throughout the day. \'ery light breeze. Treatment 4. Dusted from 10:30 to 11:30 A. M. Trees moist from a fog which (lissijjated at about 1(1:00 o'clock. Wind velocity. 8 miles per houi'. Plots 251 and 252 were dusted on the dates of spray applications. All these treatments were made on dry trees in light to moderate breeze. In each case, however, the trees in plot 251 were wet just prior to dusting by being thoroughly sprayed with clean water. This was accomplished l)y connecting the spray rig between the tractor and the duster, and heavily spraying the trees by means of rods, using the same technique described above for applying sprays. The trees were thoroughly dusted from two sides. Lime-sulphur and Bordeaux showed little difference in scab control (plots 244, 243, 261, 259, 274, 272) when the lull program was used. While Bordeaux russet was less severe than usual, it was sufficient to render the results from this spray unsatisfactory commerciall3^ On Lubsk's Queen, which was not severely scabbed, dry lime-sulphur gave about the same degree of scab control as Bordeaux and lime- sulphur (plots 248, 247, 243, 244). On Dudley and Wealthy, which were more severely affected, dry lime-sulphur gave a somewhat less satisfactory control (plots 263, 259, 261, 276, 272, 274). An unusual amount ■of russet occurred on Wealthy. A part of this was apparently oc- Epidemiology and Controi. ok Apple Scab 79 casioncd l)y injuries from ,si)riiiR frosts. Because of the irregularity of its occurrence, little significance can be attached to this russet in relation to spray injury. In every instance, the addition of application 0 to the program mater- ially increased its effectiveness in scab control (plots 244, 245, 259, 260, 261, 262, 263, 264, 272-277). The data contained in Table XV ac- count for the effectiveness of this treatment. All programs of Bor- deaux mixture or liquid lime-sulphur which included this early ap- plication controlled sepal infection adccjuatel}', whereas aljundant sepal infection developed on all plots which did not receive a pre- pink treatment. On June 28, when the data of Table XV were taken, comparativeh' little scab had developed on fruits which were not in- fected on the sepals. A high percentage of the sepal-infected fruits, however, had developed secondary infection in spite of protection by the later fungicidal treatments. Certain variations and substitutions were made in treatment 0 of the full lime-sulphur program. Where 90-10 flowers of sulphur-lead ar- senate dust was substituted in this application, slightly less scab de- veloped (plots 250, 244). No consistent difference in scab control at- tended the substitution of Bordeaux with or without casein-lime (plots 266, 265, 261, 279, 278, 274). In treatment 0 of the lime-sulphur program, no significant difference in control attended early application, increased concentration, or the addition of casein-lime (plots 270, 268, 269, 261, 282, 280, 281, 274). Slightly less scab developed where casein-lime was added to lime- sulphur in the full program than where lime-sulphur was used without a spreader (plots 244, 246). This difference appears to be insignificant. The substitution of Bordeaux mixture for treatment 3 of the full lime-sulphur program led to no important change in effectiveness of seal) control or in the amount of russeting (plots 249, 244, 267, 261), but occasioned considerable foliage injury. This substitution was planned in relation to its possible bearing on fruit burning on lime- sulphur sprayed trees. There was not enough of this type of injury on the experimental plots to give a satisfactory test in this relation- ship. Excellent control was oljtaincd from each of the five dusting materials used. These tests, however, were not fully satisfactory in view of the fact that all the dusting work was done of Lubsk's Queen, on which the development of scab was light (31% of fruit scabbed on unsprayed plot). The various dust materials controlled scab on this variety with approximately the same effectiveness as did Bordeaux mi.xture and lime- sulphur (plots 253-257, 243, 2-14). A considerable amount of russeting attended the use of copper carbonate and the sulphur-dr\- lime-sulphur- arsenate dusts, and, to a less degree, the copper-lime-arsenate prepara- tion. Where flowers of sulphur-arsenate dust was used comparatively on wet and dry trees (plots 251, 252) there was no significant differ- ence in results. The control in this experiment was not significantly 80 Wisconsin Research Bulletin JZ —iiOOO'-i-H'HO'-iO'^t^.OH^'JCO OfHr-40000'-"fHO>-i "-lO 6^ OiOOOOOOOOOO OOC^ O-' ooooooooooo oo 6^ OrJfOOOOOOOClOOi-iOCOaO O^-HOOOOOCSO CO "Oc^i-HOt^io©iOin-,ot^t^'NcDc^r~ oimcoo^htow^ccc^m t^-i< V- OOt-^— lOOOOOOOMOlcOiMCO oiOiOOOO'-<-H"00 CO .- C-) CI Q CO 05 00 (N t^ to 00 >0 1-1 M ira IN o> C^ O --I o -> ?? -f •- -^ o on; COOO'J'O^OCI'-I'-hOOOOOO (MOOOriroC^-fCl — MC-1 csc^ ^^^^COCOO'OMOCOrOrHC^OC^'-l OLO-OO'.O'.'JXOCOOt^ C". t>- CO CI rt O (N i-H -H C) OOCOOOM^t^OCOX-HOJCOOOOtDrH OM!MS5Ot^OC0i-(C»X tCt~ t- ooO'-iio^'coMcod.-iodoood od-)M«MO>— lOC^M MPO •O P3 00 O 00 ■* IN ■* 00 "O d '-' b- "-I lO IN tj< lO 00 00 00 CO lO O 03 O l^ CO t>. &^ iOOCOCOOC^'-h^OO'HOOOOOO OWINWcD'-i-"?*CO»-iC1CO -tci •-HaiioccoNt^ioco.-io i-i^ t^cocoiOt-iootoo3»i<^ T»i>n ° 2 Hi; lNC-)CmNiNM*M<'*'*'-*'J'C0-<»< CO-* IN CO r-<""N a - . . -■*' ajo . .ajCM' N (N -^^ '"' -S-^So '^.'^..^•'vmco-coSco ©•gj _CC^c*30't*50'^l M^ e^tooo-0'-i>-i"-< c^rt COCOOOOOC^OiMMt-. tOi005C^OtO-*OMt- OM d rn to rt 00 !0 r-i M r-l CJ Tjl O MOl^~^^00 tDI^ COCO lO t>.rt< lO O 0> O — 1 T)< X C^ CC CO -^ o oc ')< O) o T)i . - , , t6 1 ■ •"J" t 1 0J3 [ H Jj ' L- I ^o ; ■1 :s+ 1 S ; CO s^ iT3 loo !■* •* of° oP. I'o" iCO'^ i< '^ £ I""^ 1"* ti 101 CO CO't ^"oi d-t M r^ co"^ >iCO Soj ^ -^-^•^ C'ln ofoi w ifed- °+ -<'- O-H oo rtO >• KX kJh-J ss "S Sffla ^^ QG «ffl ^Jj ' J u ^ -^OJCOt^-O-JDI^XOO-^ n t>- r- 1>. t~ (^ t^ r- r~ 1^ X X x OOJNOIOIOJOIIMOIOIN oi .2= J S Ijio — 3 c 3 J2 ^^ 1 :_; S3C^ >> ;:: " -> II 2 OS o O 3 ■-5 co" o -a II X 03 ZJ c Q O a 3 ■-3 ^ c 2 c i-s 3 s 2 : §• •■* 3 O II G . « C^ 3 C I >>■< .'-O - ^»-n « .to M :2^ .^'^ ..o M o- --S°oi& . -S^ -■« 3 = aoco ..s . - '? £; e M d § «.:= a^ t kE-^^^"* H " ^ £:5'iO so v;-S g * = 1 S .2^* =--5 2 . «■="-; -g >>-2 ° :3 ^"3 O 'C = ^- ^— 3 2 3 =*> * pa §•» 2 M ''•a 2-1 2 >> II ^I^OO-oiill &T3 a ;3< a '-< 82 Wisconsin Research Bulletin 7^ S: a o'5 g ■ 5j= 03 c c-^c^i"'X'^!/;co-i0XNqC-f'MCXCD'*0C^10 ^ ' in po -* m in — - p^ — -- 0 r-H CO^ rt :9S^ + 0ST3 ^■^XiS i^^ 03 5=^ 1-1 "2-->:^-::;-o:h- cc ' mi! -,-c<5 a; ^ M o fi= , -- ^i" N 'i. -■*.!- "3 r-' . <-. x CO ►-.>-: ^ 1=" xo-.o-^McoTtf'Ocooocso wro-i" it^XCOC'-^OlMrfmcDt^ --T. Tr -T -T -^ -f -^ »0 IC to IQ »C »C »C >0 C^ ?^ C^ C^l 0) ^ !M C^l ^J C^ ^ nj n) C^J nj ■ - 03 o o £-2 ;S E l^PIDKMror.OGY AND CONTROL OF ApPLE SCAB 83 (litTnciU irciiii tliat of llu- same material applied according to the regular dusting program (plot 25.^). It is worthy of note, however, that few infection periods occurred in the critical stage of host develop- ment in spring and early sunmier, and that the spraying and dusting programs were very fortunately timed in relation to these periods. Under the conditions of this season, the spray applications, which were timed largely according to host developiuent, and the dust treat- ments, wliich were timed primarily according to infection periods, were applied very nearly i)ara!kl throughout the most critical time for scab infection (Fig. 5). Experiments in 1924 Condition of Plots Learned orchard. Same Lubsk's Queen trees as in previous seasons. Orchard disked twice, beginning just before blossoms opened. Sod remained practically unbroken in radius of about five feet from base of each tree. "Set" of fruit was light. Sackett orchard. Same Dudley and Wealthy trees as in previous years. First cultivation made with disk harrow May 16. Cultivations made at frequent intervals with spring-tooth harrow until July 5, when Hubam clover was sown. Seasonal Conditions Seasonal development and meteorological data appear in Figure 6 and Table 1.^ The first ascospore discharge from the seasonal development experi- ment was recorded on May 6, when the "blossom" buds were in the early green tip stage (see PI. III). The first ol)servations on ascospore development were made on May 5. The abundance of mature asci found indicates that ascospores had been mature for some days. The first infection period of the season occurred May 5 to 10. A second oc- curred May 12 to 15. Other infection periods occurred June 15 to 17 'and 28, to 29; July 7 to 9, 15, 29 to 30; and August 5 to 6 and 15 to 16. The season of 1924 was one of unusually severe scab infection, the percentage of scabbed fruits on unsprayed plots varying from 92 on Lubsk's Queen to 100 on Wealthy and Dudley. This severity of oc- currence is attributed in large measure to the abundant early es- tablishment of the fungus on the sepals. The foliage of Dudley was heavily infected, while that of Wealthy and Lubsk's Queen showed verv little infection. >In making cross-references it should be recalled that in Table I days extend from midnight to mid- night, whereas in Figures 1-6 they are taken as 24 hours preeeJino; 8 .\. M. of the dates listed. 84 Wisconsin Research Bulletin 73 Treatment and Results A summary of treatments and of the results taken at harvest appears in Table XVI. Data regarding early infection on fruit are found in Table XVII. Supplementary records pertaining to dust applications follow : Treatment 1. Dusted from 9:00 to 10:15 A. M. Wind velocity 4-5 miles per Imur. Trees were wet from fog and a light drizzle which fell during tiie operation and continued at int>.-rvals mitil next after- noon. Treatment 2. Dusted from 7:15 to 8:30 A. M. Wind velocity 5 miles per hour. Trees were partly wet, following a trace of rain at 7:00 A. M. Treatment 3. Dusted from 9:30 to 10:30 A. M. (plots 294, 293), and 2:00 P. M. (plot 292). Treatment interrupted to repair duster. Wind velocity 10 to 12 miles per hour. The trees were wet from rain which fell intermittently during the day and steadily through most of the night. Plots 290 and 291 were dusted on the dates of spray applications. All these treatments were made on dry trees in light to moderate breeze. In each case, the trees in plot 290 were sprayed with clean water and dusted by the method described on pages 60-61. On Lubsk's Queen, it would appear at first glance that Bordeaux mixture had given somewhat better scab control than liquid lime- sulphur (plots 284, 285). A consideration of the standard deviations, however, indicates that this difference is of doubtful significance. On Wealthy there was practically no difference in control between these fungicides (plots 308, 310). On Dudley, however, Bordeaux gave a strikingly m,ore effective control (plots 296, 298). Much of the in- fection on the lime-sulphur yAot api)eared in the latter part of tlie season (see Table XVII, plots 296, 298). It appears prolKible, there- fore, that the difference in effectiveness of these programs is to be at- tributed largely to a longer duralioii of effectiveness of Bordeaux in the final application under tlu' very severe conditions of scab in- fection on Dudley. Bordeaux iin'xture russeted thi' fruit severely and occasioned considerable defoliation prior to harvest. Dry lime-sulphur appeared to give somewhat less satisfactory results than the licjuid product, except on Dudley, where both programs were relatively in- effective (plots 288, 287, 285, 300, 298, 312, 310). Dry lime-sulphur, 3-50 and 4-50, used comparatively showed no signilicant difference in the control of the disease (plots 287, 288). Application 0 showed considerable variations in eff'ectiveness on the different varieties and with the different fungicides. On Lubsk's Queen in the lime-sulphur program (plots 285, 286). it showed no significant increase in effectiveness of scab control. On Wealthy, whereas treatment 0 in the Bordeaux iirogram (plots 308, 309) gave Epidemiology and Control of Apple Scab 85 > o 'O — M f'i t- — ci -f cc o c- '•'^ — — — — "' — "I c; c OM-hM — C — MM-<-<0 o — c:ocio-OO ^J cc o CO to M t>. t^ f M '-' M o 't ?3 'O u; o ^ o) X c^i x r>. o c; x t^ -i x -i •* • O c: M ro O C^ C: -^ •-'^ »■* ^ ^ C '* :C O •-' ro lO :D O ^h ^t ^) C »w O O C O w — • I '^ t^c-^TXc-.n — C'ic: a'»s o o E2 M -« -H r, — . h, 1 1~- 1^ wo M C5 — — M — . — . — M T}< :o xr'ic:£=500«:c<3-^«x .-Hr-CKlTO'*MOC<5W*OX t^ X M c: o m >-- (^ O M 'C ro !0 o '•': -* ^ m X O M » >n CO MiOXXX-^:Ct^t~C-)'^iO 5>JOMi-iC»3t^X!NmM"*C'I r-lr^ O; —I rt.-IM'-l rt '-C t» S3 X I^ h- IC C<; X X t- O 1.0 ;o iC t~ CM C^l — -J X o fo o ;MMO'^W0 1"C>5 0: CMC>5 tCiOCOCOO— 'tCX'OCi-^-.C — ■ M M C>1 M CO M -^ — •- M C! o X i-o X t- o: ■* "O ci 'o ro t^cq MC5C: tCMtOX COt^t^ CJ O X C<3 to — I "C ■* in ■* X IM ^xx^■. — — w t^'Ox — r^tt-cc^joxo cilomcoc". cc'c: :C'>i"Mt^r~xo-^CM"'0'.o -.o-f-r'r;D-.cc^c>jt^orcc; t^ a: ^ ^ -r -^ c. -r y: s ~ o noic-jC')MNCMMC-iMc>)N co■^*>^;x«T^^TOWc<5«^o^5 ^3^3eocQ^occccroM^;«« C<5« i d" "♦^ cs r= CO -J- S3 .<<<< <■ « o rt -S n f < u < ►^ >1 "^ ± u ^ d++^?"'r rO-T .- . . >.?§§ : r jq\-:J)3QxxxxxO~— ^?QShJKi2C;c— :J ir-iJ^'^^GKp-if-i '^ «^ .^- .,■ — — • CO c^i '"^ d ^ '4i 'Tf '--<■* CO COT}- . ^^■"co - - -o XX r -h" .,=3 = X X x: X CO'^'OCCt^XOiO — OICO'* XXXXXXXO:C5C-. 005 _ _ _ _ _ _ ^) c^ M c^ c^ c^ c^i ^ ^1 0^ CM M c^j c*) c-i CM c*! CO ; ^coco cocorocococococococococo -O . >> --^'^ ^< ::; ^ oIlo i IsC"^ ^ 3 ,. s X = ./ J; >>J . C.J; Bl-H &2 ^'CMCM . uS c o =S « 3 Si t3 co :^ =! Sot: - — >> c . o - > "-S t 53 ^-g ~ 53 C3_j « li-rdg'SI^I^.! ?~^-= t£.SS=3' -i-2 o ^ ^a o o C^O-tOCOOOO'OOOO C^liOOO'-OX'OOOOCOO OOOOiOOOtDOOOOS cooooooooooo 00-* O— HDOOO-* (N'* 0> O: 05 03 'H O Tt< O OCO-*OOt~05MCOr«<0-' "OCOt-XOSO— iC^COiCCO-J* ooooocxooooccoio>050> ocnoioojcoooooo (MCSINNC^C^C^C^OIINC-IC^ C^C^C^1C^4!MCOCOCOCOCOCOCO ^ > 2 >^ 2H S 2 a,S i'li'i[)i;.\iioLO(;Y AND Control of Apple Scab 87 no increase in scab control, in the li(|uid and dry Iinie-suli)luir pro- grams (plots 310-313) a similar application improved the control. On Dndley this application in the Bordeaux (plots 296, 297) and dry limc-snlphur (plots 300, 301) programs gave increased control, l)ut showed little effectiveness in the case of liquid lime-sulphur (plots 298, 299). The effectiveness of these early treatments was much marred by the fact that important infection periods occurred before they were applied (see Fig. 6 and PI. III). The variations in effective- ness appear to have been influenced largely by differences in the sever- ity of these early infections on different varieties, particularly on the sepals. On June 2 (at the beginning of the blooming period) counts of "blossom" buds showed the following percentages to be sepal- infected: Lubsk's Queen, 12; Wealthy, i7 \ Dudley, 74. These varia- tions in the amount of early infection appear to have been due chiefly to differences in the amount of the ascigerous stage of the scalj fungus developed in the fallen leaves of these varieties. On Lubsk's Queen this development was very sparse, due to the scarcity of foliage in- fection in the previous year (Fig. 5). On Wealthy the ascigerous de- velopment was intermediate in amount, wdiile on Dudley it was ex- tremely abundant. Tlie importance of early sepal infection in rela- tion to control is further shown in Table XVII. Fven in this sea- son of severe epidemic development and inadequate control, secondary infection prior to July 10 was readily held in check on the sepal-clean fruit by all the programs tried. On the sepal-infected fruits, however, the protection given by the strongest programs used was inadequate, and abundant secondary infection developed. Certain variations and substitutions were again made in treatment 0 of the full lime-sulphur program. The value of these experiments was greatly lessened because of the infection which occurred prior to the earliest treatment. Consequently, the results appear to be of doubtful significance (plots 284-289, 296-304, 308-316). In view of the high standard deviations, the results from lime-sulphur with and without the addition of lime (plots 306, 298, 318, 310) do not appear to be significant. Where sulphur-arsenate dust, 90-10, was applied in comparison with spray applications on the spraying dates the results relating to scab control appear at fir, on Baldwin following a single application. Regarding the time of treatment, they state: * * * "In many cases the buds had already burst and in some cases the leaves were well out, while in others only tlie tips of the young leaves were beginning to appear." Scott and Quaintance (1907, p. 21) state: * * * "The fungus may attack and destroy tlie l)]ossoms and even the unopened buds . . . The scab first appears early in the spring on the young buds and unfolding leaves, and new infections may con- tinue to take jilace throughout the season." Nevertheless, they rccom- nuend that the first scab treatment be applied . . . "after tlie cluster buds have opened, but prior to blooming." Cordley (1908) calls at- tention to the value in scab control shown by delayed dormant appli- cations of lime-suli)hur for scale, and states that it was such obser- vations that led him to try lime-sulphur as a .'•umnier spray. Beginning in 1913, Blair and his colleagues conducted an extensive series of spraying experiments in which programs with two pre- blossom scab treatments were compared with otiiers wliich carried only one. In l'>l,i the first pre-blossoni application, w liicli was made "when tlu' leaf buds Vv'ere unfolded slightly," played an important part in scab control (Robinson, 1914). Discussing the work of 1916, Blair 1917, ]). \M)) stalls: * * * "The results would seem to show that one spray [before bloonil may be sufficient to give reasonabi}' good scab control. The results were very similar to those obtained in l')14 and 1915. However, as i)ointed out last season, should the leaves expand early as in 1913, due to warm weather, but be delayed in their de- velopment by cool weather, the period fpre-blossom] should be covered by two applications of spray mixture rather than one." Jackson and Winston (1015?) noted tliat in 1''14 very beneficial re- sults in seal) Cdutrol followed the applic:ition of a delayed dormant lime-suli)liur sjiray in certain coniint'rcial orchards. Winston and Childs (1916?) report that in their spraxini; experiments of 1915 a de- layed dormant application . . . "in i)ractically every case produced de- cidedly beneficial results." Childs (l">17) made a valuable study of the periods of ascospore ejection at Hood River, Oregon, in 1916. He reviews previous literature and states (p. 11): "The study of ascospore discharge demonstrates beyond a doubt that the delayed-dormant ap- plication given while the leaves are small and undeveloped cannot be Epidemiology and Control of Apple Scab 91 safely dispensed with in tlie llood River and VVillanicttc Valleys, at least, since the ejection of ascospores begins before the foliage has even started. This would probably also hold true in all sections of Western Oregon and Washington ... It is n(jt at all unlikely that other sections could also be included except for our present limited knowledge." Morse (V)]() a, p. W-92 and 1916 b, p. 180-182) reports no signifi- cant advantage in scnb control following a delayed dormant applica- tion of lime-sulphur in bis experimental orchard (see also Morse and Darrow, 1913). The same writer (1918, p. 116) reports unsatisfactory scab control from all programs used in 1917, and states ; * * * "Un- doubtedly in i)ractical work an additional, earlier api)lication of a fungi- cidal spray when the leaves are about one-fourth inch in diameter . . . would be very effective in Maine under such conditions as these [1917J." Whetzel (1918) calls attention to the importance of preventing prim- ary infection, and reports that in 1917 growers who applied the de- layed dormant spray got best results. He states : "As a matter of insurance, this delayed dormant spray may well be made." In 1919, therefore, the commonly accepted recommendations for ap- ple scab control throughout the north-central and north-eastern apple belt of the United States called for the first treatment shortly before the blossoms opened, and the need appeared to exist for a more de- tailed study of pre-blossom infection iii relation to the later develop- ment and the control of the disease. In a preliminary publication on such studies conducted in 1919, the senior writer (1920) reported that the heaviest discharges of ascospores at SturLjeon Bay, Wis., occurred prior to the open cluster spray, and that on certain badly scabbing varieties programs beginning with the U'-ual open cluster application failed to control the disease satisfactorily, whereas excellent control resulted when an earlier pre-blossoni treatment was added (for further preliminary reports on this work, see: Keitt, 1921, 1922, 1923, and Keitt and Jones, 1924 a, 1925 b). The present writers (1924 1)) reported on the special significance of sepal infection in relation to the epidemiology and control of apple scab. Since 1920, a valuable body of data has been accumulated in the north-central and north-eastern apple belt of the United States, the general tendency of which is to attach increasing importance to the occurrence and the prevention of scab infection prior to the open cluster stage of bud development (e. g., Manns, 1921; W'hetzel. 1921, 1924; Button and Johnston, 1922; Bennett, 1923; Krout, 1923 1924; (.uilinan and Baker. 1924; Doran and Osmun. 1924; Thurston, et al, 1924; Walton, 1924: Williams, 1924; Butler. 1925; Morse and Fol- som, 1925). Working farther south. Schneiderban (1923, 1926) and Schneiderhan and Fromme (1924) have reported less valuable results from the delayed dormant treatment. These results show suflFicient varia- tion with season and section, however, to indicate the desirability of studies under local conditions as a basis for practice. Important gains in efifectiveness of scab control attended the addi- 92 Wisconsin Research Btjlletin 7Z tion of an early pre-blossom application to the regular spray pro- grams in four of the six years covered by the experiments herein re- ported. In the other two seasons, the pre-pink application was of little or no value. These results are readily understood in the light of the data presented in Figures 1-6, and discussed in other connec- tions. In 1919, 1922, 1923, and 1924, years in which the disease was not satisfactorily controlled on the more severely attacked varieties with- out the use of a pre-pink treatment, early spring rains of sufficient number and duration led to serious scab infection prior to the appli- cation of the open cluster spray. In 1920 and 1921, although asco- spores were matured well in advance of the open cluster stage, the early rains were of such distribution and duration that little or no infection occurred prior to the open cluster spray. From these data and from the results discussed elsewhere, it is apparent that the im- portance of green tip or closed cluster treatments is conditioned pri- marily upon the abundance and time of maturity of ascospores and the occurrence of seasonal conditions favorable for infection and disease development during the critical period preceding the open cluster stage. The experience of 1922-1924 shows that serious sepal infection may occur at a very early stage in the unfolding of "fruit" buds, and that subsequent control of the disease on fruits which sustain early sepal infection is much more difficult than on those which escape it (Pis. IIl-V, pp. 50, 54, 55, Tables XV, XVII). Under the epidemic conditions of 1924, two pre-blossom treatments failed to control early-season infec- tion on the more severely scabbed varieties. While this situation was foreseen, it was decided to adhere to the original program on the ex- perimental plots and test its effectiveness, rather than to introduce an additional application. In certain commercial orchards, however, a third pre-blossom treatnient timed between those at the green tip and open cluster stages gave good control. A program carrying three pre-blossom applications is being included in further experiments. The effects of variations in the pre-pink treatment are so greatly influenced by local conditions that conclusions are not justified from the data available. It is apparent, however, that timeliness of appli- cation was more important than the material used. No significant advantage appears to have attended the sul)stitulion of Bordeaux for lime-sulphur in this treatment, or the modification of lime-sulphur by increased concentration or addition of casein-lime. In two trials sul- phur-arsenate dust, 90-10, substituted for the pre-pink application of the lime-sulphur program gave as good results as the spray. Dust appears to have im]iortant potentialities for use in this critical period. It should be observed that even a program with three pre-blossom ap- plications is aimed ])riniarily at control of the disease on the fruit, and maj' not safe-guard against infection of leaves which unfold between fungicidal treatments (see Pis. I, II). In the presence of an abundant source of inoculum, the control of scab on the foliage of the more susceptible varieties is very similar in its difficulty to control of JUMDKMIOLOCV VXD CONTROL OF ApPLE ScAB 93 apple rust by fungicides. Often, however, scab is well controlled on the fruit, notwithstanding a considerable amount of leaf infection. The seriousness of such foliage infection in relation to epidemiology and control is discussed in other connections. The number and timing of spray applications. It is now coming to be increasingly accepted tliat no single fungicidal program for apple scab is applicable to all conditions. The immediately preceding discussion has indicated the necessity of adequate protection for the fruit, at least, during the pre-blossom period, and has shown that under severe epidemic conditions in commercial orchards three applications may be re(|uired for this purpose. Subject to minor changes to meet local needs these treatments would naturally be timed (a) in the green tip stage ( frnni the stage shown in PI. I, A, B to that shown in PI. I, C, D), (b) in the earlier part of the closed cluster stage (PI. II), and (c) in the open cluster stage (PI. II. D). Under less severe conditions, fewer ap- plications are necessary. If only two pre-blossom treatments are to be given, the most critical periods for application appear to be in the green-tip stage and just before blooming, preferably in the open cluster stage. In such cases, if good weather forecasting service is available, the timing of the tirst application may be modified to ad- vantage. If the di.-^ease is adequately controlled in the earlier part of the sea- son, its later control is comparatively easy. Supplementing the treat- ments just outlined, the following program of applications, which has long been widely used, has given satisfactory results: (d) after about three-fourths of the petals are off and before the calyx lobes close, (e) about ten days later, and (f) in summer at a time chosen for effi- ciency in codling moth control (ordinarily Aug. 1-20 at Sturgeon Bay). On early varieties, the final treatment of the program just outlined should be omitted. It should be observed that this program of applications is arranged primarily for scab control. In the Sturgeon Bay district, where in- festations of codling moth are comparatively light, it has not seemed necessary to accommodate the scab program to the insect icidal pro- gram, save in the final ajiplication. On the advice of the Department of Economic Entomology, arsenate of lead powder, 1-50, has been added to each application of spray. Further information relating to the control of insects may be ol)taineci fr')m the Department of Economic Entomology. The desirability of a mixed program of Bordeaux and lime-sulphur. The mixed i)rograms of Bordeau.x and lime-sulphur gave no significant differences in scab control from the full Bordeaux and lime-sulphur programs. While the use of lime-sulphur in treatments 2 and 3 materially lessened russeting, as compared with the full Bordeaux program, the fruit sprayed with this mixed program was ordinarily distinctly inferior in finish ;'.nd amount of russet to that from the full lime-sulphur program. 94 Wisconsin Research Bulletin 73 Variations in the lime- sulphur program in relation to fruit injury. During the six years of the experiments at Sturgeon Bay, "burning" of lime-sulphur sprayed fruit occurred in a significant amount only in 1921. In that year it appeared only on plots which received lime- sulphur in application 3, and was of minor importance when this treat- ment was timed according to schedule (ten days after petal-fall). It is evident that there is danger of this type of injury in the event that hot, clear weather occurs during or soon after a lime-sulphur appli- cation (Young and Walton, 1925). The substitution of Bordeaux mix- ture in treatments 3 or 4 of the lime-sulphur program has occasioned no significant difference in scab control. The fruit from such mixed programs, however, has ordinarily been less satisfactory in appear- ance than that from the full lime-sulphur program, and in certain in- stances considerable foliage injury has occurred. In 1922 a program in which there was a similar substitution of sulphur-dry lime-sulphur- arsenate dust gave as good results as did the full lime-sulphur pro- gram. The results from experiments designed to test the effect of the addition of lime to lime-sulphur in relation to fruit burning are in- conclusive. Caution should be exercised in the use of lime-sulphur on apples in hot weather. Effectiveness of spraying with rod and gun. The results of three years' comparative tests with spray rod and gun on trees of moderate size showed no significant differences. The effectiveness of adding certain spreaders to sprays. The ad- dition of gelatin and glue, respectively, to lime-sulphur appears to have decreased, rather than increased, its effectiveness in scab control. No significant difference in scab control appears to have attended the addi- tion of casein-lime to lime-sulphur and Bordeaux mixture, respectively. The effectiveness of certain dust programs. The dusting experi- ments were planned primarily with the aim of testing the effectiveness of the different dust schedules and materials under varied conditions. The results of 1921 are not significant because of the scarcity of scab in that season. In 1922 and 1923, fair to good control of scab was ob- tained from a dusting program in whicii applications were made dur- ing major infection periods when tlie trees were wet. In these tests there was no significant difference in scab control by ground sulphur- arsenate, coppcr-lime-arsenate, sulphur-dry limt^-sulphur-arsenate, cop- per carbonate (in 1923), and flowers of sulphur-arsenate (in 1923). On Lubsk's Queen in 1923, consideralile russet followed the use of sulphur-dry lime-sulphur-arsenate and copper carbonate in this pro- gram. In 1924 this program gave less satisfactory control than in the previous years. These results suggest that in emergencies considerable benefit might be derived from dust applications made during the in- fection periods before the fungus became established. Jm'jdkmioi.oc.y and Control of Apple Scab 95 In \^)22 ill tlu' (iot'f (ircliard, a program closL'ly paralk-lin^ tliat men- tioned above, but applied on dry foli.'igc under less favorable wind conditions and witliout immediate tinuliness in relation to major in- fection periods, failed to control scab. In 1923 and V)24. a du;.t ]iroL;ram in vvbich tlie applicatioiis were made on the same dates as the treatments in the full spray program gave essentially as satisfactory scab control as liquid lime-sulphur. Where applications of suliiluir-arsenate dust in this program were made comparatively on wet and dry foliage, therL- was no significant differ- ence in results. Experiments with sulphur and sulphur-arsenate dusts are being continued with an increased number of applications, parti- cularly in the prc-blossom period. OTHER STUDIES BEARING ON CONTROL MEASURES The studies of epidemiology which have been reported in other con- nections have shown : 1. That, under Wisconsin conditions, and probably wherever twig infectidu does not occur to a significant extent, an abundant supply of ascospores is essential to the development of difficultly controllable epidemics of apple scab. 2. That the most critical time for the control of scab by the type of fungicidal program now in general use is in the period of rapid expansion of susceptible host parts early in the season. 3. That the fruit phase of the disease is more readily controlled by fungicides than the leaf phase. 4. That the fungicidal programs now in general use are adapted primarily for the control of scab on the fruit, and may permit suffi- cient foliage infection to lead to the development of a serious asco- sporic inoculum for the following year. These facts make it increasingly apparent that great promise for the future improvement of scab control lies in preventing the develop- ment and timely discharge of an abundant supply of ascospores. On general grounds, measures directed toward this end have been recom- mended ever since the establishment of the genetic relations of the Venturia and Fusicladium stages of the scab fungus. Even before that time, destruction of fallen leaves was recommended. Yet, scab epidemics continue to take their toll, and comparatively little progress seems to have been made- towards checking them in their inception by other means that the direct protection of susceptible host parts by fungicides. Special consideration is therefore being given to methods for limiting the production and timely discharge of ascospores. The re- sults of this work will appear in a later paper. 96 Wisconsin Research Bulletin 73 SUMMARY AND CONCLUSIONS In trials made in Wisconsin in 1916 and l'^17, tlie programs then in most common use for apple scab control in the North-Central and Northeastern United States, failed to give satisfactory results. Con- sequently, work was initiated along two general lines : (1) field studies of the disease and its control in relation to the natural environment, and (2) laboratory and greenhouse studies of the development and pre- vention of the disease under conditions in which certain factors of the environment were controlled. For the years 1919 to 1924, inclusive, records of the seasonal develop- ment of the host, parasite, and disease are presented, in correlation with meteorological data taken in the experimental orchards, and records of control experiments. In general agreement with many pre- vious observations by other investigators, these data show that the moisture and temperature factors of the natural environment play a leading part in determining the severity of occurrence of the disease and the difficulty of its control. These records are further discussed in connection with the topics to which they are most pertinent. They have been especially useful in defining critical periods in epidemiology and control and in contrilsuting to a more effective orientation of control programs to these critical periods. Studies of the viability and longevity of ascospores and conidia of V. inaequalis and of relations of moisture, temperature, and light to their germination are reported. Infection experiments were conducted under conditions in whicli certain factors of the environment were contrcilkd. Asco.'^pores and conidia which were allowed to lie in drops of water in contact with hard surfaces rapidly acquired a high degree of ad- herence to the substratum, (ierni tul)es whirh developed in close con- tact with a firm substratum were firmly adherent throughout their entire lengtli. Direct ])enetration of the cnticir I)\- infection hyiihae from w(.d! difTerentiated ajipressoria, as described by .Aderliold and others, was vt'r\ eomnumlx- observed. Not in lri'(|nentl>', however, comparatively short, thick germ tubes appeared to function as appres- soria, without the development of clearly differentiated hold-fasts. In- fection hyi>liae fre(|uently developed direetl\- from ascospores. which functioned as ajjpri'ssoria without the formaticni ol elearl\- differen- tiated germ tnl)es. Relations of thi.s' type of pent'tration to the ra])id- ity of infection and to tlie action fo fungicides are discussed. Infection occurred at temperatures ranging from 6" to 26° C, with evidence that tlir tenii)eratnre most favorable for rapid development of the disease is near 20 C. (ierminalion of ascospores and conidia and growth of the fungus were observed to occur from J/. ° to 32- C. The oi)timal tenqn'rature for spore germination and growth of the fungus in watir or in certain culture media agreed closely with that for development of the disease. The temperature range for develop- ment of host plant appears to be somewhat higher than that of the parasite or the disease. I'J'IDKMIOI.OCY AND CONTROL OF Al'IM.K SCAB 97 The results thus far available suKJi;e>t that the rniniiiial periods of continuous wetting necessary for infection of api)Ie leaves by asco- >p()re-> i>\ r. !ii'Oung fruit b\- water-borne ccnidia. The sjic-cial significance of se])al infection in relation to epi- demiology and control is fliscussed. The production of conidia ordinarily begins before the scab lesion becomes macroscopically visible, and may continue under favorable conditions until the end of the season. Field observations showed that an abundant source of secondary inoculum was never lacking on un- sprayed trees after infection was once well established. In confirma- tion of the work of Frey and Keitt it is shown that conidia arc very resistant to detachment from the conidio])hores when dry, but quickly become detached in water. Their chief mode of dissemination in nature is in meteoric water moving under tlie influence of wind and gravitatif)!!. The occurrence of secondary infection is discussed. Under Wisconsin conditions the most critical i)eriod for the develop- n-rent of apple scab epidemics extends from the time the apical iiarts of the sepals are first exjiosed in the unfolding "fruil" buds until an indefinite time some two to four weeks after petal-fall. The most critical part of this period is that prior to tlu' open cluster stage of bud development. Other important infection jieriotl-, may develop at any time during the summer. A second critical period occurs in the fall, when cooler weather and sufficient moisture may permit serious late infection of fruit and abundant estal)lishnient of flu' fungus on the foliage. If jn-imary infection is adetpiately controlled, however, the disease is easily held in check later in the season. Control exiJcriments conducted during the vears 1910-1')24, in- clusive, are reported. Bordeaux mixture, licpiid lime-sulphur, dr\- lime-suliihnr, and varif)US mrixed programs of Bordeaux and liquid lime-suli)hur were tested comparatively. .Arsenate of lead was added to each >\n-:iy nn'xture. Bordeaux or li(|uid lime-sulphur, in approprialeI\- tinieil jirograms, or- dinarily controlled the disease satisfactorily. P)Ordeanx mixture, how- ever, was unsatisfactory cominercially because of injury lo fruit and foliage. Liquid lime-sulphur, 1-40, appeared to be the most satisfactory spray tried. Dry lime-sulphur, 3-50 and 4-50, in the full program of I'j'ii)i;.\i loi.oiiY AND Control of Apple Scab 99 tit-atiiKiits, ordinarily coiitrolKd tlit- (li^^(.•a-^■ in al)()Ut tlic- same degree as the liiiuid lime-sulphur, 1-40. The results of certain experiments, huwever, suggest that the dry product at these concentrations has a somewhat less margin of safety than lii|ui(l linic-sulphur, 1-40. Mixed programs of Bordeaux and linie-suliiliur showed no .significant differ- ence from the lime-sulphur [jrogram in controlling the disease, hut were less satisfactory than linii-suli)iuir from the standpoint of ap- pearance of fruit and injury to foliage. The tmie and numhcr t)f api)licati(nis of siira>- should hu arranged to meet the re(|uirement.s of individual situations. ihe necessity of pro- viding adequate protection of at least the hlossom parts during the critical pre-blossom period is discussed. For use under severe epidemic conditions in Wisconsin, the following program of lime-sulphur appli- cations is tentatively reconnnended, subject to modification in rela- tion to seasonal or local conditions: (aj in the green tip stage; (b) in the early closed cluster stage; (c) just before the blossoms open, preferably in the open cluster stage; (d) after about three-fourths of the petals are off and before the calyx lobes close ; (e) about ten days later; and (.f) in the summer at a time chosen for efficiency in codling moth control. Modifications of this program to meet varied conditions are discussed. During the six years of experimentation, "burning" of fruit which had been sprayed with lime-sulphur occurred in a significant amount only in 1921, and then was of minor importance on plots sprayed ac- cording to program. Considerable injury occurred, however, in cases where the treatment scheduled for ten days after petal-fall was de- layed a week or two. It is evident that there is danger of this type of injury in the event that hot, clear weather occurs during or soon after an application of lime-sulphur. The results of three years' comparative tests with spray rod and gun on trees of moderate size showed no significant differences. No significant increase in effectiveness of Bordeaux mixture or lime- sulphur attended the addition of g'-ue, gelatin, or casein-lime. Dusting experiments were conducted with the aim of testing the efficiency of several materials under varied conditions of application. In 1922 and 1923 all the materials used gave fair to good control of scab when the applications were made during major infection periods when the trees were wet. In 1924 this timing of applications gave less favorable results. These results suggest that in emergencies con- siderable benefit niay be derived from dust applications during infec- tion periods before the fungus becomes well established in the host. In 1923 and 1924, sulphur-arsenate dust applied on the same dates as the treatments in the spray program gave essentially as satisfactory results as liquid lime-sulphur. Th.e data on dusting reported in this paper are not considered conclusive. Experiments with sulphur and sulphur-arsenate dusts are being continued. 100 Wisconsin Research Bulletin J'S LITERATURE CITED Adams, J. F. 1925. The .spore- dischar^L- of tlie apple scab fuiif^us in Delaware. Del. Agr. Kxp. Sta. Bui. 140, lo p., illus. AlJERHOLI), R. 1894. Die J*erilhecieiiform voii ■•'usicladiuni (lendriliciini \\al. ( \'en- turia chlorospora 1. inali). Ber. Dent. iiot. desell. 12: 338- 342. 1896. Die Fusicladieii uiiserer (Jbstbaunie. I. Teil. Landw. Jahrb. 25: 875-914, illus. 1899. Ueber die Wirkungsweise der so-genannten Bordeauxbriihe CKupferkalkbnihe). Centbl. Bakt. [etc.J, 2 Abt. 5: 217-220, 254-271. 1900. Die Fusicladieii uiiserer Obstbiiume. II. Teil. Landw. Jahrb. 29: 541-588, illus. 1902. Kin Beitrag zur Frage der Empfanglichkeit der Apfelsorten fiir Fusicladium deiidriticum (Wallr.) Fuckel und deren Beziehungeu zuni Wetter. Arb. K. Gsndhtsamt., Biol. Abt. 2: 560-566. Barker, B. T. P., and Gimixgham, C T. 1911-12. The fungicidal action of Bordeaux mixtures. Jour. Agr. Sci. 4: 76-94. Beach, S. A., Lowe, V. H., and Stewart, F. C. 1899. Common diseases and insects injurious to fruits. N. Y. (Geneva) Agr. Exp. Sta. Bui. 170: 381-445. Bennett, C. W. 1923. Apple .scab and its control. Mich. Agr. Exp. Sta. Quart. Bui. 5: 130-134, illus. BiisGEN, AI. 1893. Ueber einige Eigeiischafteii der Keimlinge parasitischer Pilze. Bot. Ztg. 51 : 53-72, illus. Bl-AIR, W. S. 1917. Dominion experimental orchard work. Ann. Kept. I<"ruit Growers' Assoc. Nova Scotia. 53: 132-159. Bremer, 11. 1924. Das Auftreten der Schorfkrankheit am .Viifelbaum [Fusiclad- ium deiidriticuni (Wallr.) l-"uck.| in seiiien Beziehungen zum Wetter. Angew. Hot. d: 77-'>7, illus. Brown, W. 1922. Studies in the physiology of parasitism. IX. The effect on the germination of fungal spores of volatile substances aris- ing from plant tissues. Ann. Bot. 36: 285-300. Butler, O. 1925. Control of apple scab. N. H. Agr. Exp. Sta. Cir. 25, 8 p. 1.:i.ii)kmi()1.<»«;y and Control of Api'LE Scai 101 Vn7 New facts rc-garding the i)cTi..cl uf ascosporc discharge of the apple scab fungus. Ore. Agr. Kxp. Sta. Bui. 143, 11 p., dlu^. L'l.INTUN, Cj. V. „ , , ,, .,, lyUl. Apple .cab. HI. Agr. Exp. Sta. Hul. (./ : m-\,<>, illu.. CORDLEY, A. B. 1 t U 1 1908. The lime-sulphur spray as a preventive ul apple scab. Kurai New Yorker. 67: 202. CULLINAN, F. P., AND BaKER, C. E. 1924. Liquid lime sulphur versus .ulphur dust tor apple sprayuig. Ind. Agr. Exp. Sta. Bui. 283, 22 p., iHus. Curtis, K. M. tj -iic oiu 1921. Black-spot of apple and pear. N. Z. Jour. Agr. 23: 21b--18. 1922. Ascospore ejection of the apple and pear black-spot fungi. X. Z. Jour. Sci. and Tech. 5: 83-90. Darxell-Smith, G. p., and McKinxox, E. 1915. Fungus and other diseases of the apple and pear. X. S. Wales Dept. Agr., Farmers' Bui. 99, 45 p., illus. DicKSox, J. G. . ,. 1926. Making weather to order for the study of gram diseases. Wis. Agr. Exp. Sta. Bui. 379, 36 p., illus. UORAN, W'. L., AND OSMUN, A. V. 1924. Combating apple scab. Mass. Agr. Exp. Sta. Bui. 2\J, 17 p. DuTTox, \\. C. AND Johnston, S. . , ,n.i n,i n 1922. Dusting and spraying experiments of 1920 and 1921. Mien. Agr. Exp. Sta. Spec. Bui. 115, 14 p., illus. ^"^^911^' Die Empfanglichkeit der Apfelsorten fur Fusicladium dendriticum (Wallr. ) Fuck, und deren Beziehungeii zum Wetter auf Grund zehnjahriger Feststellungen. Jahresber. Konigel. I.eln-anst. 1. Obst- und (^artenbau zu Proskau. PHO : 104-113. Fairchild, D. G. _ » T-v- -vT-^^ 1894. Bordeaux mixture as a fungicide. U. S. Dept. Agr., Div. Veg. Path. Bui. 6, 55 p. Fischer, F. ^ 1909. Ueber die Bekampfung des Fusicladium. Ztschr. FHanzen- krank. P) : 432-434. Frank, B. , , , i .^ 1883. Ceber einige ncue und wenige bekannte PHanzenkrankheiten. Ber Deut. Bot. Gesell. 1 : 29-34. Frey, C. N. . . 1- //- Uo\ 19^4 The cvtologv and physiology of Ventuna maequahs (Cooke) Winter. Trans. Wis. Acad. Sci., Arts, and Letters. 21 : 303- 343, illus. AND KeITT, G. W. 1925 Studies of spore dissemination of Venturia maequahs (Lke..» W^int. in relation to seasonal development ot apple scab. Jour. Agr. Res. 30: 529-540, illus. 102 Wisconsin Research Bulletin 7v3 GiDDINGS, N. J. 1921. Orchard rlu-^tins; versus sprayinir. Jour. F.con. Ent. 14: 225- 230. Jackson, H. S. 1913. Diseases (if jj^macedus fruits. Ore. Agr. Exp. Sta. Bien. Crop. Pest aiKi Hurt. Ke])t. 1")11-12: 233-261, illus. AN'It WiN.STO.X, J. [\. 1 1915? I l-'.\])eriments for coiitrdl of apple seal). ( )re. Agr. Exp. Sta., Kept. IIu(h1 River Hrancli l^xp. .Sta. 1913-14: 9-lX. illus. | No date. I Johnson, J. 1921. The relation of air teni])erature to certain plant diseases. Phytopa.th. 11 : 446-45.S, illus. Keitt, G. \V. 1917. Peacli scab and its control. U. S. Dept. Agr. Bui. 395, 66 p., illus. 1918. Inoculation experiments with species of Coccomyces from stone fruits. Jour. Agr. Res. 13 : 539-569, illus. 1920. A preliminary report on apple scab and its control in Wiscon- sin. (Abstract) Phytopath. 10: 5.S. 1921. Second progress report on aii])K' scab and its control in Wis- consin. (Abstract) Phytopath. 11: 43-44. 1922. Third jirogress report on apple scab and its control in Wiscon- sin. (Abstract) Phytopath. 12: 54. 1923. Apple scab. Proc. Ohio Stale Hort. Soc. 56: 78-81. AND Jones, L. K. 1924 a. Seasonal development and control of apple scab and cherry leaf spot in relation to environment. (Abstract) Phytopath. 14: 36. 1924 b. Sepal infection in relation to the seasonal development and control of apple scab. (Abstract) Phytopath. 14: 36. 1925 a. Ere(|uencies of ascospores of Wnturia inaeiiualis in orchard air. (Abstract) Phytopath. 15: 57. 19251). I''urther studies of the seasonal development and control of a])ple scab and cherry leaf spot. (Abstract) Phytopath. 15: 57. 1926. Some relations of environment to the epidemiology and con- trol of apple scab. Proc. Nat. Acad. Sci. 12: 68-74, illus. AND Wilson, E. E. 1926. Studies of the development of the ascigerous stage of Ven- turia inaequalis in nature. (Abstract) Phytopath. 16: 77. l'J'II)i:.M lOl.OiiV AM) ("oNTKOr. OK Al'l'LE SCAB 103 K 1 1. I.I AN, K. l'M7. Ol)(.T (lit- ScxualitJit von Veiituria ina(|ualis CCookc) AH. Zcitschr. f. Hot. 9: 353-398, iUus. Kroi T. W. S. 1923. Cumhatinti ajipK- scab. Sprayin.i; and dustinj^ in 1922. Mass. Agr. K.xp. Sta. F>ull. 214: 29-41. 1924. Spraying and dusting for the control of apjilc >cal) in Massa- chusetts. Crop Protection Digest Bui. 1 (4): 30-36. L.AWRF.XtE. W. H. 1904. The apple scab in Western Washington. Wash. Agr. Exp. Sta. Bui. 64, 24 p., illus. Lowe, V. H. .'\nd Parrott, P. J. 1902. San Jose scale investigations. I\'. X. Y. (Geneva) Agr. Exp. Sta. Bui. 228: 391-456, illus. Ma.nxs. T. F. 1921. Report of fungus diseases for 1920. Dei. Bd. Agr. Bui. 10: 72-77. Morris, H. E. 1914. A contribution to our knowledge of apple scab. Mont. Agr. Exp. Sta. Bui. 96: 67-102, illus. Morse, W. J., and Darrow, W. H. 1913. Ts apple scab on young shoots a source of spring infection? Phytopath. 3: 265-269. 1916 a. Six years of experimental apple spraying at Highmoor Farm Me. Agr. Exp. Sta. Bui. 249: 81-96. 1916 b. Spraying experiments and apple disea.ses in 1915. Me. Agr. Exp. Sta. Bui. 252: 169-192, illus. 1918. App'e spraying experiments in 191^ and 1917. Me. Agr. Sta. Bui. 271 : 101-128. AND FOLSOM, D. l'J25. Apple spraying and dusting experiments 1918 to 1924. Me. Agr. Exp. Sta. Bui. 325: 125-184. Pastetr, L. 1862. Memoire sur les corpusculcs organizes qui existent dans Tatmosphere. Ann. Chim. et Phys. 64: 5-110. illus. Peace, L. M. 1910. Notes upon the clearing and staining of leaves and stems. Plant World 13: 93-96. Robinson, J. M. 1914. Spraving. Ann. Rept. Fruit drowers' Assoc. Xova Scotia. 50: 184-210. SCHNEIDERHAN. F. J. 1923. Scab and other things. Proc. Va. State Hort. Soc. Zl : 153- 162. [Discussion, p. 162-174.] 104 Wisconsin Research Bulletin 73 ANii From ME, F. D. 1924. Apple scab and its ciuitrnl in Virginia. Va. Agr. Fxp. Sta. Bui. 236, 29 p., illus. 1926. Apple disease studies in Nortliern X'irginia. Va. Agr. Exp. Sta. Bui. 245, 35 p., illus. SfOTT, W. M. AND (Jl^AINTANCK, A. L. 1907. Spraying for apple diseases and tiie Cddiing moth in the Ozarks. U. S. Dept. Agr. Farmers' Bui. 283, 42 p., illus. TiuRSTo.x, H. \\'., Jr., Walton, R. C. and Fagax. F. N. 1924. Comparison of materia's used in spra\ing and dusting for apple scab control in Pennsylvania. Pa. Agr. Exp. Sta. Bui. 190, 20 p.. illus. Trfj.rase, W. 1884. The apple scab and leaf blight. Ann. Kept. Wis. Agr. Exp. Sta. 1 : 45-56, illus. Vtx;ES, E. 1910. Die ljek;un])fung des Fusicladium. Ztschr. Pflanzenkrank. 20: 385-393. Wallace, E. 1913. Scab disease of apples. N. Y. (Cornell) Agr. Exp. Sta. Bui. 335: 543-624, illus. Walton, R. C. 1924. Dusting and spraying for disease control in Pennsylvania, 1922. Crop Protection Digest \'>n\. 1 (4): l')-28. WUKTZI-L, H. H. 1918. Apple scab and its control. Can. Horticulturist and Bee- keeper. 26: 121. 1921. Fruit diseases of the i)ast season. Proc. N. Y. State Hort. Soc. 66: 37-47. 1924. Apple scab and foliage injury. Proc. N. Y. State Flort. Soc. 69: 76-78. Williams, C. G. 1924. Report of the director. Ohio Agr. i'.xp. Sta. Bui. 382, 68 p., illu>. WlLTSHIRK. .^. P. 1915. Infection and immunilv stndit'N on the api)K' and pear scab fungi (J'cnhiria imwijindis and \'. piriuti). .Ann. Ai)pl. Biol. 1 : 335-350, illus. WiNSTOX. J. R. AND ChILDS, L. 1 1916? I The spraying experiments of rJ15 in the Hood River Valley for the control of apple scab. Ore. .'\gr. b'.xp. Sta.. Rept. Hood River Branch Exp. Sta. 1914-1«)I5: 30-40, illus. | No date.] YouNC, 11. C"., AND Walton. R. C 1925. Spray injury to app'e. i'li> loi)ath. 15: 405-415, illus. Research Bulletin 76 Februaiy, 1927 The Cleissification of Plant Viruses JAMES JOHNSON Agricultural Experiment Station of the University of Wisconsin Madison Contents Page Introfhiction 1 Material and Methods 2 Experimental Results 4 Description of the Viruses 9 The Isolation of Plant Viruses 12 Discussion 13 Summary 15 Literature cited 15 J 7^ \ ^ The Classification of Plant Viruses' ^ r James Johnson THE CAUSAL AGENCY of mosaic and related plant diseases, though unknown, is generally regarded as a filterable virus, there- by suggesting an invisible and infectious entity. The results of certain investigators indicate that only one virus is concerned with the disease on a number of widely different hosts, while the results of others suggest that a virus is plastic, the symptoms, host range, and properties being modified by existing conditions. On the other hand, the work of several investigators has shown that a single host may be afifected by a number of different virus diseases and that a number of plant species are affected with apparently specific viruses. The evidence to date on the spe- cificity of the viruses affecting a single host is based largely on symptoms and on differential hosts. In the case of the viruses affecting tobacco this evidence can, in most cases, be supplemented by differential properties of the viruses as will be shown in this bulletin. While it is admitted that a single virus may affect a wide variety of un- related hosts, and that a virus may be changed in virulence or have its properties measurably influenced by conditions, the fact seems to remain that specific viruses exist in nature with approximately the same constancy as is, for instance, exhibited by the bacteria. The failure to recognize the specificity of certain of these virus diseases has resulted in much confusion in the literature. It seems to be important, therefore, to describe adequately and to classify the virus diseases of plants as far as our knowledge of each group will permit. A system of nomenclature for "plant viruses is greatly needed. The present system of applying names on the basis of host attacked, or symptoms exhibited, is quite inadequate for present needs. The number of different viruses attacking a single host species limits at once the usefulness of host names. Nowhere in the realm of plant pathology are symptoms of less value in description than in the plant virus diseases, because of the re- markable influence of environmental factors (PI. lA.), and the possible co-existence of two or more viruses in a single plant. Symptoms have, however, some valuable diagnostic features when properly interpreted in comparative studies (PI. I, B, C). We have found it very difficult to apply a different descriptive name to all the viruses occurring in one host, and this seems to be a good reason for the use of an arbitrary system of nomenclature. In the development of such a system the established 'Cooperative experiments with the ofBce of Tobacco Investigations, Bureau of Plant In- dustry, United States Department of Agriculture. 2 Wisconsin Research Bulletin 76 term already applied to any virus disease should, however, be left for what it is ; i. e., a name for the disease, but not for the specific causal agency. An important objection may be raised to the proposal to name the causal agents in the absence of any known or visible entity. It seems most likely, however, that no visible causal agency will be found to be as- sociated with these diseases and that the workers with viruses must be content to follow the example of other branches of science where names have long been applied to definite but unseen entities. Material and Methods This bulletin will deal only with certain viruses of which tobacco and other solanaceous plants are hosts. The first of these viruses is the causal agent of ordinary tobacco mosaic. The writer has previously described two viruses affecting tobacco which were secured from apparently healthy potatoes, namely those of spot-necrosis and ring-spot (4). Recently (5) we have shown by means of differential hosts that four other virus diseases affect tobacco, namely cucumber mosaic, "speckled" tobacco mosaic, "mild" tobacco mosaic, and petunia mosaic. This bulletin will add to the list four more diseases which will be referred to as yellow tobacco mosaic, medium tobacco mosaic, bleaching tobacco mosaic, and tomato stem-necrosis, al- though it must be recognized that these terms are descriptive of the diseases in question only under special conditions. Petunia mosaic, previously des- cribed (5) was not available for property studies. 'i he separation and rlass'l^cation of ttie viruses described is based on the symptoms, if any, produced on ten or more different species of hosts, their longevity in- vitro, thermal death-points, lethal effect of chemicals, and such minor differences as relative infectivity and length of incuba- tion period. The inoculations to tobacco and other hosts have been made by in- troducing the extracted juice from mosaic plants into healthy plants by punctures and scratches with a needle, wrapped near the point with a small quantity of absorbent cotton to carry the extract more readily. Five plants were used in each series of inoculations. These plants were grown in four-inch pots, in very fertile soil under warm greenhouse conditions (27-32° C.) The inoculations were made on the plants when very young. The inoculation of old plants or young stunted plants will usually lead to symptoms with some viruses but not with others, so that such results would be of questionable value. The longevity in vitro was determined by merely aging the extracted viruses from the tobacco plants in stoppered test tubes without preserva- tives for the desired length of time. The thermal death-points were de- termined by immersing five cubic centimeters of the newly extracted virus from tobacco plants in thin-walled test-tubes into an automatically regulat- ed and strongly agitated water bath. The time of exposure was ten minutes in all cases, including the time between actual immersion and removal from the bath. Thermometer readings showed that the temperature of the The Classification- of Plant Viruses 3 extract jumped rapidly to within five degrees of the desired temperature but that it usually required two to three minutes from the time of immer- sion before it reached the desired temperature. Determinations were usual- ly made only at five degree intervals between 55 ^C and 95 ^C. Certain circumstances will vary the thermal death-points appreciably and they cannot ordinarily be regarded as significant for classification purposes unless differing by more than 5°C, even when the material is carefully chosen. The effect of chemicals on the virus was determined by exposing the virus, diluted only one-half, to certain concentrations of the chemicals used for a definite length of time, usually one hour and one day. This promis- ing field for the separation of viruses has been barely opened up in our investigations, and only the tests with alcohol and nitric acid for this purpose will be reported. As a result of Allard's work (1) and our own with other chemicals we are convinced that the lethal action of various chemicals on the different viruses will form a valuable means of classifica- tion and separation of viruses. The incubation period for any one virus naturally varies greatly, de- pending upon the the condition of the host and other environmental fac- tors. The first symptoms of tobacco mosaic may be evident under exception- ally favorable conditions three days after inoculation, whereas, under un- favorable conditions, ten or more days may elapse before symptoms develop. In our experiments symptoms were usually pronounced after five or six days. When the different viruses are compared under similar conditions on the same host, symptoms characteristically follow in a fairly definite chron- ological order ; for example, cucumber mosaic Tjsually requires considerably longer to develop symptoms on tobacco than does tobacco mosaic on tobacco. The percentage of infection secured with the different viruses varies considerably, particularly on some hosts. With some viruses, one hundred per cent infection is the rule, whereas, with other viruses, only forty or sixty per cent infection is characteristic. The conditions influencing this relation are as yet inadequately understood. We are convinced, however, that at least two factors are concerned, namely, the relative susceptibility of the host, and the source of the inoculum. A'', tabacum is, for example, much more susceptible to cucumber mosaic than is .V. glaiica. when the same inoculum is used on both species. On the other hand, cucumber mosaic from tobacco is a very good source of inoculum, whereas, cucumber mosaic from pokeweed, potatoes, and certain other plants, is a very poor source of in- oculum. Comparing good and poor sources of inoculum in one group of experiments with cucumber mosaic, the records show 94 per cent infection secured in the former case as compared with only 7 per cent in the latter case. Results of this nature led us to doubt at one time that we were securing systemic infection with a virus disease even though marked symptoms occurred, since the virus could not be recovered from the in- oculated hosts (5). We have since come to the conclusion that when symptoms are definitely secured, the species should be regarded as a host. 4 Wiscoxsix Research Bulletix 76 even though the virus cannot be again secured therefrom. Good examples of this sort are tobacco mosaic on A'^. glutinosa and several of the viruses occurring on tobacco when transmitted to the potato. It is evident that while a host may be poor source of inoculum for one virus disease, it may be a good source for another. In cross inoculation work one should, there- fore, make certain that a suitable source of inoculum is being used in cases of negative results. Experimental Results Seven of the viruses with which we have been concerned have been pre- viously described as to their source and behavior on dififerential hosts in particular (4, 5). Since these publications, four other viruses have been collected and studied along with the others in comparative trials. Two of these four viruses (yellow tobacco mosaic and medium tobacco mosaic) were secured from tobacco fields on farms near Madison, the source of the others (bleaching mosaic and tomato stem necrosis), is not definitely knoAvn, except that they arose during efforts in the experimental work to segregate the various viruses with which we were working. These viruses are described in comparison with the others in the tables and in a later chapter. It cannot be argued that some of these viruses are not closely related, and in view of recent results in which we have secured decided attenuation of some of these viruses by exposure to heat (6). it is not inconceivable that they are modified by other conditions as well. Such a behavior is characteristic of many organisms and until more is known about changes in virulence, mutations, etc., this phase of the subject will remain obscure. This does not, however, obviate the needs of a system of classification, but rather adds to the importance of it. For present pur- poses, it seems advisable to give all forms of viruses equal rank in classi- fication, although the need for grouping them into related classes seems already evident. The separation of the viruses with which we have worked is based up- on the results secured in the inoculation of several thousand plants. The data in its entirety would to too voluminous and confusing to present in detail. It is, therefore, only summarized in Table I. with relation to the differential hosts. The larger number of inoculations to tobacco 's significant only for the purpose of showing the number of times the viruses have been transferred without losing their identity. The results only are presented in the cases of aging in vitro, and thermal death-points (Table II.). In the case of the lethal influence of chemicals, the result of only one series of tests is given as an illustration (Table III.). The symptoms given in Table I, as typical on the different hosts, are in a sense only relative. The natural variations and modifications due to environment and other circumstances are so great as to render them in many cases of .little value, except in repeated comparative studies. Certain of the hosts, however, give remarkably specific symptoms for a specific virus, and these are the ones which should be selected for purpose of determina- The Classification of Plant Viruses 5 tions, and are conse(|uently indicated in tlie description nf the viruses as being particularly significant. The proper comparative host studies are sufficient in most cases to determine the viruses described in this paper, and in some cases prove the cn'.y available method. In other cases, de- terminations of the longevity of the virus in vitro, the thermal death-point, and the influence of chemicals may be necessary or will lend valuable ad- ditional information as to the certainty of the determination. The hosts shown in Table 1 represent those mojt commonly used in the experiments. The results of the previous season's work (5) are added to those secured as a basis for the present report. Solanutn nigrum and Solanum inelongeua are, however, dropped from the list of differential hosts since they are not well suited for this purpose and henbane and potato (Bliss Triumph) substituted. Symptoms with one or more of these viruses have been secured, however, upon a number of other hosts in the course of the work. Most of the solanaceous plants seem to be susceptible to one or more of the viruses described. Belladonna, S. dulcamara and Lycium seem to be the most "resistant" to virus diseases of all solanaceous plants tried in the experiments. We have secured symptoms on Belladonna with one virus, the classification of which is, however, uncertain. PhysaUs alkekengi and PhysaUs franchetfi have given fairly good symptoms of tobacco mosaic in some cases and these species cannot be regarded as symptomless carriers (2, 7). Solanutn atropurpurcum has been infected with all the viruses except ring-spot, the latter inoculation not being tried. Sdatntiii lacinatum gave symptoms with tobacco mosaic, cucumber mosaic, and spot necrosis. Tobacco mosaic has also given symptoms on Solanum miniatum. S. rostra- ium, S. sisymhrifoliutn. and N^icandra pliysaloidcs. In addition to the names which we have applied to the specific virus diseases, we are tentatively applying a name to the specific causal agency of each disease, with the expectation that it may be more useful in future classifications than in the present one. The suggestion is simply to name a virus from the host on which it is first discovered, together with a number to indicate the specific virus in question. The well known tobacco mosaic virus under this scheme is to be technically known as Tobacco virus 1, and the other viruses described in this paper, originally noted on tobacco become simply Tobacco z'irus 2. 3, 4. 5. 6, 7, S. and 9.^ The cucumber mosaic virus, though it occurs on tobacco, should remain Cucumber virus 1. since it was orginally found and described on cucumber. Question of prior- ity in the naming of or numbering of viruses, will natural'y arise, and it has been suggested that this should be regulated by some committee selected for this purpose. It is not believed, however, that simultaneous descriptions and questions of priority will greatly interfere with the prac- tical application of such a system of nomenclature. Should this suggestion meet with ?nv approval among the workers with plant viruses, the details of its anilication should be m ire adequately presented than can be done in this paper. Un naming the virus, it may be preferable in some or all cases to use the T;>''n gcnerU- name cv binomial in place of the common name of the host plant. Wisconsin Research Bulletin 76 1 1 t. ; I' 5 11 CO 1 ^ oc 1 2 2 o lO 1 !C m 1 2 oo 1 in O 1 00 m 1 ic o 1 «^ o 1 •> o ffO o 1 o O 1 o o o m 1 o o 1 CO o 1 CO m cc O 1 o 0 0 C-) C^l •— ' (N 1 1 IN 1 ^ -1 1 CO 1 i5 s ~ £ o c "< m e S E £ £ £ E *» C VI ^ fel2 o OU5 t^ o 1 fc O 1 00 in .r-« m 1 o in 1 M O 1 •<*< oO o ^ IB to in ^ 1 '^ rH 1 rt Z"' Tj" Z m"'^ ' '^ "-^ m 1 -^ 00 S £ e" E e" £ § i e" c E (» D ci U > S« ^ P~, «„ o o o o o o ,0 iT S^ H ^ fc" g^ gin VJCD ^^ 8oo §== s: 0 2 a 2 o 2 e » e » e .^■b .*.fc o> »> O »^ o t« O K. ^ I. E-c S o s fri 5 ^. S t-i S f-, S E«. S t,s ^■s ^'5 0.-S U M -s: u w u 3 £.2 ■a ^ g o C oj S III 8« '» ^8o is.2 2 o O e 0) §'3 HS 3 S Mild tobac mosai So I§1 i-c o S2E ffl £ ££§ SI The Classitication of Plant Viruses 21 "2 - o s a 5 6S Hi °-5 - I -H CI M : -H 4, — S 3 p-a; r •a o 3-2 "= 3 ucn as « O — «5 a. CO -H I - 2B = ■5 3 g Wisconsin Research Bulletin 76 TABLE II.— resistance TO AGING, IX VITRO, AND THERMAL DEATH- POINTS OF VARIOUS VIRUSES ON TOBACCO Virus disease Resistance to aging in vilro Thermal death- point ° C 10 inin. Tobacco mosaic Cucumber mosaic Speckled mosaic Mild mosaic Spot necrosis Ring spot Yellow tobacco mosaic. Medium tobacco mosaic Bleaching mosaic Tomato stem necrosis. . 2 years + 3 days — 3 mouths 4- 6 days — 14 days — 14 days — 3 months -|- 3 months -j- 3 days — 3 months -f- 90 60 90 60 70 70 90 90 75 90 TABLE III.— RESISTANCE OF VIRUSES TO THE LETHAL ACTION OF ALCOHOL AND NITRIC ACID. (FIGURES INDICATE NUMBER OF PLANTS INFECTED OUT OF FIVE INOCULATED.) Virus disease Alcohol, 50 per cent Treated 1 hour 1 day Nitric acid, 1 to 200 Treated 1 hour 1 day Tobacco mosaic Cucuml>er mosaic Speckled tobacco mosaic Mild mosaic Spot necrosis Ring spot Yellow tobacco mosaic. . Medium tobacco mosaic. Bleaching mosaic Tomato stem necrosis — The Classification of Plant Viruses Description of the Viruses The following abbreviated descriptions of the viruses with which this bulletin is concerned are presented below, with the hope that they may serve as a tentative basis for classification. It is to be expected that future investigations will necessitate additional details and modifications of the present descriptions. Other valuable diagnostic features are already known to exist, such as the variations in the cytological details of infected tissues, the details of which are now being worked out by Miss Isme A. Hoggan in this laboratory. Tobacco Mosaic {Tobacco virus 1.) PI. I, A. Type. Allard, U. S. D. A. Bul. 20, 1914. Host Family. Solanaceae. Differential hosts. On tobacco, marked mottling, malformation and stunting. On N. glutinosa, stem and leaf necrosis and stunting, no mottling. On tomato, mottling and stunting, no stem necrosis. On pokeweed, no symptoms. Resistance to aging in vitro. Several years. Thermal death-point. 90° C. 10 minutes. Resistance to chemicals. High (60% alcohol or 1 to 200 HNO^,' does not kill in one day). Cucumber Mosaic {Cucumber vims 1.) PI. HI, A and PI. VH, B. Type. Doolittle, U. S. D. A. Bul. 879, 1920. Host Families. Cucurbitaceae, Solanaceae, and others. Differential hosts. On cucumber, chlorosis, mottling, stunting, malformation, necrosis. On iV. glutinosa, mottling, malformation, stunting. On pokeweed, mottling, stunting. On tobacco, chlorosis, generally no malformation. Resistance to aging in vitro. 3 days or less. Thermal death-point. 60-70° C. 10 minutes. Resistance to chemicals. Low, 50% alcohol or 1 to 200 HNO ,, kills i-n one hour). Speckled Tobacco Mosaic {Tobacco virus 2.) Type. Johnson, Phytopath. 16: 141. 1926. Host Family. Solanaceae. Differential hosts. On tobacco, mottling or speckling. On petunia, mottling, stunting, malformation and necrosis. On henbane, chlorosis, stunting and malformation. No symptoms on N. glutinosa. pepper or pokeweed. Resistance to aging iu vitro. 3 or more months. Thermal death-point. 90° C. 10 minutes. »One part nitric acid C. P. to two hundred parts water. 10 Wisconsin Research Bulletin 76 Resistance to chemicals. Medium (withstands 50% alcohol 1 day^ but is killed by 1 to 200 HNO^ in 1 hour). Mild Tobacco Mosaic (Tobacco virus 3.) Type. Johnson, Phytopath: 16: 141, 1926. Host Family. Solanaceae. Differential hosts. On Physalis pubesccns. Marked stunting and chlorosis. On A^. gliUinosa, mild chlorosis and stunting. On A'^. rustica, mild mottling and malformation. No symptoms on N. glauca or pokeweed. Resistance to aging in vitro. About 6 days. Thermal death-point. 60° C. 10 minutes. Resistance to chemicals. Low (50% alcohol or 1 to 200 HNO^ kills in one hour). Spot-Necrosis of Tobacco {Tobacco virus 4.) PL I, C. Type. Johnson. Wis. Agr. Exp. Sta. Research Bui. 62>, 1925. Host Family. Solanaceae. Differential hosts. On tobacco, mild mottling and necrotic spots. On potato, virulent form produces basal leaf necrosis and curling on top leaves. Resistance to aging in vitro. About 14 days. Thermal death-point. 70° C. 10 minutes. Resistance to chemicals. Low (50% alcohol or 1 to 200 HNO^ kills in one hour). Ring-Spot of Tobacco {Tobacco z'irus 5.) PI. I, B. Type. Johnson, Wis. Agr. Exp. Sta. Research Bui. 63, 1925. Host Family. Solanaceae. Differential hosts. On tobacco, mottling and ring-like spots. On henbane, necrotic leaf spots. Resistance to aging in vitro. About 14 days. Thermal death-point. 70° C. 10 minutes. Resistance to chemicals, Low (50% alcohol or 1 to 200 HNO3 ^'''^ in one hour). Yellow Tobacco Mosaic {Tobacco virus 6.) PI. H. Type. This publication. Host Family. Solanaceae. Differential hosts. On tobacco, mottling and angular yellow chlorotic areas. On A'^. glauca, round white chlorotic areas. On petunia, irregular white or yellow chlorotic areas. The chlorotic areas apparently develop only under special condi- tions and are imperfectly understood, therefore negative re- sults are not always reliable. The Classification of Plant Viruses 11 Resistance to aging in vitro. Three or more months. Thermal death-point. 90° C. 10 minutes. Resistance to chemicals. High (60% alcohol or 1 to 200 HNO^j does not kill in one clay). Medium Tobacco Mosaic (Tobacco virus 7.) Type. This publication. Host Family. Solanaceae. Differential hosts. On tobacco, medium mottling and stunting. On A^ glauca, mild mottling. On petunia, mild mottling. \'ery similar to yellow tobacco mosaic except that yellow areas are not produced. Resistance to aging in vitro. Three months or more. Thermal death-point. 90*^0. 10 minutes. Resistance to chemicals. High (60% alcohol or 1 to 200 HNOg does not kill in one day). Bleaching Mosaic (Tobacco virus 8.) PI. HI C and PI. V. B, E. Type. This publication. Host Families. Solanaceae, Phytolaccaceae. Differential hosts. On tobacco, mottling and sometimes chlorosis on young plants. On iV. giutinosa. mottling and malformation. On pokeweed, mottling and occasionally necrosis. On Physalis pubcscens, chlorosis, necrosis and mottling. Very mild, if any symptoms on pepper. Resistance to Aging in vitro. About three days. Thermal death-point. 75° C. 10 minutes. Resistance to chemicals. Medium. (Withstands 50% alcohol but is destroyed by 1 to 200 HNO, in one day). Tomato Stem-Necrosis (Tobacco virus P.) PI. IV. Type. This publication. Host Family. Solanaceae. Differential hosts. On tobacco, mild mottling. On young tomato stems, necrosis. On Physalis pubcsceus. very mild if any symptoms. No symptoms on A'', glauca or pokeweed. Resistance to aging in zntro. Three or more months. Thermal death-point. 90° C. 10 minutes. Resistance to chemicals. Medium. (Withstands 50% alcohol, but is destroyed by 1 to 200 HNO^ in one day). 12 Wisconsin Research Bulletin 76 The Isolation of Plant Viruses The co-existence of two or more viruses in a single plant is not an un- usual occurrence. In the case of the viruses described in this bulletin it is frequently possible to separate such combinations, though it is not possible in several other cases with the methods at present available. It is believed, however, that the possibilities in this direction are especially promising along the line of the application of the lethal or inactivating action of chemicals. The present methods of isolation are, of course, based on the host range and the properties of the various viruses. In order to test the possibility of the separation of certain viruses, combinations of known viruses were made in test-tubes or on plants and subsequently isolated in "pure culture" by such methods. As an illustration, the case of a combination of the ordinary tobacco mosaic virus and the cucumber mosaic virus may be cited. This separation was accomplished by inoculating the mixed viruses to poke- weed. Since pokeweed is susceptible to cucumber mosaic and not to ordinary tobacco mosaic, the cucumber virus was secured free from the tobacco virus. Another portion of the combination extract was * aged for four days and then inoculated to tobacco. Since aging in vitro for three days or more destroys the cucumber mosaic virus, the inoculation to tobacco resulted in ordinary tobacco mosaic alone. The latter separation could be ac- complished in other ways, as for instance, by the use of heat or chemicals destructive to the cucumber mosaic virus but not to the tobacco mosaic virus. A combination of three viruses may in certain cases be separated. If, for instance, speckled tobacco mosaic, cucumber mosaic, and mild tobacco mosaic are combined, the isolation may be accomplished by heating a portion of the extract to about 75 °C. for ten minutes to destroy the in- fectivity of the two latter viruses. Another portion of the extract may be aged for three days, destroying the cucumber virus, followed by inoculation to A^. glutinosa plants, which will become infected with the mild mosaic virus but not the speckled tobacco mosaic virus. Finally, a third untreated portion on inoculation to pokeweed should yield cucumber mosaic alone. In this manner it seems possible to separate combinations of even four or five viruses, but, of course, with greater difficulty and uncertainty of positive results. However, the separation of more than two viruses is rarely of importance in the present state of our knowledge of virus diseases. In the case of the virus diseases of the potato, the development of some such methods of isolation would be of the greatest value at present, but unfortunately several difficulties would present themselves according to our preliminary attempts in that direction. In the first place, the potato viruses are quite generally limited, so far as we know, to the potato as a host, and the absence of a wide host range naturally reduces the possibilities of isolation. Secondly, the potato mosaics, at least, are re- latively short lived in extract outside the host, reducing in considerable measure the ease with which they can be handled and isolated on the basis The Classification op" Pt.ant Viruses 13 of their comparative properties. When the properties of the various potato viruses are known in more detail, however, it is quite likely that certain combinations may be separated as readily as "spot necrosis" can now be separated from rugose mosaic of the potato, since the former and not the latter is transmissible to tobacco according to our experience. Discussion The evidence for the existence of several specific viruses affecting tobacco has been practically limited to symptomatology on differential hosts in previous contributions on this subject. In this bulletin it has been shown that certain properties of these viruses offer in most cases a sufficient basis for their separation. In the present unsettled state of the plant virus problem there are those who evidently believe that all mosaic diseases of a certain host or group of hosts are caused by a single virus, or on the other hand, that a single virus is capable of such modification in its hosts and properties as to develop forms of a nature which are described as specific viruses in this paper. Walker (9), in a comparative study of the mosaic diseases of cucumber, tomato and Physalis, concludes "that the properties of the mosaic virus of a given plant maj-'be decidedly changed by transferring it to another host. The properties of the viruses from mosaic plants of a certain species also appear to be the same, no matter what source of infection. This fact indicates that there may be a single causal agent for all the mosaic diseases studied here", (namely, cucumber, tomato, and Physalis mosaic). Our results are entirely con- trary to this suggestion. We believe the results obtained by Walker could only be obtained by w^orking wnth mixed viruses in certain experiments. At the time Walker's experiments were conducted, the difference between cucumber mosaic and tobacco mosaic on solanaceous hosts w^as not clearly recognized. In our experiments, we have carried cucumber mosaic serially through a dozen different plant species, over a period of two years, but as far as our observation goes, we have not changed its behavior or properties in any respect. The question of virus specificity in another group is best illustrated bv the potato degeneration diseases which have concerned the workers on this subject for many years. While much skepticism still exists among the workers themselves as to the possible synonomy of some of the diseases described, there is no doubt whatever of the occurrence of several different virus diseases of the potato in nature, and according to Schultz and Folsom (8), there are at least seven. The potato virus diseases have been described entirely on the basis of symptoms, on one or another variety of potatoes. The host range of these potato viruses is apparently restricted to the potato, and it still remains to be determined whether or not the properties of -these viruses can be used as a basis of separation. An interesting feature of the present studies is the artificial infection of the potato with all the viruses described in this paper, with the exception of ring-spot. It is not likely, however, that these viruses are responsible for any potato diseases in nature. The fact 14 Wisconsin Research Bulletin 76 that they are not readily if at all transmissible from the potato back to other solanaceous species, however, renders this point difficult to determine. Fernovv (3) has recently separated a group of viruses on the basis of differential hosts, but he apparently did not concern himself with then- identity in relation to the known or unknown virus diseases. The probability of a combination of two or more viruses being mistaken for one specific virus always exists. This possibility has been considered from various angles in the separation of the viruses described in this paper, and we believe this possibility has been largely eliminated. That certain of the viruses are so closely related, however, that separation may not be justified in some cases, is granted. In the cases of known attenuated viruses, we believe, of course, that they should retain a designation in- dicative of their origin. Aledium tobacco mosaic may be regarded as an attenuated form of tobacco mosaic, but in the absence of proof of this fact, we believe it should be placed in a class by itself. Yellow tobacco mosaic re- sembles medium tobacco mosaic as far as virulence is concerned, but it possesses in addition a yellow character, which may or may not be de- pendent upon a separate virus. In the case of this disease, the classifica- tion is dependent entirely upon the yellow symptoms. The occurrence of decided yellow symptoms is very commonly associated with tobacco mosaic in the field, especially on the sucker growth following harvest of the crop. We are inclined to believe, however, that in this case, the yellow symptoms are usually color modifications, brought on by the existing environmental conditions, and that our yellow mosaic is quite different, and that its ex- pression is determined by a special condition or conditions yet imperfectly understood. The significance of the present investigations does not lie in the eco- nomic importance of the viruses described, since perhaps only the tobacco mosaic and cucumber mosaic may be said to be of much concern in this respect. Such importance as they may have lies ratlier in the demonstra- tion that a number of different viruses exists in nature, and that they can be classified on the basis of the symptoms produced on various hosts and on the basis of their properties. Whether or not we wish to regard some of these viruses as closely related, and to be actually strains or forms of other viruses, is not particularly pertinent in the present stage of our knowledge of the subject. It is evident that workers with plant viruses who are concerned with studies on the nature and properties of viruses should be certain of the strain with which thej' are working, and that those who are reporting a virus on a new or old host should describe the virus sufficiently to indicate whether or not the causal agency has previously been reported. The suggested classification of the group of viruses which may affect tobacco and related hosts is of course only a fraction of the entire problem. A great deal of investigation into the host range and properties of the virus diseases of other plant families is necessary before a satisfactory system of classification can be established. In the meantime it may be of help to speak of a virus disease as due to a specific virus, and to designate this virus by the name of the host The Classification of Plant Viruses 15 on which it was first described together with an arbitrary number. Thus Physahs mosaic may be written Physalis mosaic (Cucumber virus 1), Physalis mosaic (Tobacco virus 3), or in other ways depending upon which, if any, of the previously described viruses is determined to be the one concerned. Such a system should eventually aid greatly in reducing the confusion in plant virus literature. Summary The existence of eleven different viruses on tobacco and related plants is shown on the basis of their behavior to various factors applicable as tests. The most useful of these tests are the symptomatology on differential hosts, the longevity in vitro, the thermal death-points, and the lethal action of chemicals. It is suggested that these factors form a basis for the description of a virus, and that some form of classification and nomenclature be established for plant viruses. On the basis of the behavior of plant viruses to various conditions, it is possible in many instances to separate two or more viruses, where they co-exist in a single plant. While it is possible to attenuate and to increase the virulence of some plant viruses, the experimental evidence indicates that they are relatively stable and specific entities. 16 Wisconsin Research Bulletin Id Literature Cited (1) Allard, H. a. Effects of various salts, acids, germicides, etc., upon the infectivit}' of the virus causing the mosaic disease of tobacco. Jour. Agr. Research 13: 619-637. 1918. (2) Elmer, O. H. Transmissibility and pathological effects of the mosaic disease. Iowa Agr. Exp. Sta. Research Bui. 82, pp. 39-91. 1925. (3) Fernow, K. H. Interspecific transmission of mosaic diseases of plants. Cornell Univ. Agr. Exp. Sta. Memoir 96, pp. 3-34. 1925. (4) Johnson, J. Transmission of viruses from apparently healthy potatoes. Wis. Agr. Exp. Sta. Research Bui. 63. pp 1-12, 1925. (5) Johnson, J. Mosaic diseases on differential hosts. Phytopath 16: 141-149, 1926. (6) Johnson, J. The attenuation of plant viruses, and the inactivating influence of oxygen. Science 64: (No. 1652) 210,. 1926. (7) Nishimura, M. a carrier of the mosaic disease. Bui. Torrey Bot. Club. 45: 219-233, 1918. (8) ScHULTZ, E. S. and Folsom, Donald. Infection and dissemination experiments with degeneration diseases of potatoes. Observations in 1923. Jour. Agr. Research 30 : 493-528. 1925. (9) Walker, M. N. A comparative study of the mosaic diseases of cucumber, tomato and Physalis. Ph3'topath. 16: 431-457. 1926. ^ PLATE I. I'ai-iahilitx in Syiiipioins of I'ints Diseases A The svinpt..ins ol'a.iy virus disease may vary remarkably as illustrated hv tlu tendeiuv of this tobacco plant to recover from the mallormed condi- Uon of the lea?es Symptoms are^ theref..re, often of little diagnostic value m indicating the specific virus conctMiied. , . , , ,,.„ «xn,r,ionis irp how- B. Ill the case of certain viruses on certain hosts, the s>niptoms aie now eve-, quite specilic as illustrated by the "ring-spof ^^^^;-J'\'r^[;^; ,„„. (■ ''Soot necrosis" symptoms are often quite specific, but the mius con ceined nmy vary in virulence so that this introduces an additional complica- tion. Pi.ATi: II. ')'('ll<>:^' Tohiicri) Mosiiic Vpjier- YclJDW lohticio iiMisaic is cIkii acliri/cd liy tlic developnienl of irregular yellow areas in liir leal in addilicui Id a iiKilllin;; eharaelerislie ol' niediuiii lobaceo mosaic. LoiiU'r—Tbc vcliow iliaiailei- ol' yellow lohaeeo mosaic may develop on dilTerent liosls as sliow ii itotiana gluitca usually gives a low pei'i'i'ntaj;e of infection, jjut when symptoms are oljtained they are fairly distinctive. (A.) Cucumlier mosaic. (B.) Yellow tobacco mosaic. (C.) Bleaching mosaic, (D.) Tobacco mosaic. (E.) Control. PLATE IV. Toiiialo Slciii Xccrusis When applied In \hr sicnis (iC vdiiiiK lonialo plants the virus causing this • u o cs 'si ^'5 ^ e.^ ^ 73 ~-S = _ — O V - ? 3 i:u to •an c Oj o '--•~ £ a fc .-^ ■^"^'3 o' 3.2 "^ 2 •- -5 sp ^ fa ^ =_ i ^ .11 o . . i? •^ ;* — 3 -; — ;; o Research Bulletin 82 February, 1928 Spraying Versus Dusting To Control the Potato Leaf hopper in Commercial Potato Fields o f Wisconsin John E. Dudley, Jr., and C. L. Fluke, Jr. (Co-authorship) Agric ultural Experiment Station of the University of Wisconsin Madison CONTENTS Introduction 1 Distribution and Importance of the Potato Leafhopper 2 Plans of Experiments 2 Experiments at Spooner 5 Experiments at Waupaca 6 Results of 1922 - 6 Results of 1923 7 Results of 1924 8 Results of 1925 8 Bordeaux Mixture Increases Percentage of No. 1 Potatoes-10 Comparative Yields in Severe and Light Hopperburn Years 10 Relative Costs of Spraying and Dusting 13 Discussion 13 Summary 15 Literature Cited 16 Spraying Versus Dusting To Control the Potato Leafhopper in Commercial Potato Fields o f Wisconsin John E. Dudley, Jr.,* and C. L. Fluke, Jr. (Co-anthorship)^ THREE YEARS of experiments in Wisconsin on a small scale, from 1919 to 1921, showed that the potato leafhopper^ can be con- trolled by liquid Bordeaux mixture (3, 4, and S).^ When the agitation for dusting began, commercial experiments were planned to deter- mine the relative effectiveness of liquid Bordeaux mixture and copper-lime dust upon the leafhopper and hopperburn ; upon the resulting yield and grade of potatoes ; and to compare the cost of each operation when the work was actually done under farming conditions. These experiments extended over a period of four years, from 1922 to 1925. It has already been shown by Chambers (2) in Wisconsin, Stewart and Parrott (7) in New York, and others, that either spraying or dusting will increase the yield of potatoes considerably above that of untreated acreages. Most of the results indicate that spraying will increase yields to a greater extent than dusting (1, 6, 7). However, very little attention has been given heretofore to the ease and speed with which dusting can be done as com- pared with spraying. Boyd (1) studied the efficiency of copper dusts and sprays in the control of diseases and certain insects, principally fleabeetles. Folsom and Bonde (6) were experimenting largely with diseases, and they state that the leafhopper was not a factor. Stewart and Parrott (7) in an experiment with diseases and insects in general, used a hand duster to apply the dusts and an orchard type power sprayer to apply the liquid Bordeaux mixture. They state in eflFect that the dust did not give satisfactory control in their experiment, while the spray showed high efficiency. They observe, however, that dust may be advisable tmder certain conditions, where water is difficult to obtain and in small fields where it is necessary to use hand machines. Chambers (2) secured higher yields with copper-lime dust than with liquid Bordeaux mixture while carrying on demonstration experiments in potato fields in northern Wisconsin. Associate Entomologist. Bureau of Entomology. U. S. Department of Agriculture. nVe wish to acknowledge the courtesy of J. G. Milward of the Department of Horti- culture, University of Wisconsin, who arranged for the co-operative experiment at Spooner, Wisconsin. Also to express our appreciation to E. M. Searls and T. E. Bronson of the Govern- ment force who gave much of tlieir time to assisting with the experiments. We wish to give special credit to C. J. Schrock, "manager of William Gulp Farm upon whose land the experiments at Waupaca were carried on for four years. Both Mr. Gulp and Mr. Schrock showed unusual appreciation for research work and ofiFered the utmost co-operation in carrying on experiments and taking yields. -Empoasca fabae Harris, order Hemiptera. family Gicadellidae. 'Reference is made by number (italic) to "Literature cited," p. 16. 2 Wisconsin Research Bulletin 82 It is not intended in this bulletin to review all the literature on the subject as that has been excellently handled by Folsom and Bonde (6). A study of resulting yields, grades, and costs as here presented shows little or no difference between dusting and spraying. Distribution and Importamce of the Potato Leafhopper The potato leafhopper has been known as a pest of potatoes for thirty or more years, with serious outbreaks occurring periodically. It is gen- erally distributed in practically every state in the Union and in parts of Canada and Mexico. Its greatest damage to potatoes occurs in the North Central states, and regularly in Wisconsin. FIG. l-FIELD OF EARLY OHIO POTATOES PRACTICALLY DEAD FROM HOPPERBURN Picture taken on July 17. Leafhoppers were extremely numerous. The field has not been sprayed with Bordeaux. The economic loss to the potato grower caused by the attacks of the potato leafhopper and the accompanying hopperburn is probably greater than that caused by any other potato insect (Fig. 1). As the insect re- produces very rapidly under favorable weather conditions, it is essential that the grower be equipped and read\- to cope with an outbreak before it threatens his crop. The Experiments Locations — The experiments were carried on at Spooner and Waupaca, Wisconsin in 1922. During the next three years, they were conducted only at Waupaca. Spooner is the location of a sub-experiment station in Wash- burn County in the northwestern part of the state. This county has a sandy Spraying Versus Dusting 3 soil; a moderate rainfall, averaging 3.34 inches per month from^ April to September; and moderate temperatures, the normal being 64.3°F. from June to September— the growing period. Waupaca is in Waupaca County, in the east central part of the state, and also has a sandy soil, but the normal rainfall and temperature are slightly higher than at Spooner, the precipita- tion averaging 3.68 inches per monh from April to September and the temperature 66.2°F. from June to September. FIG 2-SPRAYING UNDER HIGH PRESSURE PRODUCES A VERY FINE MIST With three nozzles per row, it is possible to apply 100 gallons of Bordeaux mixture per acre. Equipment— A triple pump traction sprayer was used at both places for applying the liquid spray. The boom, especially made for these experiments, covered four rows with three nozzles per row, one above the foliage point- ing downward and two nozzles below, one on each side, directed upward. The discs of the nozzles were replaced once a year with new ones having very small apertures in order to secure a fine mist (Fig. 2) and keep the pressure close to 200 pounds. To apply the dust, an engine-driven, four-row duster was used at Spooner and a four-row traction duster at Waupaca. Each was equipped with two nozzles per row (Fig. 3). Varieties— In all, five varieties of potatoes were sprayed and dusted: Triumph, Green Mountain, and Early Ohio one year; King three years; and Rural New Yorker four years. Formulas— \. Home made Bordeaux mixture — + pounds copper sulfate, 5 pounds hydrated lime, 50 gallons of water. Wisconsin Research Bulletin 82 FIG. 3— TRACTION DUSTER USED IN COMMERCIAL EXPERIMENTS With two nozzles per row properly set, the entire foliage can be covered with dust. 2. Copper-lime dust— -containing either 20 per cent or 25 per cent mono- hydrated copper sulfate according to whether or not an arsenical was in- cluded for the Colorado potato beetle. This copper-lime dust is not powdered Bordeaux but a mixture of mono- hydrated powdered copper sulfate, hydrated lime and filler. In the pres- ence of moisture, this combination turns to Bordeaux mixture. Other insects — As these experiments were carried on primarily to learn the effect of sprays and dusts upon the potato leaf hopper and hopperburn, the Colorado potato beetle was kept under control on all plots both treated and check by the use of calcium arsenate, except at Spooner where lead arsenate was used. On the treated plots, the arsenical was combined with the spray or dust whenever beetles were present. On the check plots, the beetles were controlled by spraying or dusting with the arsenicals alone. As beetles were nearly always present, the arsenicals were always used in the first two and often in the third applications. At no time in all four years were the potato flea beetle or potato aphid factors. Diseases — In these experiments, little or no attention was given to the effect of different treatments upon potato diseases. Two diseases, however, were present in the fields. Early blig-ht (Alfeniaria solan i) occurred to some extent each year, especially on the check plots although at no time was it a serious factor in the experiments. Late blight (Phytophthora infestans) also made its appearance, but at no time, in the locality where these experiments were carried on, was it a serious factor in potato pro- duction. Mention is made of these two principal foliage diseases because Spraying Versus Dusting the relation of yield to spraying and dusting is, of course, the net result of these treatments upon all insects and diseases of potatoes which are in any way alleviated by Bordeaux mixture. Time of day— The spraying was done at any convenient time during tlie day when little or no wind was blowing. Dusting was performed early m the morning if possible, while the air was still and the plants moist with dew (Fig. 4). F,G. ^ABSENCE gr^wmD A^ ^ESENCE O? DEW INCREASE dust soon turns to Bordeaux mixture. Experiments at Spooner Three varieties were treated at Spooner in 1922: Triumph, Green Moun- tain, and Rural New Yorker. Sprayed and dusted plots consisted of one- half acre each; the check plots were smaller, averaging about one-fourth acre Three applications were made, on July 12, 24, and Aug. 7. The amomit of liquid Bordeaux mixture (on an acre basis) ran from 80 gallons, for the first spray up to 120 gallons for the last. Applications of copper-lime dust containing 25 per cent monohydrated copper sulfate were altogether too heavy due partly to the use of a power duster and partly to inexperience in the amount of coverage necessary for protection. The amomit averaged 40 pounds per acre for the first and 86 to 100 poiinds for the last two applications. The temperature averaged about normal for Jmie and July with abundant precipitation. The infestation of leafhoppers was very light during these two months and hopperburn was scarce. An unusually hot, dry period 6 Wisconsin Research Bulletin 82 occurred the latter part of August, however, continuing into September. Largely on account of this condition, leaf hoppers reproduced at a rapid rate with the resulting increase and spread of hopperburn on both treated and untreated plots of the Triumph and Green Mountain varieties. Hopper- burn was more general and severe on the checks than on the treated plots. Very little hopperburn occurred on the Rural New Yorker plots, treated or untreated, although leafhoppers occurred in moderate numbers the latter part of August. A summary of yields is given in Table I. Table I. — Potato Yields in Experiments at Spooner, Wisconsin, 1922. Variety Treatment Yield in bu. per acre* Total No. 1 No. 2 Gulls Triumph do do Spray — home made Bordeaux Dust — copper- lime dust Check 38.4 43.4 34.7 34.0 36.7 30.9 No. 2'8 not sep- arated^ 4.4 6.6 3.8 Green Mt. do do Spray — home m,ade Bordeaux Dust — copper-lime dust Check 84.85 87.5 60.4 66.9 73.2 41.0 15,3 12.6 16.4 2,7 1.8 3.0 Rural New Yorker do do Spray — home made Bordeaux Dust — copper-lime duet Check 111.15 121.9 89.1 Not graded Not graded Not graded ^In grading potatoes, the U. S. Standard grade was always employed. This calls for a screen with \% inch apertures for round varieties and If^ inch aperatures for long varieties. All that do not pass through these screens are known as No. 1 potatoes. No. 2 potatoes (of all varieties) are those which do not pass through screens of 1% inch apertures. ^As the Triumphs were wanted for seed stock, they were not machme graded but picked over by hand into two sizes, "Salable" (No. I's) and "Small" (Culls). Experiments at Waupaca A commercial experiment was started at Waupaca in 1922, running four years. The amount of materials applied ran from 65 to 80 gallons of liquid Bordeaux mixture per acre for the first sprays and from 85 to 110 gallons for the last ones. No dusting was done in 1922. It was com- menced in 1923 and continued to the conclusion of the experiment. The average rate of application was from 18 to 25 pounds per acre for the first two and from 25 to 32 pounds for the last one or two applications. Waupaca Results — 1922 In 1922 four applications of liquid Bordeaux mixture were made, on July 7, 17, 27, and August 5, to one acre each of Green Moimtains and Rurals, using the special boom with three nozzles per row. These tests were com- pared directly with like acres sprayed by the grower five times but with only two nozzles per row and both of these pointing downward. The temperature averaged about normal for June and July with abundant precipitation. A hot, dry spell commenced the latter part of August and continued into September, Sprayinc; Versus Dusting 7 On account of the weather conditions the latter part of the summer, hopperburn became general in August on both varieties and severe on the Green Mountains. On August 18 the untreated Green Mountain vines were nearly dead; those sprayed with two nozzles per row, little better; while those sprayed with three nozzles per row were in the best condition of any although showing considrable hopperburn. On the Rurals, hopperburn was much less severe. The Green Mountains were dug on September 14 ; the Rurals the last of the month. Table II. — Potato Yields in Experiments at Waupaca, Wisconsin, 1922 Variety Treatment Yield in bu. per acre Total No. 1 No. 2 Culls Green Mt. Spray — liquid Bordeaux 3 nozzles per row 4 applications 186.5 144.0 36.0 6.5 do Spray— liquid Bordeaux 2 nozzles per row 5 applications 136.2 97.5 33.0 5.7 do Check 83.2 49.5 29.2 4.5 Rural New Yorker Spray — liquid Bordeaux 3 nozzles per row 4 applications 265.0 Not graded do Spray — liquid Bordeaux 2 nozzles per row 5 applications 250.0 Not graded do Check 180.0 Not graded The Green Mountain plot sprayed four times with three nozzles to the row yielded 50.3 bushels per acre more than the one sprayed five times with two nozzles per row. The corresponding difference in the Rural plot was only 15 bushels per acre in favor of the three nozzles. Treated plots of both varieties yielded far more potatoes, however, than the check plots. Waupaca Results — 1923 In 1923 it was necessary to make only three applications of spray and dust. These were put on July 13, 25, and August 6 to approximately one- half acre each of Kings and Rurals. The weather conditions in 1923 were nearly the opposite of those in 1922. The temperature was slightly above normal for June and July with a light rainfall. During August and September it was cool with a rainfall slightly above normal. The generally cool, wet season was most favorable to the growth of potatoes. At the same time it held back the development of leafhoppers, and consequently hopperburn did not assume serious proportions. An unusually early and severe frost on September 14 killed all the vines before the treated and check plots showed as much difference in the amoimt of hopperburn as occurred in 1922. The potatoes were dug Sep- tember 27, Wisconsin Research Bulletin 82 Table III. — Potato Yields in Experiments at Waupaca, Wisconsin, 1923 Variety Treatment Yield in bu. per acre Total No. 1 No. 2 CuUb King Spray — home made Bordeaux 231.65 209.0 21.25 1.4 do Dust — copper-lime dust 223.4 198.3 23.5 1.6 do Check 181.4 157.8 22.6 1.0 Rural New Yorker Spray — home made Bordeaux 237.4 209.1 24.9 3.4 do Dust — copper-lime dust 228.6 199.7 26.1 2.8 do Check 211.1 187.0 22.4 1.7 If the heavy frost had held off another two weeks, it is quite certain that a greater difference in yield between the treated and check plots would have been realized. Waupaca Results — 1924 Kings and Rurals were again used in the experiments of 1924, when four applications of spray and dust were made, on July 6, 17, 30, and August 20, to one-acre plots. This was an unusually cool, wet year ; the monthly mean temperature from June to September averaged 3.8°F. below normal. The rainfall in June and July was slightly below normal but for the next two months to- gether was 7.64 inches above normal. This year, while too cool and wet to expect any great number of leaf- hoppers and resulting hopperburn, was also too much below normal for the best production of potatoes. In accordance with what was expected, very little hopperburn appeared even on the checks and the yield of potatoes was below that of 1923. They were dug on October 3. '1 able 1\/. — Potato Yields in Experiments at Waupaca, Wisconsin, 1924 Variety Treatment Yield in bu. per acre Total No. 1 No. 2 Culls King Spray — home made Bordeaux 205.85 165.6 35.65 4.6 do Dust — copper-lime dust 192.2 154.7 33.7 3.8 do Check 184.3 144.5 37.2 2.6 Rural New Yorker Spray — home made Bordeaux 183.0 157.2 23.0 2.8 do Dust — copper-lime dust 210.1 181.7 25.3 3.1 do Check 188.55 156.4 28.7 3.45 Waupaca Results — 1925 In 1925 three varieties of potatoes — Early Ohio, King, and Rural New Yorker — were sprayed and dusted four times ; July 7, 16, 25, and August 5. Spraying Versus Dusting • 9 The summer of 1925 would be called generally a warm, rainy season. The temperature was considerably above normal in June and September and averaged normal during July and August. The rainfall was 2.7 inches above normal in Jtuie, 1.2 above in July, and below normal only in August. Hopperburn was quite prevalent in August and increased until by the middle of September there was a clearly marked difference between the treated plots and the checks. Potatoes were dug on October 2. Table V. — Potato Yields in Experiments at Waupaca, Wisconsin, 1925 Variety Treatment Yield in bu. per acre Total No. 1 No. 2 Culla King do Spray — home made Bordeaux Dust — copper-lime dust 211.1 190.4 159.6 143.0 51.5 47.4 Practi- cally do Check 151.5 92.9 58.6 none Early Ohio Spray — home made Bordeaux 190.4 141.9 48.5 do do Dust — copper-lime dust 162.6 117.7 45.0 do ' do Check 138.5 84.1 54.5 do Rural New Yorker Spray — homemade Bordeaux 185.7 144.2 41.5 do do Dust — copper-lime dust 170.9 128.6 42.3 do do Check 124.7 78.8 45.9 One of the unexpected results of 1925, discovered when potatoes were graded, was the large proportion of No. 2 potatoes to the total yield, of all varieties in treated and check plots alike. There were so few culls that it was not necessary to separate them from the No. 2's. It is evident that the generally warm, moist season resulted in an unusual development of the tubers so that practically all grew to at least No. 2 size. Table VI. — Percentage of No. I's to Total Yield of Potatoes in Experiments at Spooner and Waupaca, Wisconsin Locality Variety Percentage of No. I's to Total Yield Year Sprayed Dusted Check Difference between check and best treated plot 1922 1922 Spooner Waupaca Waupaca Waupaca Waupaca Waupaca Wa-upaca Waupaca Waupaca Green Mt. Green Mt. King Rural King Rural King Early Ohio Rural , 78.8 77.2 90.2 88.1 80.4 85.9 75.6 74.6 77.6 83.6 67.8 59.5 86.9 88.5 78.4 82.9 61.3 60.7 63.1 15.8 17.7 1923 1923 1924 1924 1925 1925 1925 88.7 87.3 80.5 86. 4 75.1 72.4 75.2 3.2 -0.4 2.1 3.5 14.3 13.9 14.5 Avera ge percentages 80.9 81.15 74.9 9.4 10 Wisconsin Research Bulletin 82 Bordeaux Mixture Increases Percentage of No. 1 Potatoes Protection of potato vines from hopperburn, whether by spray or dust, not only consistently increases total yield, but at the same time increases the percentage of No. 1 potatoes, which are the cream of the crop, Table VI. In these experiments the treated plots yielded a noticeably higher per- centage of No. 1 potatoes to their total yield than did the checks to their total yield, with one exception in which the percentage of No. I's was slightly higher in the check. The difference in favor of the series of sprayed plots over the checks is 6.0 per cent and the difference in favor of the series of dusted plots over the checks is 6.2 per cent of the total production of No. 1 potatoes. The average increase in percentage of No. I's of sprayed and dusted plots taken together, was 6.1 per cent higher than in the checks, which equals 8.4 bushels per acre of No. 1 potatoes. This increase is for the average year but the increase is much greater, as shown by the table, during severe hopperburn years (1922 and 1925) when fancy No. 1 potatoes usually command a better price. Table VII.— Total Yields and Percentage of No. I's in 1922 and 1925 at Waupaca, Wisconsin. {The Two Severe Hopperburn Years.) 1922 Total yield n bu. per acre Percentage of No. I's to total Variety Sprayed 3 nozzles per row Sprayed 2 nozzles per row Check Difference between check and best treat- ed plot Sprayed 3 nozzles per row Sprayed 2 nozzles per row Check Difference between check and best treat- ed plot Green Mt. Rural 186.5 265.0 136.2 250.0 83.2 180.0 103.3 85.0 77.2 71.5 Notg 59.5 raded 17.7 Averages 225.75 193.1 131.6 94.15 77.2 71.5 59.5 17.7 1925 Total yield n bu. per acre Per 3entage of No. I's t( ) total Variety Sprayed Dusted Check Difference between check and best treat- ed plot Sprayed Dusted Check Difference between check and best treat- ed plot King Early Ohio Rural 211.1 190.4 185.7 190.4 162.6 170.9 151.5 138.5 124.7 49.6 51.9 61.0 75.6 74.6 77.6 75.1 72,4 75.2 61.3 60.7 63.1 14.3 13.9 14.5 Averages 195.7 174.6 138.2 57.5 75.9 74.3 61.7 14.2 Fig. 5 portrays the temperatures and rainfall during two severe and two light hopperburn years. Thus it is evident that in 1922 and 1925 temperatures were above normal, with scant rainfall — especially during the latter part of the summer. In the other two years, just the opposite occurred. Spraying Versus Dusting 11 FIG. S— CHART OF TEMPERATURE AND RAINFALL IN WAUPACA, WISCONSIN. Comparing the two j-ears of severe hopperburn (1922 and 1925) with the two years of light hopperburn (1923 and 1924), a considerable difference in yield may be noted. Severe hopperburn years All treated plots 194.9 bu. All check plots 135.6 bu. Difference 59.3 bu. per acre. 12 Wisconsin Research Bulletin 82 Light hopperburn years All treated plots 214.0 bu. All check plots 191.3 bu. Difference 22.7 bu. per acre. A complete summary of the Waupaca experiments, including the torn years, is as follows: All treated plots 203.4 bu. All check plots 160.4 bu. Difference 43.0 bu. per acre. Table VIII. summarizes all experiments at both locations. Table VIII. — Summary of all Spraying and Dusting Experiments at Spooner and Waupaca Wisconsin, 1922-1925 Year Locality Variety Total yield in bu. per acre Sptfayed Dusted Check 19^ do do Spooner do do Triumph Green Mountain Rural New Yorker 38.4 84.85 111.15 43.4 87.5 121.9 34.7 60.4 89.1 1923 do Waupaca do King Rural New Yorker 231.65 237.4 223.4 228.6 181.4 211.1 1924 do Waupaca do King Rural New Yorker 205.85 183.0 192.2 210.1 184.3 188.55 1925 do do Waupaca do do King Early Ohio Rural New Yorker 211.1 190.4 185.7 190.4 162.6 170.9 151.5 138.5 124.7 167.9 163.1 136.4 The experiment at Waupaca in 1922 could not be included in Table VIII for it was concerned with two methods of spraying and not with spraying versus dusting. I'lG. f,— YirXD OF KINGS FROM SPRAYKD, DUSTED, AND CHECK PLOTS Yields here illustrated from two rows in each plot Left, check; Center, sprayed; Right, dusted. Placards are on stacks of No. 1 potatoes; No. 2's are on the right in each case. Spraying Versus Dusting 13 Thus, final results show that spraying is very little if any better than dusting, the difference amounting to 4.8 bushels per acre in favor of the former, which difference, however, is not outside the bounds of experimental error. Both spraying and dusting, however, increased average yields over the checks by 31.5 and 26.7 bushels per acre, respectively (Fig. 6). A final presentation of all treated plots compared with all check plots, at both locations for the four years on five varieties of potatoes including 14 sprayed plots, 10 dusted plots, and 12 check plots, shows : All treated plots , 171.4 bu. All check plots 135.6 bu. Difference 35.8 bu. per acre In considering these results, it must be kept in mind that the Colorado potato beetle was controlled on all treated and check plots alike. Relative Costs of Spraying and Dusting A simplified analysis and statement shows that dusting actually costs very little more than spraying^ We have not taken into consideration depreci- ation, interest, repairs, or other like items because such charges would appear to be quite similar for both machines. The actual labor including team, and cost of materials, for the number of applications given, should form the basis of criteria in the relative cost of spraying versus dusting. The prices of the sprayers and traction duster were the same. Tabular Comparison of Spraying and Dusting Spraying Dusting Unit — one acre Four applications, Amt. per application, per acre 88 gallons. Total— 352 gallons Unit — one acre Four applications, .\mt. per application per acre, 24 pounds Total — 96 pounds Copper sulfate 28.16 lbs. at 9c.. Hydrated lime 31.68 lbs. at Ic. Calcium arsenate 13.2 lbs. at 16c.. Cost $2 2 ur for two men and team for all operations for -. 6 $11 54 32 12 28 26 Cost Copper-linie dust, 20% monohydrated copper sulfate, 25% calcium arsenate 72 lbs. at $11.60 per cwt $8.35 (3 applications) Copper-lime dust, 25% monohydrated copper sulfate 24 lbs. at $8.80 per cwt... . 2. 12 (3 applications) Labor, $1.2."> an ho 75 minutes average spraying each acre Total (1 application) Labor, $0.75 an hour for man and team. 28 minutes average for dusting each acre 1.50 Total $11.97 Discussion Although the experiments recorded in this paper indicate that spraying increases the yield of potatoes to a slightly greater extent than does dust- ing, particularly in years when the leafhopper and hopperburn are severe, yet the difference shown in average yields is not outside the bounds of 14 Wisconsin Research Bulletin 82 experimental error. It is also apparent that the increase in yield alone is not the only factor which must be taken into consideration when recom- mendations are prepared for potato growers. A rather surprising reaction was experienced several times. Progressive growers familiar with our detailed experiments and the results thereof became enthusiastic, first, from the decided increase in yields from treatment with Bordeaux mixure and, second, from the greater advantages of dusting over spraying. An analysis of this rather common decision is interesting. Those growers most familiar with our experiments witnessed the whole spraying opera- tion as a rather detailed, complex, and difficult piece of work. On the other hand, the dusting operation could be carried on with only a frac- tion of the attention to details and immediately appealed to them from the standpoint of the following factors : Elimination of use of water Simplicity of preparation Reduced weight of outfit Speed of application Consequent willingness to repeat applications when necessary. Protection of potato vines from hopperburn whether by spray or dust, not only consistently increased total yield but at the same time increased the percentage of No. 1 potatoes. Finally it would appear that the net results of the experiments have been, first, to greatly increase the treatment of potatoes with some form of Bordeaux mixture ; second, to increase the proportion of dusting to spray- ing on account of its greater speed and adaptability. Several growers who formerly sprayed but two or three times are now dusting from four to six times, are increasing their yield more than ever before, and appear to be entirely satisfied with this method of hopperburn control. Table IX.— Mean Monthly Temperatures and Ramfall at Waupaca, Wisconsin, April to September, 1922-1925 Year Mean temperatures April May June July August Sept. 44.4 44.6 41.6 41.8 50.0 56.2 63.4 54.7 48.6 52.7 65.7 68.2 67.4 .59.7 70.0 71.0 69.4 71.5 67.2 70.6 67.9 70.0 64.2 66.4 68.4 60 4 1922 . 63 0 1923 - ... . 59 0 1924 56 5 1925-- .- 63.6 Rainfall Normal. 1922 1923 1924-... 1925-.. 2.66 5.07 3.13 4.61 1.16 4.53 3.70 1.87 5.07 1.81 4.43 5.86 3.06 3.84 7.13 3.35 4.37 4.05 2.74 4.64 3.44 3.27 4.17 11.06 2.28 3.83 2.63 3.68 3.85 3.99 Spraying Versus Dusting 15 SUMMARY Scientific workers arc generally agreed that the use of Bordeaux mixture on potato foliage will increase yields. The potato leafhopper is probably the most injurious insect pest of the potato plant, and it regularly causes serious damage in Wisconsin. Spraying and dusting experiments were carried on in commercial potato fields of Wisconsin for four years under regular farm conditions. The object of these investigations was to find out the cost of each treat- ment and how much each treatment increased the total yield and improved the grade. Five varieties of potatoes were included in the e.xperiments : Triumph, Green Mountain, Early Ohio, King, and Rural New Yorker. A definite increase in the percentage of No. 1 potatoes was secured both by spraying and dusting. The average increase of No. 1 potatoes in the treated plots was 6.1 per cent greater than in the check plots, which equals 8.4 bushels per acre. With one exception, yields were always increased by spraying and dusting. The average increases for spraying and dusting combined, varied from 22.7 bushels per acre in the two years of little hopperburn to 59.3 bushels per acre in the two years of heavy hopperburn. The average yields of all sprayed, all dusted, and all check plots when comipared, show the following distinctive results : All sprayed plots 167.9 bu. per acre All dusted plots 163.1 bu. per acre All check plots 136.4 bu. per acre Computations show the cost of spraying and dusting for four applications per acre to be: Spraying $11.26; dusting, $11.97. The yields from dusting and spraying were equal within the limits of experimental error. The cost of dusting was little more than that of spraying. • Dusting appeals more strongly than spraying to many growers on account of its greater speed and adaptability. 16 Wisconsin Research Bulletin 82 LITERATURE CITED (1) Boyd, Oran Cecil 1926. The Relative Efficiency of Some Copper Dusts and Sprays in the Control of Diseases and Insect Pests. Cornell Agr. Expt. Sta. Bui. 451. 68 p. illus. (2) Chambers, E. L. 1924. Potato Dusting and Spraying Demonstrations. Wis. Dept. Agr. Bui. 69:83-85 illus. (3) Dudley, J. E., Jr. and Wilson, H.. F. 1922. Combat Potato Leafhopper with Bordeaux. Wis. Agr. Expt. Sta. Bui. 334:32 p. illus. (4) Dudley, J. E., Jr. 1926. The Potato Leafhopper and How to Control it. U. S. Dept. Agr., Farm Bui. 1462:12 p. illus. (5) Fluke, C. L., Jr. 1919. Does Bordeaux Mixture Repel the Potato Leafhopper? Jour. Econ. Ent. 12:256-7. (6) Folsom Donald and Bonde, Reiner. 1926. Potato Spraying and Dusting Experiments 1921 to 1925. Me. Agr. Expt. Sta. Bui. 334:85 p. illus. (7) Stewart, F. C. and Parrott, P. J. 1924. Experiments with Potatoes. New York Agr. Expt. Sta. Bui. 518:36 p. Research Bulletin 85 June, 1928 A Fusarium Wilt of Peas in Wisconsin Maurice B. Linford Agricultural Experiment Station of the University of Wisconsin, Madison. CONTENTS Introduction 1 The Disease Geographical range 1 Economic importance 2 Description of the disease 3 Etiology 7 Pathological histology 7 Other diseases of the pea caused by species of Fusarium 10 The Pathogen Taxonomy 11 Relation of temperature to growth in pure culture 12 Pathogenicity 13 Dissemination 13 InJfluence of Soil Temperature and Moisture Upon the Disease Greenhouse experiments 14 Discussion of experimental results 21 Relation of Soil Temperature and Moisture to the Occurrence of Pea Wilt 23 Other Factors Influencing Field Occurrence Number and frequency of previous crops of peas 26 Soil type 21 Resistance to Fusariumi Wilt Relative varietal susceptibility 2S Resistant peas obtained by selection 30 Degree of resistance to wilt i^ Practical usefulness of wilt-resistant varieties 34 Control of Fusarium Wilt of Peas 36 General Discussion ol Summary 40 Literature Cited 43 w A Fusarium Wilt of Peas in Wisconsin Maurice B. Lini'oku 'HERE PEAS ARE GROWN intensively during a period of years, avoidance of disease becomes a problem of major concern. This has proved particularly true in the production of green peas for canning. Since 1910, investigations of the causes of pea failure have been in progress almost continually at Madison, supported by the Wisconsin Agricultural Experiment Station, the United States Department of Agricul- ture, and by the canners themselves. From these and other investigations it was soon recognized that such failures are generally the result of one or more diseases caused by parasitic fungi and bacteria, and recommendations for their control gave partial relief, but up to the present time it has not been possible to eliminate them entirely. The diseases that have been found most frequently important are those which attack the underground parts of the plant. In Europe such injuries are generally grouped under the name of "St. John's Disease" (11) which is there attributed to several species of Fusarium. In the United States the chief root disease is the rootrot caused by Aphanomyces euteiches Drechsler CIS), but numerous other parasites are commonly associated with it in causing serious root decay, including: Fusarhim rnartii App. & Wr. var. pisi F. R. Jones (14) ; Mycosphaerella pinodes (Berk. & BIox.) Stone, and a "micro" form of this same species (25) ; Rhizoctonia solani Kiihn : and Pythinm spp. (16). During the summer of 1924 in a survey of Wisconsin pea fields (16) for the study of field occurrence and relative importance of these diseases, a Fusarium wilt (24) hitherto unrecognized, was discovered as an additional cause of important pea failures. A study of this disease was begun by the writer during the winter of 1924-25 and has been continued to include field, greenhouse, and laboratory investigations during 1925-26 and 1926-27. THE DISEASE Geographical Range Fusarium wilt has been found generally distributed in the older pea grow- ing districts of southern Wisconsin, but detailed information is still lacking concerning its prevalence in the northern part of the state. Of a total of 59 localities which have been searched. 35 have been found infested. These The writer Kratefully acknowledges the vahiable advice and assistance of Fied R. Jones and L. R. Jones throughout this investigation. Wisconsin Reseakcm IjurjAmx 85 i-i(i. 1- (.i:<)(;i?aphi(;al uanc.k of i usarium wilt of peas in Wis- consin AS DETEHMINED OCRING 1924, 1925, AND 1926. Bhick (lols rcpi'esciit localities in \\hic'h will hi\s been found. arc di.stril)utc'(l, as shown in Figure 1, in 21 counties which represent the most important pea producing areas in Wisconsin. Outside of this state, wilt has been reported from only two localities. Dr. J. B. Kendrick collected it at Peru, Indiana, in 1926 and sent the writer specimen plants and cultures of the wilt fungus in confirmation. In the same year the writer found it associated with rootrot at McMillan, Michigan. These observations, although narrowly limited, suggest that this Fusarium wilt nia\- he found ratlicr generally distributed at least in the older pea prixhicing regions of tlie northern United States. Economic Importance Fusarium wilt of peas is destructive more frequently in association with several diseases of the rootrot type than as an isolated disease, a fact that A I-'USARIL'M W ILT OF PeAS 3 makes dilticull a precise statement of its ccmiomic importance, in over a dozen of tlie older cannery districts in southern Wisconsin, however, it caused more important losses during the period of this investigation than all other diseases combined. Here it has probably been, for several year> at least, the cause of a considerable part of the injury hitherto attributed to the Aphanomyces rootrot. In Wisconsin as a whole it appears to be less important than that rootrot Init more destructive than any of the other pea diseases. Wilt reduces (|uantity of yield more than quality of product. Affected plants die (luickly and in infested areas the crop may be destroyed com- pletely. During 1925 in a survey of 693 fields with a combined area of 4,564 acres, crop reduction from wilt was estimated at 6.6 per cent. This figure is too high for the state as a whole, but much too low to represent conditions in the most severely infested localities. With the increasing age of intensive pea culture, diseases are gradually rendering unfit for the crop those limited areas best adapted to pea growing. Continued success of the pea industry in regions where it is now established is therefore partially dependent upon the development of adequate means of disease control. Description of the Disease l-'usarium wilt of peas occurs characteristically in scattered spots of circu- lar outline and variable size whicli mav be distributed in large numbers KUl. 2.- FIELD OF WINNEF. VARIF.TY I'EAS SI'OTTi:i) WITH WILT. The palp areas, dislribiiied Ihiough the field. conlaincHl plants in various .stages !!|' the wilt dise.l^e. 4 Wisconsin Research Bulletin 85 through a field (Fig. 2). These spots sometimes consist of only a few- plants but they may enlarge both during the season and from year t(j year until they involve the entire field. They first become discernible through a bluish paleness of the vines which gives way to yellow as the plants die. Plants are afifected first at the center of such spots and then progressively outward towards the margins which may continue to advance until harvest or maturity. An apparent radial spread of three feet or more during the season is frequently observed. Symptoms The initial and most characteristic symptom of this disease is a recurving of the margins of the younger stipules and leaflets which leads to separation of the tips of the stipules in the terminal bud before their bases begin to diverge, and to rolling of the leaflets (Fig. 3). Simultaneously the upper IIG. 8.— EARLY SYMPTOMS OF \YILT IN HORSl'ORU PEA .Vllcttcd plant at riglil shows leiiii v iiig of margins of stipules and Icaflcls of tlie iippei- leaves, loss of color and withering of lower leaves, retardation of teiniinal gi-o>\th, and slightly increased stem diametei- at the giound line. Roots of this plant showed no conspicuous exteinal injury. (Slightly jeduced.j A FusAiuiM Wii.T OF Peas fk; L-LATK STAGES OF WILT IN PERFECTION PEAS. tin. gie.nh..us.- ;.t the variable tenipen.tin e l(.-20 degitcs C. parts of the plant mav become m'.e anl develop a bluish sheen, the terminal bud mav be checked in its growth, the stem and upper leaves may become more rigid than normal, and the roots more cnsp and brittle, whde the lower leaves turn pale and commence to wither. Sometimes the enfre plan. becomes yellowish and the leaves wither progressively upwards to tne terminal bud. Characteristically, however, after the collapse of a tew basal leaves, the upper part of the plant wilts abruptly and may become dry while still green in color. After such wilting the stem shrivels downward from the tip toward the basal internodes which remain firm and turgid until the end (Fig. 4). The disease may affect one side of a plant earher than the other, giving rise to a unilateral expression of symptoms which i. a useful diagnostic characteristic of this Fusariuni wilt as of some related diseases of other plants. . . „ , The roots of plants affected with wilt alone characteristically shovv a tew dead rootlets and a limited superficial browning but otherwise are white and clean externallv. The dead rootlets which have undergone a dry decay generallv point" at their base to the most pronounced vascular discoloration fo be found ,n the taproot. Vascular discoloration varying from an almos imperceptible yellow to a rich orange brown, deve ops m the "PP^^ P^^^* the taproot (Fig. 5) and may extend into several internodes of the .tern. General root de^ay seldom begins until the wilted ^te- has dried a^so its base. Then, in wet soils, the succulent cortex of the basal inte.nodc, mav become covered with a white fluffy grcwth of mycelium. A plant may wilt within two days after the first symptoms ^PP- • ^^^ even under conditions very favorable for the disease, an interval ot ten to Wisconsin Research Bulletin 85 I-IG — VASCILAR DISCOLORATION IN THE BASE Of UPPER PART OE THE TAPROOT. THI': STEM AND This plant, which showed only preliminary wilt symptoms, lias been split lengthwise to expose the discolored stele. Note that the root cortex and the nodnles e\-eii at this stage of \ascnlar discoloration are still white and sonnd. twelve days is m>vv frcqiit-ntly observed. During cool weather in the early spring, symptoms develop slowdy, bnt on sncceeding bright warm daj'S the plants wilt rapidly. When progress of the disease is slow, ])lants may die withont a detinite wilt phase. On snch ])Iants, one or more lateral shoots may arise from the basal nodes and grow to a height of several centimeters liefore each in its turn collapses. The increased rigidity of the plant in the earliest stages of the disease is associated with a marked change in the water relationships of the plant. The basal internodes, especially near the soil line, become distinctly swollen : one series of measurements showed a 20 per cent greater cross sectiona' area in the second internode of diseased plants as compared with healthy A l^^jsAKiu.M Wii.r OK Pkas 7 plants grown under similar conditidns. Such dibt-astd plants arc particularly retentive of water and, when removed from the soil, wither much less rapidly than do healthy plants. The swollen basal portion of the stem is especially resistant to drying. These symptoms collectively are clearly distinct from those of other known diseases of the pea. L'p to the present, the characteristic rcjlling of leaflets and stipules, and the checking of terminal growth have been seen induced in the greenhouse by only one set of conditions other than wilt infection, and that is calcium deficiency maintained experimentally by Miss Dorothy Day in studies, at the University of Wisconsin, not yet published. In calcium starvation she found rolling of stipules and leaflets and increase in rigidity of leaves and stems indistinguishable from the early symptoms of wilt. The degree of calcium deficiency required to produce this condition is not to be expected in agricultural soils. Etiology Preliminary examination of plants affected with this disease revealed the presence of abundant fungous mycelium in the discolored vascular tissues, and numerous isolations have yielded cultures predominantly of one species of Fusarium. The pathogenicity of the fungus has been demonstrated by numerous inoculation and reisolation experiments, several of which are described in following sections of this paper. During the summer of 1925 isolations'- were made from 68 collections of affected plants from many localities. From the cultures thus obtained, 161 were selected for an inoculation experiment in which peas were grown in pots of steamed soil into which the fungus was introduced, in culture on Melilotus stems, at the time of planting. Of the 118 cultures of the pea wilt fungus, 75 produced the disease, while none of the several other species of Fusarium tried proved significantly pathogenic. This and other similar experiments have been accepted as adequate evidence that one species of Fusarium causes pea wilt throughout the known geographical range of this disease. Pathological Histology The most conspicuous and apparently most significant development of the pea wilt fungus occurs in the xylem of both roots and stem. The parasite, to attain this position, must pass through cortical and undifferentiated tissue, but, with the exception of the occasional rootlets of entry which are thor- oughly invaded, it occurs only sparingly in the cortex, chiefly in the outer ^Several nu'thods were found expedient for isolating the fnnsiis troin roots and stems, including; surface sterilization in mercuric chloride followed by washing and then cutting into segments with llamed scissors before plating on acidified potato-dextrose agar; cutting the cortex from the stele with a tlamcd razor blade and plating fragments of the stele; and pulling the xylem core out from the coitex by grasping the tap-root in one hand antl the stem base in ihe other, and plating' fragments of the core thus exposetl. ^These studies were made with stained paraffin sections and with free-hantl sections either fresh or stained with cotton-blue in lacto-phenol as outlined l)v Klcbahn ( 21 i . W'lscoxsix Research Bulletin 85 l-IC. (i— MYCELIUM IN XYLEM OE HORSFORD PEA PLANT DYING IROM WILT, SECOND INTERNODE. a. Transvcisp section, showing niyccliuni crowding some vessels and occur- ring more sparingly in others and in the xylem parenchyma, x 340. b. Longitudinal section, showing a vessel crowded with mycelium, and hy- phae passing through jiils in the walls of this vessel and into the suirounding elements, x 375. layers. No conspicuous cortical lesions such as characterize the Fusannm viartii pisi footrot are formed by this vascular parasite. Within the xyletn the fungus distributes itself chiefly in the long, con- tinuous tracheae, but it also invades freely the surrounding xylem paren- chyma. Thin walls arc penetrated directly; thickened walls chiefly if not solely through pits (Fig. 6b). Thorough invasion of the xylem is gen- erally accompanied by a limited invasion of other stelar tissues. The fungus frequently gains access to the cambium where it causes complete dis- organization (Fig. 7). From there it may invade the phloem and pericycle, but the endodermis constitutes an eff^ective barrier which prevents the fungus from growing out through the cortex from an invaded stele until very late stages of the disease. A FusARiuM Wir/r of Peas IIG 7— BREAKDOWN OK CAMBIUM REGION IN A YOUNG HORSKORD PEA PLANT DYING FROM WILT. In this section of the first intcrnode, the xyleni is tlioroughly invaded and the phloem sparingly as indicated by localized dark areas, x 130. A few days after the first symptoms appear, the wilt fungus may be found readily in the xylem of the taproot near the base of affected lateral rootlets, and somewhat later throughout the entire upper portion of the root system and in the basal internodes of the stem. In late stages of the disease the fungus usually extends through the lower half of the stem, frequently through five or si.x internodes, but it has not been found in any instance within two internodes of the lowest blossom or pod. and its entry into even the lowest leaves is rare. When plants die slowly without a distinct wilt phase, the fungus may be found only sparingly within the xylem. In the usual course of the disease, however, pea wilt is characterized by the accumulation of mycelium in the vessel? in greater amounts than has been described in related diseases of other plants. Vessels sometimes become crowded with closel}^ packed masses of mycelium which might, apparently, partially obstruct the passage of water, but this condition does not appear to have any bearing upon the production of the characteristic preliminary symptoms of the disease. 10 Wisconsin Research Bulletin 85 Other Diseases of the Pea Caused by Species of Fusariutn There are numerous reports in literature of the occurrence of species of Fusarium on the pea, but a parasitic relationship has been demonstrated by adequate experimental evidence in only a few instances, and in each well attested case the disease produced is primarily a cortical rot with very limited vascular invasion. The reports of Fusarium species on peas have been summarized so recently by Jones (14 J that a complete review will not be necessary here. In European literature, Fusarium on peas is mentioned chiefly in con- nection with the "St. John's Disease," a foot disease described by van Hall (11) in Holland in 1903, and attributed by him to Fusarium vasinfectum Atk. var. pisi (variety not described). Gueguen (10) and Capus (3) in France later found what they considered to be this same fungus associated with the disease but concluded that its entry into the pea roots was secondary to other injury. Schikorra (31) isolated a different fungus which he mistook for van Hall's organism, and concluded on the basis of inadequate inoculation experiments that it was the cause of the disease in Germany. This fungus was later placed by Appel and Wollenweber (1) in the new species Fusarium falcatum. Wollenweber (38, 39) while in the United States, described F. redolens Wr. as the cause of a wilt and foot disease of peas without presenting any of his experimental evidence. Since then he has mentioned this fungus (41, 42) as probably the chief cause of the St. John's disease in Europe, but has concluded that at least two other species, F. viartii App. and Wr., and /''. falcatum App. and Wr. may also cause the disease. Turreson {Z7) has given a more detailed statement concerning a foot disease of peas in Sweden caused by F. viticola Thiim., characterized by a reddish-brown decay of the base of the stem beginning at a point near the seed. In the United States, the only species that has formerly been shown to be an important parasite of the pea is F. martii App. and W. var. pisi F. R. Jones (14). This fungus causes a cortical decay of the stem base and roots and may, at high temperatures, invade the vascular system of the stem for a short distance. Jones made numerous isolations from the vascular systems of plants showing root decay and thereby secured a number of species, but of the following fungi from this and other sources none proved significantly parasitic : Fusarium oxysporum Schlecht., F. solani Mart., F. sclerotiodes Sherb., F. vasinfectum Atk., and F. redolens Wr. More recently Togashi (36) has reported wilt of peas in Japan caused by three undetermined species of Fusarium. This also, from his descriptions, is a footrot rather than a typical vascular disease. All of these reports are concerned with diseases which dififer in symptoms, etiology, and pathological histology from the disease under consideration in this paper. A FUSAKILM W'll.T ()!■ I'KAS 11 THE PATHOGEN Taxonomy Four pathogenic cultures of the pea wilt fungus were submitted to Dr. H. W. Wollenweber of the Biologische Reichsanstalt fiir Land- und Forst- wirtschaft at Berlin-Dahlem, Germany, for determination. He kindly re- plied that they might be placed as Fusariuiii orthoccras App. and Wr. but observed that they showed some divergence in pigmentation from the type of the species and indeed were not constant in this character among them- selves. One of them, showing no bright colors when grown on steamed rice, agreed closely with F. conglutinans Wr., the cabbage yellows parasite. Fusarium crthoceras (38, 40) is a species of the section Elegans that has been isolated by different workers from potato, sweet potato, and other plants including the pea (23) and has been regarded as a cosmopolitan saprophyte. It forms no sporodochia or pionnotes, and the microconidia are always more numerous than the straight, triscptate macroconidia. Chlamydospores are usually abundant. For comparison with the fungus from pea wilt, the writer obtained sub- cultures of authentic F. nrtlwceras from four different sources, cultures which had come originally from Europe and Honduras by way of the Fusarium Conference (43) of 1924. The eight cultures thus obtained, when grown on standard substrata (43), were in excellent agreement among them- selves, showing only minor distinctions in cultural characters, but all of them were totally different in appearance from any of the numerous cultures of the pea wilt fungus which have been studied in this investigation. With a single exception they were sporulating freely and all of them were pro- ducing suflficient triseptate conidia of typical form for purposes of identi- fication. The pea wilt fungus, on the contrary, sporulates very sparingly. Micro- conidia may sometimes be found in fair abundance, but even in fresh isola- tions and on diverse substrata held at different temperatures, the writer has been unable to find sufficient macroconidia of uniform type in "normal con- dition" (40) to permit satisfactory morphological comparison with F. orthoceras or other species. In the opinion of the writer the pea wilt fungus is distinct from the fungi that have been placed in the species, F. orthoceras App. and Wr. In the absence of an adequate morphological basis, however, it can perhaps be best considered a variety of that species because of its predominantly nonseptate conidia, the abundance of thick walled chlamydospores, and the presence of pigmentation in culture on rice. Therefore it is described tentatively as a new variety, with the description based upon growth in culture on standard substrata. Fusarium orthoceras App. and Wr. var. pisi (n. var.). Differs from F. orthoceras in paucity of microconidia and almost complete absence of macroconidia, also in absence of vinaceous colors on potato - dextrose agar, in dominance of other than vinaceous colors on rice, and in absence of salmon colored spore masses on various media. Macroconidia rarely present, irregular in size and form : microconidia continuous or rarely 12 Wisconsin Research Bulletin 85 septate, irregularl,v ellipsoidal to elongate, somewhat curved, borne on the mycelium either above or within the substratum, usually in very small riumbers and sometimes wholly absent ; chlamydospores relatively abundant, typical of the species, simple or compound, thick-walled, tuberculate ; my- celium well developed, variable in diameter and septation, at first white in mass, becoming tawny to variously tinted ; pigmentation of substratum highly variable from strain to strain, but including the following colors (30) in well developed cultures of the strongly pigmented strains grown three weeks on boiled rice: (above, drj^ growth) hydrangea pink, orange-vina- ceous, coral pink, ocher red, dark vinaceous, victoria lake; (below, moist growth) pale pinkish cinnamon, buff pink, glaucous, malachite green, clay color, and Isabella color. Some strains of the fungus remain colorless or develop only the least intense of the colors named above. The cause of a vascular disease of Pisum sativum L. On plain potato agar the cultures remain dirty white or develop small amounts of blue-green pigment. This color is distributed characteristically in tube cultures as a narrow lip where the lower edge of the slant surface touches the glass ; in petri dish cultures it occurs in irregular areas in the older part of the colony, most clearly visible from below. The agar remains essentially unchanged in color. On potato-dextrose agar the characteristic blue-green lip is accompanied by irregular pink areas in the mycelium and by pigmentation of the sub- stratum, ranging from light golden brown to very dark purplish brown. Sometimes pigmentation of the mycelium and substratum are completely lacking. On oat agar, blue-green is less frequent than on potato. The mycelium is white, pink, or tawny : the substratum strongly colored, brownish, rich vinaceous, or olivaceous. On Melilotiis aiha stems the m\celium is tawny or dirt\' white. On potato plugs the mycelium remains tawny white or becomes spotted blue-green. Relation of Temperature to Growth in Pure Culture For comparison with other wilt-producing species, Fusariiiiii ortlwccras var. pisi has been grown on petri dish plates of potato-dextrose (2 per cent) Tabic I. — Results of CJiaractcristic Groii'th-tcmpcrature Experiment in JVhich Fiisarium nrtlwceras 7 maintain the desired soil moisture content. Upon emergence, the seedlings were thinned to ten vigorous and well spaced individuals. A layer of mineral wool or of white sand was spread over the soil for insulation. Experiment I. — At each of seven soil temperatures (8, 13, 17, 21, 25, 2^). and 33 degrees C.) one can was planted with Horsford peas without inocu- lation and four were inoculated at time of planting, each with a different culture of the wilt fungus from a difi^erent locality. These cultures, on A FusARiuM Wilt of Peas 15 Table II. — Siiminary of Results of Soil Temperature Experiment I. In- fluence of Soil Temperature Upon the Production of Wilt in Horsford's Market Garden Peas by Four Different Cultures of the Wilt Fungus Plants afifected in 36 days Soil temperature Culture number of inoculum Average 7 21 251 255 °C. % % % % % 8 0 0 0 0 0 13 50 40 10 10 27.5 17 60 40 40 70 52.5 21 80 70 60 70 70.0 25 80 60 40 40 55.0 29 10 50 50 10 30.0 Average 56 52 40 40 Melilotus stems, were distributed through the two inches of soil immedi- ately below the seed. The water content of the soil was held in all cans alike at 75 per cent of holding-capacity. Wilt symptoms were noted first at 21 and 25 degrees C, 23 days after planting. At all times a larger percentage of plants was affected at 21 degrees than at other temperatures, with 25 and 17 degrees C. ranking second and third respectively. At the conclusion of the experiment after 36 days (Table II), no wilt .symptoms had appeared at 8 degrees, but at all other temperatures disease had developed in the cans inoculted with all four cultures of the fungus. Differences in wilt production by these cultures were slight. Except at 33 degrees where both controls and inoculated plants died soon after emergence, the controls remained thrifty during the period of the experiment. Experiment II. — In this experiment the effects of soil moisture as well as temperature were studied. At each of five temperatures (Table III) five cans of Horsford peas were planted, three inoculated with infested soil (culture no. 255 from experiment I above) and two left uninoculated. At Table III. — Summary of Results of Soil Temperature Experiment II. The Influence of Soil Temperature and of Soil Moisture Upon the Develop- ment of Wilt in Horsford Peas Plants affected in 31 days Plants dead in 31 days Soil temper- ature Soil moisture holding- capacity A\-erage Soil moisture he capacity Iding- Average 40% 65% 90% 40% 65% 90% °C. 13 17 21 25 29 % 0 50 90 90 40 % 10 80 90 100 20 % 20 90 100 100 60 % 10 73 93 97 40 % 0 10 40 20 16 % 0 0 80 70 io % 0 10 70 60 0 % 0 7 63 50 7 Average 54 60 74 17 32 28 16 Wisconsin Research Bulletin 85 each temperature the soil in the three inoculated cans was adjusted to 90 per cent, 65 per cent, and 40 per cent of its water holding-capacity, and in the two controls to 90 per cent and 40 per cent. During the experiment, ^vater was added daily until the daily requirement exceeded 15 grams per can, and then twice daily. In spite of this at the conclusion of the experi- ment, the uneven distribution of water in the soil made it apparent that these figures were approximate only : soil was wet, medium-wet, or rela- tively dry. These soil moistures all permitted approximately normal growth of the plants ; but the plants came up earlier, produced longer internodes and larger leaflets, and had a brighter green color in the wet and medium- wet soil than in the dry. Control plants remained free from wilt through- out the experiment. The influence of temperature upon the disease was much as in the first experiment. Disregarding the effects of soil moisture (see averages of the three soil moisture contents. Table III) there were more plants affected after 31 days at 25 and 21 degrees C. than at higher or lower temperatures. This is still more evident in the average counts of numbers dead in 31 days. The only difference from Experiment I is that here 25 degrees appears slightly more favorable than 21 degrees for the early appearance of symptoms but, in conformity with that experiment, 21 degrees appears most favorable for early death of the plants. Considering now the influence of soil moisture, it is apparent, first, from Table III, that the cardinal soil temperatures for the disease remain essenti- DAYS AFTER PLANTING FIG. 8— RELATION OF SOIL MOISTIRF (.ONTKXT TO RATE OF DEVELOP- MENT OF FUSARIUM WILT OF PEAS. "Wet" soil contained approximately 90 per cent of its holding-capacity; "medium -wet," 65 per cent; and "tlry,' 40 per cent. Soil temperature experi- ment II. A FusARiuM Wilt of Peas 17 lilly the same in dry as in wet soil. There are differences, however, in the rate at which the disease develops. As shown in the table, more plants were affected in 31 days in the wet soil than in the medium-wet or dry. Figure 8 shows this to have been apparent also at other times, and much more conspicuous early in tl-.e experiment than late. Table III shows, however, that more plants were dead after 31 days in medium wet than in either wet or dry soil. This means that the course of the disease after inception is much more rapid in the medium-wet than in the wet soil. Rec- ords of the disease in individual plants showed that in medium-wet and dry soil the average interval betw-een the appearance of first symptoms and the death of the plant w-as six days, as compared with nine days in wet soil. This experiment thus shows: (a) that soil moisture content does not modify markedly the influence of soil temperature upon the development of pea wilt; (b) that the influence of soil moisture content is relatively slight in comparison with the influence of soil temperature ; (c ) that wet soil favors the earliest appearance of symptoms; (d) that medium-wet soil favors the earliest killing of plants ; and (e) that dry soil, while delaying the appear- ance of initial symptoms, leads to rapid dying of plants after they show symptoms of the disease. Clayton (5) found that soil dry enough to retard the growth of tomato plants delays the appearance of tomato wilt, but Tis- dale (33) found that the driest soil used in his experiments was more favorable for the development of cabbage yellows than either medium or wet soil. Experiment III. — In this experiment the influence of soil temperature upon the disease was studied in two varieties of peas comparatively : Alaska, an early or short-season variety, and Horsford's Market Garden, a late variety. At each of seven temperatures from 12 to 30 degrees, Horsford peas were planted in two soil-temperature cans of infested soil and one Tabic IV. — Suuunary of Results of Soil Temperature E.vperitiient III. The Influence of Soil Temperature Upon the Development of Wilt in Hors- ford's Market Garden and in Alaska Variety Peas Days Days Plants affected Plants dead Total until until after after 8oil temperature plants first symptom first death 20 24 1 29 36 20 24 29 36 days days dajs days days days days days °c. No. No. No. c- '0 C", % % w % % Horsford 12 20 36 -- 0 0 0 10 0 0 0 0 15 " 22 36 0 20 85 100 0 0 0 .5 18 " 18 26 40 90 100 100 0 0 25 70 21 ■' 12 20 85 100 100 100 15 35 70 100 24 13 24 70 95 100 100 0 10 25 95 27 15 24 20 80 95 95 0 5 25 70 30 20 30 5 15 25 40 0 0 0 10 Alaska 12 10 .. 0 0 0 i 0 0 0 0 0 15 " 26 0 0 10 i 80 0 0 0 0 18 24 36 0 20 60 1 90 0 0 0 40 21 " 15 20 60 90 90 90 10 10 60 90 24 " 16 22 20 60 70 80 0 10 40 60 27 " 20 29 20 30 50 60 0 0 10 40 30 36 "' 0 0 0 10 0 0 0 0 18 W'iscoxsix Research Bulletin 85 of steamed soil, and Alaska peas in one can of infested soil and one of steamed. Soil from Experiment II, sifted and mixed to insure uniformity, was used as inocu.um. Soil moisture was adjusted in all containers alike to 70 per cent of holding-capacity. Detailed records of the progress of the disease from inception to death were kept separately for all individual plants, and similar records were made to trace the development of the healthy con- trol plants. Table IV, presents the chief numerical results of this experiment, and figures 9 and 10 illustrate the results with the Horsford variety. In both varieties alike the disease developed in fewer days at 21 degrees than at any other soil temperature. Throughout the experiment more plants were affected at this than any other temperature, but after 36 days all the Hors- 12 15 18 21 24 27 SOIL TEMPERATURE DEGREES CENTIGRADE 30 FIC. !) — RKLATION Ol' SOIL TEMPKRATL I\E TO THE DKVELOP>n:N'r Ol" PEA WILT. Per cciil of pl.iiils till'efted with wilt nl iliU'crciit soil tdiipcratiiics whoii grown tlie indicaled iiuiiiIkt ol' days in soil inoiiihilcd \silh pine cultnics of the pea will I'unRiis. Horsford's .^la^k('l (iaidiM) v.nicty; soil tcuipcraliii'f cx- pciiniiMit III (Tal)lc IVl. ford plants were affected at 15, 18, 21, and 24 degrees C. At temperatures above 21 degrees the disea.se developed more rapidly in both varieties than at temperatures correspondingly below, but it was less regular and ultimately aft'ected fewer plants than in the lower range. As in the first experiment. 24 degrees was distinctly less favorable for the disease than 21 degrees, l)ut more favorable than 18 degrees C. The disease developed earlier in Horsford than in Alaska peas at the optimal temperature, and likewise developed over a wider range of tempera- tures. At the close of the experiment, after 36 days, some Horsford plants were dead at all temperatures except 12 degrees, w^hile all Alaska plants were still alive at 12, IS, and 30 degrees C. A FusARiuM Wilt ok Pkas 19 a: rt H '5 ■A K : ~ 3 1/ 20 Wisconsin Kkskakch IUilletin 85 la this as in the former experiments, the most rapid germination and early growth of the control plants occurred at 24 to 27 degrees C. At 30 degrees emergence of the seedlings was only slightly less rapid, but at low temperatures it was much retarded. Judging from the appearance of tlic plants, both Alaska and Horsford made normal growth from 15 to 24 degrees, and approximately normal even at the extremes of this experiment. Leitch (22) and Jones and Tisdale (13) had shown earlier that continued growth of peas may not be expected at temperatures above 30 degrees, and this was confirmed in Experiment I in whicli peas germinated but soon died at 33 degrees C. Dry weight determinations made on the control plants in this experiment after 39 days found the greatest weight per plant at 18 degrees in Alaska and at 21 degrees in Horsford. Dry weights of roots alone were greatest at 12 degrees in Alaska and at 18 degrees in Horsford, and were much reduced at the higher temperatures in both varieties. These determinations suggest that the Horsford is adapted to temperatures slightly higher than is Alaska, but this is probably an illusion arising from the fact that the Alaska plants, already in blossom, were much closer to maturity than were the long-season Horsfords which were not yet showing their first blossom buds. Comparist)n of these results with dry weight determinations made by Rich- ards (29) and Gilchrist (9) on Alaska peas grown shorter times indicates a gradual shift of the greatest dry weight to lower temperatures as the plant approaches maturity. It seems probable from this and other evidence that during a longer time, Horsford peas would develop their greatest dry weight at temperatures below 21 degrees C. The rate of shoot growth, as measured by the rate at which nodes are expanded from the terminal bud, was not significantly modified by soil temperature where the shoots themselves were exposed to the same air tcmperatin-e. At the end of ten days, during which all plants came up at even the lowest temperature, there were marked differences in the average numbers of nodes exposed at the different temperatures, but during succeed- ing intervals imtil blossoming began in the Alaska and until the close of the experiment in Horsford, the rate of node expansion was equal at all tempera- tures, and slightly more rapid in Alaska than Horsford. At medium low temperatures, favored by larger and more active root systems, sliriots grew coarser and thus, eventually produced a greater dry weight tlian the more frail shoots produced by the weaker roots in the warm soil. Since the pea wilt fungus makes its entry into the root system from the soil, and, until the late stages of the disease, develops chiefly in the sub- terranean portion of the plant, it is probable that the temperature effects upon the host that are most important in the study of pea wilt are the direct effects upon the root system. Experiment II'. — In a fourth experiment, Morsford peas were grown 50 days at 10, 12, 14, and 16 degrees C. in three lots of soil inoculated each with a separate culture of the wilt fvmgus, and in one lot of uninoculated steamed soil. One of these strains of the fungus proved less pathogenic than the other two and gave only 70 per cent infection after 50 days at 16 degrees, 60 per cent at 14 degrees, 10 per cent at 12 degrees, and no A FUSARIUM Wll.T OF PeAS 21 lutcclidii at 10 degrees C. 'Jhe other two strains gave 100 per cent infection at both 16 degrees and 14 degrees, 60 per cent and 40 per cent respectively at 12 degrees, and questionable symptoms on a few plants at 10 degrees C. During the 50 days, plants died from the disease at 16 and 14 degrees only ; at 12 degrees the course of the disease was extremely slow. Below 16 degrees under tlic conditions of this experiment, afifected plants showed the symptoms of stipule and leaflet roll, yellowing, slow withering, and defolia- tion, but no true wilting. Plants that showed doubtful symptoms at 10 degrees after 50 days were washed from the soil and their roots subjected to culture studies and micro.scopic examination. The wilt fungus was ob- served sparingly in the root cortex but not within the stele. The pathogen was isolated readily from these roots after surface sterilization in mercuric chloride. At 12 degrees the fungus was observed within the stele of the taproot, but not abundantly. The experiment indicates that 12 degrees is an approximate but not absolute minimum soil temperature for the develop- ment of pea wilt. Expcriiiioit V. — In a fifth experiment, Horsford peas were grown at 18, 20, 22, and 24 degrees C. in soil moculated with the same three cultures of the pea wilt fungus as in Experiment IV. In all three alike, the disease developed earlier at 22 degrees than at either higher or lower temperatures. Discussion of Experimental Results The relation of soil temperature to the development of pea wilt, determined by these experiments, is strikingly similar to the temperature relations of several closely allied diseases (Table V). In each of these there is a Tabic V. — Comparison of the Cardinal Tevipcraturcs for Development of Some Vascular Diseases Caused by S'pecies of Fiisarium Cardinal temperatures for disease Host Pathogen Authority Minimum Optimum Maximum °C. °C. °C. Tomato F. lycopersici 19 28-29 above 33 (4,6) Tobacco F. oxysporum Var. nicolianne 17-18 28-31 34-35 (12) Cabbage F. conglutinans 17 26-29 35 (33, 3.5) Flax F. lini 14 24-28 34-38 (19) Pea F. orthoceras var. pisi 10-12 21-22 above 30 critical temperature below which the host develops in infested soil and remains healthy ; a distinct intermediate optimal range ; and a maximum temperature at which the host escapes the disease. Compare Figure 1 with the corresponding figures of Tisdale {iZ) and of Jones and Tisdale (19). Pea wilt, however, shows certain individual peculiarities. The minimum or critical temperature of pea wilt is not sharply drawn. At temperatures progressively below the optimum the symptoms gradu- ally become less complete until at 12 degrees, actual wilting is never ob- served and the disease assumes more nearly the character of cabbage yel- lows. Such partial symptoms develop in the Horsford variety in 50 days even at 10 degrees, and an experiment reported below indicates that even 22 Wisconsin Research Bulletin 85 below that the pathogen may gain entrance into the root cortex. A sharply defined critical temperature probably does not exist, but an approxi- mate minimum for the development of disease lies between 10 and 12 degrees in the Horsford pea and probably slightly higher in the Alaska. No true maximum temperature can be defined, for the disease still de- velops, though tardily and feebly, at 30 degrees, the highest temperature at which peas will grow. In the Alaska this is very nearly a true maximum. A clearly defined optimum for the early and severe development of pea wilt lies at 21 to 22 degrees C, but the optimal range for severe develop- ment extends about from 18 to 24 degrees C. At 24 degrees the disease develops in fewer days than at 18 degrees, but at the lower temperature the plants become affected while they are in an earlier stage of their de- velopment. This is illustrated in Figure 11 in which comparative curves are drawn representing the relation of disease to temperature (a) after 20 days from planting and (h) when the plants at each temperature had developed an average of 5.8 nodes* each. This was the stage attained at • 15 18 21 24 27 SOIL TEMPERATURE DEGREES CENTIGRADE 40 ]|(".. 11.— INFLUENCE OF SOIL TEMPER.VTURE UPON Till. 1)I;VI:L0PM1-;NT OF FUSAHIUM WILT OF PEAS. The solid liiie represents numbers of plants affected 20 days after planting; the broken line, computed numbers of plants affected at the S.8 node stage. The latter curve is drawn from graphic interpolation of data on rate of node expansion at the different! soil temperatures. Horsford's Market Garden variety; soil tempeiatiirfr experiment TIL 21 degrees in 20 days. The retarding effect of low soil temperature is thus more marked upon the host than the disease. This interpretation helps to *Peas tend to blossom after developing a ralher constant iiunibt'r of nodes, a numJjer which is characteristic of the variety. In Experiment III above, dif- ferent plants of Alaska set their first blossom at from the tenth to the twelfth node regardless of soil temperature. Thus the number of nodes ex- posed at any time serves as a fair measure of the stage of development of the pea plant towards maturity. A FusARiuM Wir/r ok Peas 23 offset the error introduct'd in this study by the more rapid germination and early growth of plants at high temperatures. The cardinal temperatures of pea wilt are distinctly lower than those of the related diseases (Table V) : its critical temperature is below that of flax wilt which formerly stood lowest in the series, and its optimum is little above the minimum temperature for tomato wilt. Until this present case, the optimal temperatures for the wilt diseases all were near 28 degrees, and likewise the wilt fungi, when grown on potato-dextrose agar, made their most rapid growth at or near 28 degrees C. Thus it has been concluded, "that the influence of temperature upon the disease development (that of the Fusarium wilt disease in general) is primarily due to its direct effect on the parasite" (20, page 129). The optimum temperature for the pea wilt fungus on this agar, 27 to 30 degrees C, is in close agreement with the related fungi, but is distinctly above the optimal range of temperatures for pea wilt. Thus, while the slow development of the disease below the optimum may probably be explained in terms of the development of the fungus, some other explanation must be found for the retardation of disease development at temperatures most favorable for the fungus. This condi- tion may conveniently be regarded as one of resistance induced in the host by the operation of external factors. Jones and Tisdale (19) observed that flax wilt develops at lower tempera- tures than cabbage yellows and tomato wilt and, in correlation, that the healthy flax plant is distinctly a low temperature plant as compared with the tomato. Now the pea is another low temperature plant, as witnessed by the seasonal and geographical distribution of its culture, and pea wilt is favored by still lower temperatures than flax wilt. The healthy pea plant which fails to grow above 30 degrees is injured more quickly by high temperatures than even flax and cabbage which were grown by Jones and Tisdale (19) and by Tisdale (33) respectively at 38 degrees C. This cor- relation of temperature relations of hosts and diseases suggests the opera- tion of a causal factor, the nature of which is not known. The temperatures at which the disease declines in severity are those at which germination and early growth are most rapid, leading to early maturity, but also those at which the plant shows a distinct weakening in its later growth. Which of these factors is the more important cannot now be said. The latter fact, correlating w^ith the retardation of the disease in soil too dry for most vigorous growth of the plant as observed in this work with peas and in Clayton's (5) with tomato, appears suggestive. On the other hand, the weakening of pea growth and the decline of the disease at high temperatures may be the result of early maturity of the roots. Without further study, however, it is impossible to say in what way the high soil temperature renders the pea resistant to this disease. RELATION OF SOIL TEMPERATURE AND MOISTURE TO THE OCCURRENCE OF PEA WILT The soil temperature relations of tomato wilt, cabbage yellows, and flax wilt as summarized by Jones, Johnson, and Dickson (20), have been found to explain the seasonal and geographical distribution of disease. Cabbage yellows especially is limited in its occurrence to those localities and seasons 24 Wisconsin Research Bulletin 85 in which the temperature is suitably liigh, and is absent from the cooler cabbage districts. Pea wilt, however, being favored by lower temperatures than these diseases, would be expected to occur differently. If conditions aside from soil temperature are as favorable in the field as in the greenhouse, it should occur wherever the soil is warm enough for the Aphanomyces rootrot (15), and that disease occurs freely in northern Wisconsin and northern Michigan. A complete survey of northern Wisconsin has not been made, but wilt has been found in the most northerly districts searched. It is most destructive, however, in the older, more southerly pea areas of Wisconsin, a fact which may have resulted from other factors as much as from temperature rela- tions. ' , ! " 15 20 25 APRIL 30 15 20 25 30 k 9 14 UAY JUNE I'lC. 12.— MEAN J)A1LV SOIL TEMPERATURE. This was taken at a depth of two inclu's and recorded at MadLsoii, Wiscon- sin, I'roni April 15 lo June 29, i;>2,'> and 1!)20. In southern Wisconsin there is no indication of escape from the disease through early planting. The canning crop is planted here chiefly during late April and May. Comparison of the soil temperature records in Figure 12 with similar records presented by other workers for earlier years (28. ^3, 15, 16, and 20) reveals the fact that generally the mean daily soil temperature reaches the minimum for the development of pea wilt not later than the middle of May. Harvest rarely begins before the last week of June, and long before this time the soil temperature is practically optimal for the disease. Actually, in both 1925 and 1926 the disease was first ob- served in this area during the first week of June. Escape from the disease through early planting would seem possible, therefore, only in case peas planted in cool soil were to acquire a degree of resistance which remains effective when the soil later becomes warm enough to favor the disease. A FusARiuM Wjlt ok Pkas 25 TIic following cxpcrinu'iit was dcsifiiitd to tost tliis possibilit.v. Four lots of Horsford peas were planted in infested soil at intervals of one week, and were held at the soil temperature of 8 to 10 degrees C until the first planting was five weeks old. The oldest plants were in the five node stage. Three inoculated and one control can of each age group were then transferred to the soil temperature of 22 degrees, and two inoculated and rne control to 18 degrees C. where they were maintained to the end of the experiment. Tabic VI. — Comparative Rates of Development of Wilt S'yi)iptoms in Plants of Horsford's Market Garden Peas, in Soil Inoculated With Pure Culture of the Wilt Fungus, Grown Two to Five Weeks at a Soil Tem- perature of 8-10 Degrees and Then Transferred Abruptly to the Tempera- tures of 18 and 22 Degrees as Indicated; Air Temperature 15-20 De- grees C. Soil temperature 18° C. Soil temperature 22° C. Age in days at Total Plants affected Total Plants affected time temp- plants after plants after erature was raised 9 11 13 9 11 13 days days days days days days No. % % % No. % % % 14 20 0 0 0 30 0 13 23 21 20 10 30 50 28 7 43 68 28 20 40 50 80 30 71 90 90 .3.5 20 50 80 100 29 86 97 100 From the sixth day, when first symptoms appeared in the oldest plants, to the end of the experiment, the older plants always showed more advanced stages of the disease than did the younger. The differences were most apparent early in the experiment as shown in Table VI. This experiment does not prove that the oldest plants were the most susceptible (although that possibility merits consideration) for the fungus had already begun to enter the root cortex at the low temperature of 8 to 10 degrees C. The practical importance of the experiment, however, rests on the fact tliat this low temperature is below that which prevails in the fields in southern Wis- consin after the earliest plantings. Field observations are in direct agree- ment with this experiment, for the earliest development of the disease in this area has been found in the earliest planted peas. The single experiment on the influence of soil moisture upon the develop- ment of pea wilt suggested a slightly favoring influence of wet soil a.s compared with dry, but this has not been apparent in the field. Abundant field evidence shows that the w'ilt fungus develops aggressively in both wet and dry soils within the extremes which favor grow-th of the pea, but suggests that soils which remain excessively wet for long periods may bs less favorable for long persistence of the fungus. 26 Wisconsin Reseakcii FUh-U'Itin 3 24 38 34 19 06 22 14 64 Clav loams and clavs 14 1 7 15 3 20 10 3 30 3 1 33 2 1 50 Totals of all soils 234 12 o 159 22 14 116 39 34 53 22 42 37 21 57 28 Wisconsin Research Bulletin 85 soils in the older and most thoroughly infested canning factory districts. The character of a soil which appeared to influence the disease most significantly is its content of organic matter. Soils naturally rich in organic matter appear to favor more rapid enlargement of infested areas and more uniformly severe injury to plants than poorer soils. The plowing under of large amounts of raw vegetable matter seems also to favo"- the disease, probably through increase of the fungus in the soil. RESISTANCE TO FUSARIUM WILT Wisconsin pea growers have recognized for a long time that certain varieties, notab y Yellow (White) Admiral, Green Admiral, and the newer Horai, are less subject to failure from disease than other canners' varieties. In fields where other varieties fail completely these hardy sorts sometimes yield very well ; at other times, however, they too may fail. This irregular performance has prevented, up to the present, the widespread planting of these hardy peas on infested fields and has hampered progress in the develop- ment of other hardy varieties better adapted to the needs of the canner. When several pea root diseases were found among the causes of pea failures it was expected that these hardy varieties would prove highly re- sistant to one or more of them, but, until this Fusarium wilt, this expectation has not been realized. Turresson {2>7), Jones (14), and Gilchrist (9), found minor differences in varietal susceptibility to footrot diseases, but not sufficient to account for the differences sometimes seen in field trials on naturally infested soil. Important degrees of resistance to the Aphanomyces rootrot were reported by Jones and Drechsler (15) and later by Jones and Linford (16) on the basis of field observations, but these workers recognized that even the most resistant varieties might fail completely ander conditions which favor severe development of rootrot. Later, after failing to detect significant resistance to rootrot itself, Jones (17) concluded that the resist- ance observed under field conditions is probably less resistance to rootrot than to secondary invasion of the vascular system following rootrot, and to Fusarium wilt. Relative Varietal Susceptibility The first evidence of resistance to Fusarium wilt of peas came from the observation in 1924 that the Green Admiral remained free from this disease in localities where other varieties were attacked. Again in 1925 this variety and also Pedigreed Extra Early remained unaffected in wilt infested dis- tricts, even when directly adjoining other varieties which were dying from this disease. During these two seasons, all the more common canners' varieties were seen to be susceptible to wilt (see list below). Resistaace has been found so constantly associated with the related vascular diseases of other plants (2, 26, 18. 27) that with this observational evidence as a basis, greenhouse and field studies were undertaken to determine the extent and importance of resistance to this disease. In the first experimental demonstration of resistance to pea wilt in January, 1926, the four varieties Alaska, Horsford, Green Admiral, and A FusARiuM Wilt of Peas 29 Horal, were planted in suil inoculated with pure cultures of the wilt fungus and were grown at a soil temperature favorable for the disease. All the Horsford plants died early from wilt, most of the Alaska plants died but a few remained healthy, while all the Horal and Green Admiral plants were free from wilt symptoms at the end of 46 days. In a more extensive test, peas of both early and late varieties were grown in a bench of freshly sterilized and inoculated soil. Wilt developed tardily in this planting, but striking differences were observed in varietal susceptibility. The late varieties especially showed clear-cut resistance and susceptibility. Both Green Admiral and Yellow Admiral remained entirely free from any evidence of the disease, while all plants of Horsford and Perfection were affected. Only one of the 23 Horsford plants set seed ; and only three of the 71 Perfection even blossomed, and none set seed. Susceptible early varieties did not develop the disease as early and severely as did the Horsford and Perfection, and, due to their earlier maturity, almost all plants had begun to blossom when symptoms first appeared. In the irregular wilting that followed it was sometimes difficult to distinguish between death from disease and drying from maturity. Varietal dift'erences were therefore less clearly marked than in the late varieties. Eight common stocks of Alaska and two lots of seed obtained from Dr. Wilbur Brotherton Jr., selected from a cross of Rice's 330 on Surprise, proved susceptible, with more than half of the plants dying from the disease. Complete freedom from the disease, indicating high resistance, .•i'J: w: 'iT^ . S^- 1^^^ -'*/':■,/ H^u- •^ -^^ 9r i^ ■ i.' ' * ^a\ m\ -v r^-«e Sf" -v^f%/' ^"' ''■-f^ii- BhP^*^ ■v -** - .- P4k^'' fei""' " ' %v^ FIG. 14.— WILT RESISTANCE IN PURE CULTURE INOCULATION PLOT, 1926. Horal (left) and Green Admiral (right) show no evidence of injury from ■wilt, but the susceptible Horsford variety (center) has been killed by the disease. 30 Wisconsin Research Bulletin 85 was found in two early peas, one a strain of Alaska selected for disease resistance by Prof. C. E. Temple of Maryland, and the other a commercial stock of Pedigreed Extra Early. In the spring of 1926, a resistance test was conducted in a garden plot of well drained Miami silt loam at Madison where peas had not been grown before, again using pure culture inoculation to avoid confusion with rootrot and footrot so commonly met with on naturally infested soil. Eight kilo- grams of inoculum, consisting of soil inoculated in the greenhouse with a mixture of three different strains of the wilt fungus, were distributed in tach 21 foot row before the seed was placed. One row each was planted with the following varieties in the order named : Horal, Horsford, Green Admiral, Pedigreed Extra Early, Brotherton's Selection "A," Alaska, Brotherton's Selection "B," and Rice's 330. Another planting which ad- joined this directly and was inoculated in the same manner (described in the following section) included numerous separate plantings of Horsford and Perfection peas. Wilt developed early and with uniform severity in Horsford (Fig. 14) and Perfection, and more tardily and less regularly in Alaska. By July 6, all Horsford plants were dead and 61 per cent of the Alaskas were affected or dead. Many more Alaska plants died before maturity, but all the other varieties remained uniformly free from the disease to maturity. These varied tests indicate that the more common canners' varieties of peas are either highly susceptible or highly resistant to wilt, with the exception of Alaska which appears to be somewhat intermediate and to contain some highly resistant individuals. The following lists of susceptible and resistant varieties are based upon : (A), field observations; (B), plot trials on naturally infested soil; (C), pure culture inoculation trials in the open ground; and (D), pure culture inoculation of steamed soil in the greenhouse. The plot trials on naturally infested soil are reported below. Varieties listed as resistant have shown no more than a trace of susceptible individuals. Susceptible Resistant Horsford's Market Garden A,B,C,D. Green Admiral A,B,C,D. Perfection A,B,C,D. Yellow (White) Admiral.... B, D. Advancer A. Horal A,B,C,D. Gem A. Rogers' K B. Rice's 13 A. Rice's 330 B,C. Ashford D Pedigreed Extra Early A,B,C,D. Badger A, D. Temple's Alaska D. Alaska A,B,C.D. Brotherton's Sel. "A" B. Winner A. Brotherton's Sel. "B" B. Resistant Peas Obtained by Selection Plants which remain free from wilt symptoms on wilt infested soil are found freely in most commercial stocks of Alaska and sparingly in com- mercial wilt-susceptible late varieties. That such plants might be inherently resistant individuals from which new resistant strains or varieties might A FusARiuM Wilt of Peas 31 Table VIII. — Results of Test for Wilt Resistance in Progenies of Single Plants Selected for Resistance in 1925; Inoculated at Time of Planting With a Mixture of Three Cultures of the Pea Wilt Fungus; Madison, Wisconsin, 1926 Variety Total plants Plants wilted June 22 July 16 No. 157 23 183 558 161 547 No. % 65 41 No. '5 183 67 1.59 276 % Selections Horsford Commercial Selections Perfection Commercial Selections 0 155 18 129 163 0 85 3 80 30 0 100 12 99 50 be developed was suggested by the findings of numerous other investigators working with related vascular diseases (2, 26, 18, 27). To test this possibility, numerous single-plant selections were made from diseased fiields of Horsford and Perfection peas at Columbus and Chilton, Wisconsin in 1925. After gathering seed separately from the most promis- ing individuals, a mass selection of each variety was made from the re- mainder. Survivors were more numerous and more vigorous in Horsford than Perfection. IK.. l.">.— WILT RESISTANCE IN THE PROGENIES OF SINGLE PLANTS SELECTED FOR RESISTANCE OUT OF THE SUSCEPTIRLE VARIETY, HOJiSFORD. This photograph, which ^^as taken at right angles to the direction of the rnw.*;, shows healthy plants in the resistant progenies at left and right, and dead plants of coniniercial Horsford (indicated by the bhick crosses) at Hie center. See Tal>le VIII and IX. 2>2 Wisconsin Research Bulletin 85 The single-plant selections were planted at Madison, May 12, 1926, in a plot adjoining the last described above and inoculated in the same way. Four single-plant selections of Alaska from the greenhouse were also in- cluded in this planting. Ten seeds (rarely fewer) from each selection were allowed one foot of space in the row, and every fifth foot was planted with ten seeds of the commercial variety from which the selection had been made. Wilt appeared early and uniformly throughout the plot in the commercial variety control plantings. By June 22, many of these plants were dead, and on July 16 very few of them remained alive (Table VIII). Among the selected progenies from Horsford and Perfection, only a few plants w^ere affected as early as the controls. As the season advanced some entire pro- genies wilted uniformly, and several of them indicated, by the tardiness of their collapse, an intermediate degree of resistance. A few others contained some plants that wilted and others that remained free from the disease. As shown in Table IX, however, all selections from Alaska, 86 per cent of those from Horsford, and 45 per cent of those from Perfection remained uniformly free from wilt (Fig. 15). Table IX.— Progenies of Single Plants, Selected for Resistance to Wilt in 1925, Grouped According to Their Susceptibility to Wilt as Determined by Trials in Pure Culture Inoculation Plot, Madison, Wisconsin, 1926 {Com- pare With Table VIII Parent variety of selection Single-plant progenies tested Progenies with — .411 plants affected Some plants affected No plants affected Alaska _ No. 4 71 60 No. 0 7 29 % 0 10 48 No. 0 3 4 % 0 4 7 No. 4 61 27 % 100 86 Perfection 45 Parallel with this test, thirteen selections from Horsford and seven from Perfection, with appropriate controls, were planted in naturally infested soil which contained a mixed flora of pea parasites. The Aphanomyces rootrot developed tardily and caused little injury. Wilt was therefore the chief disease present but it was less uniformly severe than in the inoculated plot. All plants of commercial Horsford died from wilt. A single selection from Horsford and three from Perfection developed the disease: these same four progenies wilted in the inoculated plot. The remaining progenies developed no wilt, in full agreement with their behavior in the inoculated plot. The mass selections were tested on naturally infested soil. In the selec- tion from Perfection only two per cent of the plants survived, but in that from Horsford, 85 per cent of the plants remained free from wilt to maturity. It was thus demonstrated that although some of the plants which survive disease on wilt infested soil may do .so through the chance escape of infection, many of them are inherently resistant and may serve as a basis for developing new wilt resistant peas. A FusARiuM Wilt of Peas 35 With the possible exception of one Alaska progeny, all of these selections differed from the parental varieties in some gross morphological characters, and most of them were distinct from all varieties now grown for canning. In general they were off-type plants which, when occurring in canners' peas, are called "rogues." These resistant off-type plants appear to have origin- ated chiefly other than as admixtures from established resistant varieties. In the Alaska pea, resistance is associated with several well marked rogue types, and the greater frequency of resistant individuals here is associated with much more abundant occurrence of off-type plants than in the late varieties. Whether all off-type plants in Alaska are resistant is not known, but in other varieties they are not. In the resistant Horal, susceptible indi- vidual plants occur which differ in other respects from the true Horal. The majority of resistant progenies obtained in the 1925 selections were inferior in type to the resistant varieties at present available, but these re- sults show that by selection resistant stocks may be obtained which may be used either indirectly through hj'bridization or directly through propaga- tion and possible reselection in the development of new resistant varieties of better quality. Resistant individuals selected from susceptible varieties are not always of poor type. Field observations in 1926 indicated that the Alcross, an especi- ally uniform strain of Alaska developed by Prof. E. J. Delwiche, contained about 50 per cent of resistant plants of excellent type. Greenhouse and plot tests conducted since that time by Mr. E. J. Renard indicate that these resistant plants breed true for resistance and are indistinguishable in other respects from the susceptible plants within this variety. Degree of Resistance to Wilt In the field and greenhouse trials reported above, the wilt resistant vari- eties sbciwed a degree of resistance which amounted, practically, to immun- ity. They did not develop wilt symptoms and showed no apparent reduction of vigor. Working with cabbage yellows, Tisdale (33) and Tims (35), found that resistance to that disease breaks down at the temperatures most favorable for the disease. Resistance to pea wilt shows very little break down of this sort, as shown by the following experiments. Three resistant varieties, Horal, Green Admiral, and Pedigreed Extra Early, were grown in comparison with the three susceptible varieties, Hors- ford, Perfection, and Alaska, each in three lots of soil inoculated separately with three strains of the pea wilt fungus. All were maintained alike at the soil temperature of 21 degrees C.. most favorable for rapid development of disease. The susceptible varieties all failed utterly except a few resist- ant individual plants in the Alaska variety, but at the close of the experi- ment when the Extra Early peas were just maturing, none of the resistant varieties had developed wilt, although a few lower leaves had turned yellow and fallen. In another test, Horal peas were grown 50 days in inoculated soil at 16, 21, and 26 degrees C. without developing the disease. Even at 21 degrees the plants remained as vigorous as the controls grown in steamed soil, but showed some injury in the loss of lower leaves. The fungus was found 34 Wisconsin Research Bulletin 85 sparingly in the cortex of these roots, but not in the stele of the taproot. Besides such nearly complete resistance there appear to be some intermedi- ate degrees of resistance, as in a few of the single-plant progenies described above and in the Alaska variety. In this early variety the disease requires several days longer to develop and is more sharply inhibited at high and low extremes of soil temperature than in the late variety, Horsford. The Alaska, developing more rapidly, approaches maturity more closely before the disease appears. Such resistance is slight in comparison with the resist- ance described above, and is not to be confused with the fully resistant plants that occur freely in common stocks of this variety. Practical Usefulness of Wilt-Resistant Varieties Since the varieties which are almost immune to wilt are not resistant to the Aphanomyces rootrot it appeared probable that the variable degree of resistance shown in former field trials might have resulted from variation in the relative severities of these two diseases. To test this hypothesis and to determine the extent to which the value of resistant peas may be predicted for a known piece of infested soil, varietal tests were conducted during 1926 in five localities in separate counties of Wisconsin. On the basis of observations made in 1924 and 1925, fields were selected where rootrot had occurred alone, where rootrot and wilt had occurred to- gether, and where wilt alone had been present. In the selected fields'' the following varieties" were planted in drill width strips, usually in duplicate or triplicate : Alaska, Rice's 330, Perfection, Horal, and Green Admiral. The trial grounds in two localities were thought to be free from rootrot, but that disease appeared in all five and wilt occurred in all except at Owen. There rootrot alone was devastatingly severe as it had been in 1924 and Tabic X. — Summary of Resistance Trials Conducted on Naturally Infested Pea Fields in Different Localities During^ 1926 , Diseases" Diseases* Date of Resistance Locality Soil type recognized 1924 or 192,5 recognized 1926 planting demonstrated Columbus Miami silt loam rootrot'' rootrot and wilt May 7 Marked -■ Madison Carrington rootrot rootrot ^L■lv 1 1 Marked silt loam and wilt and wilt Ripen Carrineton silt loam wilt rootrot and wilt May 20 Marked Owen Colby silt loam rootrot rootrot May 22 • None; all varieties failed Chilton Kewaunee clay loam wilt rootrot and wilt May 27 Marked "Other diseases of minor iinportanre were present in some or all of these plots, including Ascochyta blight, Fusarium footrot, bacterial blight, and downy mildew, but none were suffi- ciently prevalent to have influenced in any important way the outcome of these trials. ''The rootrot caused by Aphanomi/re.s euteiches, identified microscopically by the presence of characteristic oospores in decayed root cortex. "Varieties susceptible to wilt failed; those resistant to wilt matured a fair crop. 'The trial grounds, except at Madison, ^ve^e provided gratis by the local canning companies. "At Coluinl)iis, Yellow Admiral was planted instead of Green Admiral, Rice's 330 was omitted, and Horsford was added. A FUSAKIUM W'U.T OF PeAS 35 1925, and the test varieties all failed utterl}- with no evidence of resistance in any of them. In the other localities, however, there were clearly defined differences between varieties. In each instance, Aphanomyces caused earl\- and extensive decortication of the roots of resistant and susceptible peas, but the plants were little harmed by this. Wilt later destroyed the suscepti- l)le varieties, but the wilt-resistant peas remained vigorous to maturity. See Table X. In each of the trials where rootrot and wilt both occured, it was possible to find plants of susceptible varieties that showed advanced stages of wilt while their roots were still free from rootrot symptoms. It was clear from this that wilt is not to be regarded as secondary to rootrot. To determine any influence of possible resistance to rootrot upon the outcome of these trials, search was made for varietal diflferences in earliness and thoroughness of root decay. Early in the season, plants of each Table XI. — Percentages of Plants With Roots Infected With Aphanomyces Rootrot and of Plants Dead After Specified Intervals, Based on Counts of 100 and 300 Plants Respectively, for Each of Five Varieties of Peas in Resistance Trials at Ripon and Chilton, Wisconsin, 1926 Varietal Chilton Ripon Variety Plants with root- Plants dead Plants with root- Plants dead type* rot 22 days after 42 days after rot 28 days after 50 days after plantingb planting" planting^ plantings % % % % Rice's 330 E, R 69 6.6 97 21.3 Alaska E, S 77 82.0 99 83.3 Horal L, R 54 2.0 91 11.7 Perfection L, S 77 98.3 100 99.7 Green Admiral L, R 73 7.0 98 30.7 »E, early;- L, late; R, resistant to wilt in pure culture inoculation tests; S, susceptibk. •iPlants in 5-7 node stage. "=Pod3 set on Rice's 330; Horal and Green Admiral in 11-14 node stage. •^Plants in 7-9 node stage. •Rice's 330 four days till fit for canning; Horal and Green Admiral in full blossom. variety were removed from the soil with care to preserve their roots, and counts were made of the numbers of infected and clean root systems. Table XI shows the results of such counts in two representative trials. Jones (17) had found that Horal sometimes but not always becomes in- fected with Aphanomyces slightly more tardily than Horsford. In three of the four trials where this count was made the Horal showed this same tendency, but the differences between it and susceptible varieties were slight, and no correspondingly delayed infection could be found in the other resistant peas. This possibly indicates a minor degree of resistance to rootrot in Horal, but too minor to be significant in interpreting the out- come of these trials. The formation of periderm, described by Jones (17) as a possible factor in resistance to rootrot, was found in resistant varieties but not to any extent in the susceptible in all of the trial grounds except at Owen. There, where rootrot occurred alone, periderm was present in susceptible varieties as well as resistant. Late varieties had developed it somewhat more strongly than the early ones. At Madison, likewise, in plots where rootrot occurred 36 Wisconsin Research Bulletin 85 alone, periderm was developed almost equally in susceptible and resistant ^arieties, but where wilt and rootrot occurred together it was limited to the resistant. The formation of periderm thus appears to be a response to cortical decay which may be prevented, in varieties susceptible to wilt, by a simultaneous attack of wilt which weakens the plants before the re- sponse can begin. A well developed periderm may enable plants to with- stand Aphanomyces rootrot, but its formation, under common field condi- tions, appears to be more an indication of resistance to wilt than to rootrot. In all of the trials except at Owen the difference between resistant and susceptible varieties was the difference between a fairly satisfactory crop and complete failure. Where the two diseases occurred together, the resist- ant peas were weakened in proportion to the severity of rootrot and did not yield as well as on clean soil, but plants were killed outright only in wet pockets. The difference between resistance and susceptibility was greater in all cases in late than early varieties. Thinning of the stand from irregu- lar dying of scattered plants was one of the chief weaknesses of resistant varieties in two of the trials. Such dying may have been due in part to the rootrot fungus itself but it appeared to be chiefly attributable to the two footrot fungi, Fusarmm niartii var. pisi and Mycosphacrclla piiwdes micro form, which were isolated many times from such plants. Fifteen acres of Horal peas adjoining the trial at Columbus, planted on three successive dates, yielded very satisfactorily in all three plantings even though the roots were extensively decorticated at an early stage. The outcome of these trials provides a fair basis for judging the practical value and the limits of usefulness of wilt resistance in peas. Resistance to this disease now appears to have been the chief factor in the observed resistance of certain varieties formerly reported, at least in Wisconsin. Wilt has been found with rootrot in the trial grounds in which several of the earlier examples of resistance, supposedly to rootrot, were observed. The behavior of resistant peas in the field, which was formerly baffling in its irregularity, appears to be readily understood on this basis. If wilt occurred by itself, complete relief could be attained by planting resistant varieties, but this condition is seldom found in old pea fields. Rootrot and footrot usually occur with it and may either destroy a crop independently or thin out the stand. The value of wilt resistance is thus diminished in proportion to the severity of these other diseases, being great- est where they are absent, and negligible where they are severe enough to destroy the crop. Where rootrot occurs destructively, as in the Colby silt loam area of north-central Wisconsin, there appears to be nothing gained by the use of wilt resistant peas. This is true generally of very wet soil types and poorly drained fields in other localities as well. For well drained soils in southern Wisconsin, and elsewhere where wilt is known to be the major factor among pea diseases, they may be recommended. Fven here, Viowever, they may fail from rootrot in wet years. CONTROL OF FUSARIUM WILT OF PEAS Recommendations for the control of pea wilt follow closely those already given (15, 16) for the control of rootrot. Obviously they must take into A FusARiuM Wilt of I-'eas ' i7 account the control of other diseases which may occur destructively with wilt. The following brief summary seems, therefore, sufficient for present purposes when considered in connection with the previous publications cited. Recommendations Rotate crops systematically with as long an interval between peas as practicable, preferably as long as five or six years. Avoid planting peas where failures from disease have occurred. Avoid transfer of soil from infested to clean fields. Avoid planting seed grown on infested fields. Carefully dispose of pea vines, particularly from infested fields. Vines should be cured in silos or silage stacks; refuse should not be used as manure where peas are even to be grown. Resistant varieties may be planted where wilt occurs by itself or where other diseases are of minor iniportante. They are not aependable where lootrot occurs severely with \\ d\.. In the control of wilt, rotation of crops is advisable in spite of the long persistence of the parasite in infested soil, and for the control of other pea diseases it is an essential precaution. In the case of wilt, even if rotation should not permit more crops to be grown than will continuous cropping before the disease becomes troublesome, it will at least delay the establish- ment of centers of infestation from which the parasite may be spread to other fields and farms. Cannery districts in which systematic rotation has been followed from the beginning of pea culture are today avoiding much of the trouble from wilt and some other diseases which is being experienced in neighboring districts w'here rotation has been practiced less diligently. Losses can be very largely avoided at present by careful selection of fields for pea culture to avoid planting where the crop has once failed. Except in the oldest and most intensive pea growing areas there is still sufficient disease-free land for the production of the pea crop. In such localities the need for suitable wilt-resistant varieties is acute. Elsewhere, if cannery field agents and fanners would keep records of pea failures and, upon the appearance of small amounts of disease, abandon pea culture in infested fields during two or three rotation cycles, important losses could be almost eliminated. GENERAL DISCUSSION The recognition of Fusarium wilt marks an important advance towards an understanding of those factors which cause peas to fail when they are planted repeatedly in the same fields. Instead of complicating the problem of pea disease control, the finding of this disease has actually simplified an already complex problem through analysis, and has opened the w^ay to the elimination of considerable loss. Wilt of peas is probably not a new- disease in Wisconsin : its present occurrence suggests that it has been present for years, obscured by its frequent association with the Aphanomyces rootrot. Where the two diseases have occurred together, rootrot has been regarded as the cause of the total injury, and thus has been considered of greater economic importance in 38 ' Wisconsin Research Bulletin 85 wilt-infested localities than it really is. It is now evident that the severe occurrence of rootrot is restricted to wet soils and wet seasons even more closelj- than was reported in 1925 (16). At the time of that stud}' the two diseases were still largel}' confused and the injur}- from their combined attack was then attributed to rootrot alone. It is now apparent that on medium to light, well drained soils, rootrot may lead to extensive root decortication without causing serious losses if it is not complicated by other diseases. Special encouragement lies in the significance of wilt-resistant varieties. Their adoption by the canning industry promises to bring wilt fully under control, but since wilt is only one of several factors which jointly cause pea failures, the problem of reducing the damage caused by the other diseases must be attacked with renewed vigor. In extremely wet soils which favor the greatest severity of rootrot, the avoidance of crop failures appears to depend entirely upon withholding pea culture, but elsewhere, where rootrot and related diseases are less destructive, any slight reduction in their severity may be of the utmost importance together with complete control of wilt in permitting the continued production of peas. One of the possible means of accomplishing this which merits close attention is the com- bining of resistance to these other diseases with resistance to wilt. While wilt-resistance appears to have been the chief factor in the disease resistance observed earlier in Wisconsin and northern Michigan, there may also be some degree of resistance to rootrot itself. This was not found by Jones (17) in significant degree in the greenhouse, but it has not yet been searched for adequately in the field. The earlier varietal trials (15, 17) in which resistance was sought, were conducted chiefly or entirely where rootrot and wilt were both present, as indicated by the recent isolation of the wilt fungus from plants grown in these trial grounds, and since the presence of wilt would completely obscure any minor degree of resistance to rootrot that might have been present, the failure to detect such resistance in these trials is not conclusive proof of its absence. Minor degrees of resistance to the footrot diseases have been reported (37, 14. 9) ; they merit further study. Furthermore there is the possibility, not yet tested experi- mentally, of resistance to secondary invasion following rootrot (17). If such vascular invasion does occur, resistance to it might be expected to correlate with resistance to wilt. It is probable from earlier investigations that any varietal differences in susceptibility to rootrot or to footrot or to secondary invasion following these will be slight, but any which may be found, even if too slight to be important by itself, may be of considerable value when combined with resistance to wilt. Furthermore, any means other than the developing of resistant peas that may reduce the losses catised by these other diseases will expand markedly the scope fif usefulness of wilt-resistant varieties. In addition to its practical importance, pea wilt is of considerable interest biologically. Comparison with related vascular diseases of other plants caused by allied species of Fusarium reveals several important points of similarity and contrast. A FusARiuM Wilt uk Peas 39 The syinptums of pea wilt are somewhat intermediate between tliose of cabbage yellows and tomato wilt, with more actual wilting than in the cabbage disease but less than in the tomato. Wilting is not the most char- acteristic symptom and may, indeed, be entirely absent, as at low soil tem- ])eratures where affected plants become yellow and wither slowly, leaf by leaf. At favorable temperatures, the earliest symptoms are directly opposed to wilting and include increased rigidity of the entire plant, rolling of the leaflets and stipules while still turgid, and an increase in diameter of the lower internodes of the stem. These preliminary symptoms are more characteristic of the disease than is wilting alone. The distribution of the pea wilt fungus through the aerial parts of the plant is relatively limited in comparison with tomato wilt and cabbage yellows. In the pea the upper half of the stem and even the lowest leaves are generally free from mycelium. On the other hand, the fungus appears to accumulate in the xylem vessels in the upper subterranean parts more abundantly than in the other diseases. While this fungus is characteristically an invader of the xylem vessels, as are the allied parasites, it appears less closely limited than some. It damages the cambium rather early and in- \ades the phloem and pericycle at numerous points well in advance of death of the plant. The earliest symptoms appear at a time when the fungus occurs in the •Stele in very small amounts, indicating that the pea is highly susceptible to injury. Actual plugging of the vessels with mycelium to such an extent as mechanically to obstruct the passage of water cannot be the cause of any but the final stages of the disease. The most characteristic preliminary symptoms are probably caused by the action on the host cells of injurious products of fungous metabolism. In its relation to soil temperature, pea wilt is of particular interest because of the detailed study given several related vascular diseases by other investigators who have found remarkable similarity between them. The disease-soil-temperature curve of pea wilt is strikingly similar to those of the allied diseases, but differs notably in that it is transposed to the low temperature side. The most rapid and severe development of this disease occurs at temperatures (21 to 22 degrees C.) several degrees below the optimal range of the other diseases. No detailed study of the influence of temperature upon growth of the parasite on diverse media has been at- tempted, but on potato-dextrose agar its growth-temperature relation is closely similar to that of the related vascular Fusarium species. In these other diseases, the soil temperature most favorable for rapid development of disease has been almost precisely the same as for most rapid linear growth of the fungus on this substratum. In the pea, however, the optimum for the disease lies distinctly below that for the fungus, and the disease is much reduced in .severity at the temperatures most favorable for the fungus on this medium. This case appears particularly favorable for a detailed study of the metabolism and temperature relations of the fungus in relation to cr)mposition and physiology of the host. 40 Wisconsin Research Bulletin 85 SUMMARY A Fusarium wilt of peas which was first observed during the summer of 1924 occurs widely in Wisconsin and at least localh- in >tlichigan and Indiana. In some localities this is the most destructive disease which affects peas grown for canning. In Wisconsin as a whole it is second in importance only to the Aphanomyces rootrot. Fusarium wilt of peas is probably not new. It appears to have been present for several years, obscured by its frequent association with other diseases. This disease typically occurs in scattered patches, approximately circular in outline, but it may infest entire fields uniformly. In infested areas the crop is generally destroyed completely. Plants affected with wilt show symptoms which are diagnostically specific from other pea diseases in Wisconsin. Some of them are characteristic of vascular diseases of other crops caused by species of Fusarium, while others appear to be associated with no related diseases. One fungus, described here as Fusarium orthoceras App. and Wr. var. pisi (n. var.), is associated with this disease throughout its range. This alone, of the fungi tested, causes the disease. The pathogen invades chiefly the xylem of the roots and the lower half of the stem, but other tissues of the stele may be invaded, and the mycelium may occur sparingly in the root cortex. This is more strictly vascular than other diseases of the pea caused by species of Fusarium. Early symptoms of wilt appear before the pathogen is present in large quantities within the stele. Later, the mycelium may accumulate abundantly in the xylem vessels. The most significant phases of development of this disease cannot be attributed to obstruction of vessels by mycelium. Growth of the pea wilt fungus on potato-dextrose agar may occur from 6.5-8 degrees to 34.5-35 degrees C, but it is most rapid at 27 to 30 degrees C. This is in close agreement with the fungi which cause related vascular diseases. This fungus is pathogenic in some varieties of Pismn safiznim and in Vicia gic/antca, and, weakly, in V. faba. No hosts have yet been found out- side of Pisuni and Vicia. Dissemination of the pea wilt fungus is accomplished by any agencies which transfer infested soil or refuse. In Wisconsin pea culture one of the most important of these is the improper disposal of pea vines from infested fields. The influence of soil temperature upon growth of the pea plant and upon development of wilt has been studied in Wisconsin soil temperature tanks. Peas grow well from 8 to near 30 degrees C. Most rapid germination and early growth occurs from 24 to 27 degrees C, but 18 to 21 degrees C. is near the optimum soil temperature for growth of the healthy pea plant over a long period. Pea wilt may develop from 10-12 to 30 degrees C, but severe injury occurs chiefly between IS and 27 degrees C. In less susceptible varieties, with less aggressively pathogenic strains of the parasite, or under conditions A FusARiuM Wilt of Peas 41 otherwise less favorable for wilt, the total range of favoring temperatures is shortened at both extremes. At 21 to 22 degrees C. wilt develops in fewer days and affects plants w'hile they are developmentally younger than at any other temperature. At temperatures slightly above this optimum, wilt develops more rapidly but in fewer plants than at correspondingly sub-optimal temperatures. At the lower temperatures, plants are affected more uniformly and while they are in an earlier stage of their development. Below 16 degrees C, actual wilting is not a characteristic symptom. The cardinal temperatures for pea wilt are lower than for the relaterl diseases of other plants. The optimum for this disease is below that for rapid growth of the pathogen in pure culture, and very near that for growth of the healthy pea plant. The retardation of wilt at temperatures which favor fastest growth of the parasite in pure culture probably indicates a condition of temporarily induced resistance in the host. Soil moisture has less influence than temperature upon pea wilt. Wet soil slightly favors early . development of symptoms, but drier soil favors more rapid death of affected plants. In southern Wisconsin early planting favors the earliest seasonal develop- ment: of this disease. Pea seedlings grown to the five node stage in soil too cool for the disease acquire no resistance that remains effective when the soil temperature is later raised. In Wisconsin, low soil temperature may not be expected to permit escape from the disease in late susceptible varieties. Wilt occurs in the most northerly localities yet searched. Severe losses from wilt occur chiefly where peas have been grown repeatedly. The wilt fungus may persist in the soil almost indefinitely. Soil type does not restrict importantly the occurrence of wilt, but a high content of organic matter favors most severe development of the disease. Varietal differences in susceptibility to wilt are clearly defined. The leading varieties of canning peas are subject to complete failure when planted on wilt-infested soil. Several varieties including Green Admiral, Yellow Admiral, Horal, and Rice's 330, are strongly resistant to wilt. Susceptible varieties usually contain a few plants that are highly resistant to wilt. Selection of such plants has yielded numerous resistant progenies. Such progenies are chiefly not of true varietal type but the}' include valuable parent stock for hybridization in the production of new resistant peas. Selection by itself promises, in some instances, to yield resistant peas of good quality. Wilt-resistant peas remain free from wilt symptoms under conditions very favorable for the disease in susceptible varieties. The fungus may invade their rootlets but it does not progress freely into their vascular systems. Resistance to pea root troubles, formerly observed in practical field trials in Wisconsin, appears to have been chiefly resistance to wilt. Wilt-resistant peas are not significantly resistant to rootrot, and the variations in resist- ance formerly observed probably resulted from variability in the relative 42 Wisconsin Research LJulletix 85 importance of wilt and rootrot. \Vhere rootrot and other diseases occur with wiU they reduce the yield of wilt-resistant peas in proportion to their severity. Rootrot is less fre- quently destructive on well drained, light to medium light soils, tiian form- erly supposed. On such soils, the planting of wilt-resistaiit peas reduces losses from disease by eliminating wilt. Further reduction of losses depends upon the control of the rootrot and footrot diseases. Slight degrees of resistance to these diseases may be of much value if combined with resistance to wilt. The resistant varieties in use by canners at present are less desirable in type and quality than the best susceptible varieties, and therefore can not be recommended for use except under conditions of known wilt infestation. The possibility of developing new resistant peas has been demonstrated in this work, and it appears to be only a matter of time until, through the perfection of suitable new varieties, the general adoption of wilt-resistant peas for intensive culture will effectively control this disease. Brief recommendations are formulated for present guidance in the con- trol of wilt. Comparison of pea wilt with related diseases of other crops reveals in- teresting and biologically significant points of similarity and contrast, par- ticularly in the production of symptoms, in pathological histology, and in I elation to temperature. A Fu.sARTU^r W'ii.t oi- I't.as 43 LITERATURE CITED 111 Appt-l. ()., ;m(l Wollcuwcljer, H. \V. ]!)l(l. I p.. ilius. (7) Elliott, J. A., and Crawford, R. F. 1922. The spread of tomato wilt by infected seed. I'hytopalh. 12: 428- 434, illus. 18) ■ 1923. Cotton wilt, a seed l)orne disease. Jour. Agr. Research 23: 387-39."!, illus. (9) Gilchrist, G. G. 1926. Nature of resistance to roolrot caused by Ascochyla sp. and some other, fungi in the epicotyl of the pea. Phytopath. IB: 2(59-271), illus. (10) Gueguen, F. 191.'j. Sur inie nialadie du collet du pois. Ann. Serv. l^piphytics, Mem. et Rap.. 2 (19131 : 302-309, illus. (11) Hall, C. J. J. van. 1903. Die Sankt-Johanniskrankheit der Erbsen, verursacht von I'usai-iiini vasinfectnm Atk. Ber. Dent. Bot. Gesell., 21: 2-.'), illus. (12) Johnson, J. 1921. F\isai-ium wilt ol tobacco, .lour. Agr. I\cseaich 20: .')1.''>-.'i3.'>, ilho (13) Jones, F. R., and Tisdale, \V. B. 1921. Flffect of Soil tcnipciature upon the development of nodules on ihr roots of certain legumes. Jour. Agr. Reseai'ch 22: 17-.'!1, illus. 1923. Stem and rootrot of peas in the t_"nitetl States caused by species of Fusarium. Jour. Agr. Research 2fi; 4.)9-47.j, illus. (l.-)) , and Drechsler, C. 192.J. Rootrot of peas in the United States caused by .\phanomyces euteiches (n. sp). Jour. Agr. Research 30: 293-32."), illus. -, and Linford, M. B. 192."). Pea disease survev in Wisconsin. Wis. .\gr. I-^xpt. Sta. Rcscaich Bui. 64, 30 p., illus. (17) 1926. Resistance of peas to i-ootrot. Phytopath. Ki: l.'.<>- 16."). (18) Jones, L. R., Walker, J. C., and Tisdale, W. B. 1920. I-"usai-ium resislant cabbage. Wis. Agr. l^xpt. Sla. licscaich I>u!. 48, 34 p., illus. (19) , and Tisdale, W. B. 1922. The influence of soil tenipei atuic upon the df\ clopiiieiit ol fla:; wilt. Phytopath. 12: 409-413, illus. (20) , Johnson, J., anil Dickson, J. G. 1926. Wisconsin studies iipcn the relation f)f soil lenipiM ature to plan; disease. Wis. .\gr. F.xpt. Sta. Research Bui. 71, 1 It p., ilhis. I 21 I Klebahn, H. 1!:10. Krankheitcn des Selleiies Zeitschi", f. Pllaiizeiiki iiiikli. 2(1: 1-10. illus. 44 Wisconsin Research Bulletin 85 (22) LeiUh, L. lillC). Some experiments on the influence of temperature on tlie late of growth in Pisuni sativum. Ann. Bot. 30: 2.')-4C, illus. • 23) Lewis, C. E. 1913. Comparative studies of certain disease producing species of Kusar- ium. Maine Agr. Expt. Sla. Bui. 211): 203-258. illus. (24) Linford, M. B. 1926. A wilt disease of peas in Wisconsin. (Abstract) Phytopath. l(i : ".'). (25) , and Spiague, R. 1927. Species of Ascochyta parasitic on the pea. Phytopath. 17: 381-^97, illus. (26) Orton, W. A. 1909. The development of farm crops resistant to disease. U. S. Dept. Agr. Yearbook 1908: 453-164, illus. (27) Pritchard, F. J. 1922. Development of wilt-resistant tomatoes. V. S. Dept. Agr. Bui. 1015. 18 p. illus. (28) Richards, B. L. 1923. Eurther studies on the pathogenicity of Corticium vagum on the potato as affected by soil temperature. Jour. Agr. Research 23: 761-770, illus. (29) 1923. Soil temperature as a factor affecting the pathogenicity of Corti- cium vagum on the pea and the bean. Jour. Agr. Research 25, 431-449, illus. (30) Ridgway, R. 1912. Color standards and color nomenclature. 43 p., illus. Washington. D. C. (31) Schikorra, G. 1907. Lusarium-Krankheiten der Leguminosen. Arb. K. Biol. Anst. Land- u. Eorstw. 5: 157-183, illus. (32) Sherbakoff, C. D. 1915. Fusaria of potatoes. N. Y. (Cornell) Agr. Expt. Sla. Mem. 0: 87- 270. illus. (33) Tisdale, W. B. 1923. Influence of soil tempeiatuie and soil moisture upon the Fusaiium disease in cabbage seedlings. Jour. Agr. Research 24: 55-86, illus. (34) Tisdale, W. H. 1917. Relation of temperature to the growth and infecting pcnxer of Fusarium lini. Phytopath. 7: 356-360, illus. (35) Tims, E. C. 1926. The Influence of soil temperature and soil moisture on the de- velopment of yellows in cabbage seedlings. Jour. Agr. Research 33: 971-992, illus. (36) Togashi, K. 1926. On three species of I'usariuni >\hich cause the will-disease ol' pea. Jour. Soc. Agr. and For., Sapporo, .lapan. 18: 149-154. (37) Turesson, G. 1920. Fusaiium vilicola Thiiiu. inrccling peas. Bot. Xotiser 1920: 113- 125. illus. (38) Wollenweber, H. W. 191.3. Studies on the Fusarium problem. Phytopath. 3: 24-50, illus. (39) 1913. Pilzparasitare Welkekiankheiten der Kult((ri)llan/en. Ber. deut. Bot. Gesell. 31: 17-34, illus. (40) (41) (42) 1914. Identification of species of Fusarium occurring on the sweet potato, Ipomoea batatas. Jour. .\gr. Research 2: 251-285, illus. 1922. Tracheomykosen und andere NNelkt kraiikheiten nebst Aussuchl;'ii ihrer Abwehr. Angewandte Botanik I: 1-11. 1923. Fusarium-\\'elkeii oder Tracheomvkosen. (Sorauer, P., Haudbucii d. Pflanzenkrankh., Aufl. 4. Bd. 3:" 169-181. Berlin.) -, and others. 1925. Fundamentals foi- taxonoiuic studies ol I iis;iiium. Jour. .\ni . H( search 30: 833-843, illus. Research Bulletin 87 January, 1929 The Classification of Certain Virus Diseases of the Potato James Johnson Agricultural Experiment Station of the University of Wisconsin, Madison CONTENTS Introduction 1 The Present Status of the Problem 1 Viruses Studied in Experiments 3 Experimental Methods 7 Diagnostic Features Studied 8 Experiniiental Results 8 Traiismissihijity by Leaf Mutilation 9 Incubation Period 10 Aging in vitro 10 Thermal Death-point 1 1 Tolerance to Dilution 12 Influence of Chemicals 12 Varietal Susceptibility 13 Symptom Expression 16 The Identity of the Rugose Mosaic Virus and the Spot-Necrosis Virus 20 Discussion of Results 22 Summary 23 Literature Cited 24 The Classification of Certain Virus Diseases of the Potato' ■^ r p^|l~N^ HE TRANSMISSION of t>ne or more viruses from apparently heahliy potatoes of standard varieties was reported by tlie writer in Ji 1925 (5). Subsequent investigations in this and other laboratories have confirmed this observati(.n. It naturally became of special interest and importance to investigate this phenomenon further, and particularly to deter- mine the possible relationship of these viruses to other viruses causing disease in potatoes. In this connection it became necessary to make com- parative studies with other known potato viruses, and while doing so, an attempt has been made at the same time to contribute some information which may aid in their classification. The investigations of Schultz and Folsom (10), Quanjer (9), and others have shown that a considerable number of virus diseases affect the potato. The separation of these virus diseases is based largely, if not entirely, upon symptomatology, this being practically the only system available. The numerous disadvantages and difiiculties of this system are too well known by pathologists to warrant detailed discussion. The evidence lies in the exii,ting confusion in virus disease literature and in the difficulties encount- ered by active workers in interpreting the descriptions of others, as well as in the not infrequent difficulty of interpreting their own descriptions. Any clarifying information which may be added to the present descriptions, especially if it is in the form of a measurable characteristic, should, there- fore, prove useful in the classification and identification of plant viruses. In a previous investisation (6), this was attempted for certain viruses affecting tobacco in particular, by determination of the properties of the viruses concerned. The present paper deals with a similar attempt with respect to certain of the more confusing virus diseases of the potato. The Present Status of the Problem A detailed review of the literature bearing on the description and classi- fication of the potato viruses, to be of most value in the present connection, would need to be presented in an argumentative form. The potato virus problem itself is not, however, actually as complicated as it may appear 'In cooperation with the Office of Tobacco and Plant Nutrition, Bureau of Plant Industry, United States Department of Agriculture. The writer wishes to acknowledge the assistance of Mr. N. Mogendorff in conduct- ing the greenhouse work involved. 2 Wisconsin Research Bulletin 87 from a brief study of the literature. Some important questions remain to be conclusively established by repeated corroboration in different parts of the world ; whereas others may best be solved by bringing material from various sources under one set of conditions. Other difficulties may be remedied by a more general agreement on nomenclature. No doubt, new virus diseases and new problems may arise, but these will be the more readily handled when the confusion in the older problems is reduced. While potato diseases of the virus type, under the names of "Krausel- krankheit" and "curly-dwarf," have been known for a long time, the literature on this subject has little or no bearing on the development of the present subject of description and classification of the viruses, since these names were probably indiscriminately applied to a single disease or to a group of diseases. It is only within the last fifteen years that serious attention has been given to the potato viruses. During this time the chief interest has centered around recognition of the fact that several different virus diseases exist on the potato, some of which belong to the "mosaic" group, i. e., those which are more or less similar to the well known tobacco mosaic ; whereas others exhibit symptoms of quite a different nature, being more comparable to the "yellows" group. Nearly twenty apparently different viruses of the potato are now said to exist, although, fortunately, probably only three or four are economically import- ant. Schultz and Folsom (10) in this country, and Quanjer (9) and his associ- ates in Holland, were the first to separate and describe a number of virus diseases of the potato. Their unrelated work was carried on almost simul- taneously and consequently an agreement in nomenclature was not to be expected. The nomenclature of each group of workers is usually followed on the respective continents. \Vhile somewhat inconvenient, this synonomy is not important if the names used can, in each case, be definitely applied to a specific disease. Other workers have reported new virus diseases of the potato, or have added new names, some of which are accepted by one school, while others regard them as synonymous with previously described diseases. Schultz and Folsom (10) either describe or accept eight different potato virus diseases on the basis of their own work, namely : mild mosaic, crinkle mosaic, leaf-rolling mosaic, rugose mosaic, leaf-roll, streak, spindle tuber, and unmottled curly-dwarf. In addition, they accept auculja mosaic, yellow top, and giant hill as belonging to the virus group. Quanjer (9) has described or accepted the following types: common mosaic, interveinal mosaic, aucuba mosaic, crinkle, marginal leaf-roll, leaf- roll, stipple-streak, and leaf-drop streak. The description of symptoms by these authors is not and probably cannot be made sufficiently adequate for satisfactory comparison to be made as to the identity of the causal agency concerned, for reasons already referred to. The principal confusion exists, however, as regards the mosaic group. It is generally believed that Schultz and Folsom's rugose mosaic is the same as the crinkle of Murphy and Quanjer. Quanjer's common mosaic may conceivably be either the mild or crinkle mosaic of Schultz and Folsom. Virus Diseases of the Potato 3 The hitter's leaf-rolling mosaic may or may not be related to Quanjer's interveinal mosaic or marginal leaf-roll. As soon as adequate descriptions and comparisons of the various viruses described by the different workers can be made, it should not prove difficult to reach an agreement on nomenclature, at least for the viruses them- selves, possibly usins some such system as suggested by the writer for the viruses affecting tobacco (6). Viruses Studied in Experiments Our first interest in potato viruses in connection with the present investi- gations developed as a consequence of a study of the influence of environ- mental conditions on the expression of mosaic symptoms in various plants (4). In these experiments, the type of mosaic occurring very commonly in Wisconsin on the Bliss Triumph variety was largely used. In publishing these results, no mention was made of the specific mosaic concerned. Fol- lowing continued work on the Wisconsin Triumph mosaic, both with rela- tion to tuber indexing and further environmental work by Tompkins (12), it was decided to refer to this type of mosaic as "rugose" mosaic. This designation unfortunately came to be used by us generally in conversation and in publications from the laboratory. Following the present investigation, in which we have made special efforts to conform in our usage to the nomenclature of Schultz and Folsom, we have come to the conclusion that the common mosaic occurring on Bliss Triumphs in Wisconsin should be called "crinkle mosaic," a term actually introduced in literature by Schultz and Folsom (11) following our first studies on this disease. According to our present conclusions then, in all preceding publications from this labora- tory referring to "rugose" mosaic (3, 7, 12), this disease should be desig- nated as "crinkle" mosaic, identical with the "crinkle" mosaic of Schultz and Folsom, but in no way identical with the "crinkle" of Murphy and Mc- Kay (8) and Quanjer (9), resembling, however, more closely their simple or common mosaic. "Crinkle mosaic" is a very serious disease of Triumphs in Wisconsin, commercial stocks often being infected to the extent of 25-50 per cent. Field inspection, roguing, seed certification and indexing are, however, ho'ding the disease in check in a considerab'e measure. The symptoms are, of course, frequently more or less masked in the field, and crinkle mosaic may often be confused with mild mosaic for this reason. As far as can be judged from very extensive index records in the greenhouse and from observations in the fie'd, no other virus disease commonly occurs on Triumphs in Wisconsin. Leaf-roll, spindle tuber, leaf-rolling and mild mosaic may occur in rare instances or in small percentages of commercial stocks. Crinkle mosaic may also be the more common disease on the Early Ohio variety as grown in the State, but limited studies indicate that mild mosaic is most likely to be found on Green Mountain, Burbank, and Irish Cobbler varieties. The Rural appears to be comparatively free from marked symptoms of virus diseases. This is in accord with the apparent relatively high resistance of the variety to disease. The main interest has centered around the virus which we have previously described as obtainable from healthy potatoes (5), namely "spot necrosis," 4 Wisconsin Research Bulletin 87 which is believed to be a virulent form of the "mottle" virus securable from all healthy potatoes of standard varieties. Since both the mottle and spot necrosis forms were found to be transmissible to various solanaceous plant species, previous claims that various virus diseases of the potato were transmissible to other plant species (9) were, therefore, open to question. Following repeated attempts at correlating this problem with that of Schultz and Folsom, we have now been led to conclude that our soot necrosis virus is identical with their rugose mosaic virus. A large and im- portant problem is involved in this connection which will be discussed in some detail in a later chapter. It may help to clarify the subject, however, if the reader will bear this possible relationship in mind throughout the discussion. If we accept rugose mosaic and spot necrosis as synonymous, the relationship of the "mottle" virus to the former naturally assumes a particularly interesting position as far as the control of the rugose mosaic disease of potato is concerned. Judging from several years of indexing records at Wisconsin with large numbers of potatoes, it appears that no typical symptoms of true rugose mosaic have developed, and we are not acquainted with the disease as such in the field. The virus with which we have worked was, therefore, either secured from apparently healthy potatoes after passage through tobacco, or was secured from Drs. Schultz and Folsom on Green Mountain tubers. Certain other viruses, especially those of ordinary tobacco mosaic and of potato "streak," have been used in comparative inoculations. The "streak" virus has behaved very erratically in our trials, w-ith the result that its properties could not l)e definitely determined. The rugose mosaic virus has yielded typical streak symptoms and streak has yielded rugose mosaic symptoms, although more often no infection was secured. It is not unlikely that the "streak" virus, if it exists at all as a separate entity, may be a form of the rugose mosaic virus. Such a possibility has already been suggested by Murphy and McKay (8) and Atanasoff (1). The viruses with which we have worked have, therefore, been mainly those of crinkle mosaic, mild mosaic, leaf-rolling mosaic, and rugose mosaic. These have been secured in part originally from Wisconsin potatoes, but also in part from material kindly supplied to us at different times for com- parative purposes under the above names by Dr. I{. S. Schultz and Dr. Donald Folsom. (See Figure 1.) Expermental Methods The plan of the investigation of the potato viruses was based largely upon experience gained in an earlier study on the classification of various viruses affecting tobacco (6). The carrying out of the proposed investiga- tion with the potato viruses was found to be much more difficult in manv respects than that with the tobacco viruses. Artificial transmission with virus extract is naturally necessary in property studies. Failures to secure infection, or the comparatively low percentages of infection secured by artificial inoculation with certain of the potato viruses, accounts for m.ost of the difficulties preventing rapid progress of the work. The com- parative sensitivity of the potato viruses themselves to unfavorable condi- '> 9. £-0 «^ ;£ b/ 3 ■-'^ 5 .5 be -yi '-H ^i; cu ;- - (Ju ;{2 ?^ 11 OJ o o tn ci OJ bjo O '^ C aj ;3 C ' bX) -I-- ,(_, 3 <-hJ C. !- o i3 ^ n "^ u r/l o ^— ^ rt o rt O O CTJ c O ^ ?: o O. < ^ 1- u r-' o .£ ^ O rt ID U tn O O •'^ ^< a; O r- O U, .5^ -a H ■ § CM S ^ ^ c h tjo ti '/i o •-- S i) Q^ :=; Ms " £ ••" II ^ V ■^ i. u c/. ^ 9 f^-C — 03 '^ Oh bjO O rt Qj "7 O Virus Diseases of the Potato 7 tions, when in extract, added to the necessity of rapid and careful methods of technique. The requirements in the way of satisfactory plants for inocu- lation and of desirable environment are in many respects exacting. While these difficulties may in some cases serve as diagnostic features in them- selves, they make necessary the accumulation of a considerable mass of data as a basis for reliable conclusions, since negative resuhs cannot be regarded as convincing without very adequate controls. There are many advantages in the comparison of plant viruses under greenhouse conditions, where a reasonable constancy of environment may be secured, even though such studies do not apply directly to the recognition of these viruses under field conditions. The conclusive determination of a virus may eventually be best accomplished by bringing the virus from the field into the greenhouse for detailed study of its behavior and properties. (See Figure 2.) The present investigations were, therefore, conducted for the most part during the months from November to May inclusive, and under greenhouse conditions. Everything considered, the months of February, March, and April are most favorable for the work. Two greenhouses were used, one running at a temperature favorable for the normal development of the potato and for the expression of mosaic symptoms (65-75^ F.), the other at a higher temperature (85 = -90° F.), favorable for the development of symptoms of mosaic on other solanaceous plants and apparently for the incubation period of the viruses on potatoes. Ordinarily, the potatoes, grown in six-inch pots, were inoculated in the cold house, where they were left for 12-24 hours. They were then transferred to the hot house for ten days, with the purpose of shortening the incubation period, and finally returned to the cold house, w'here the development and expression of symptoms were observed and recorded at intervals up to four or five weeks. The results are, therefore, all based on current symptoms, following from two to five weeks after inoculation. Tubers were in some cases preserved and grown from inoculated plants, but the added information secured did not seem to justify the additional work involved. It is recognized, however, that sec- ondary symptoms in some cases have diagnostic value, though the interval between inoculation and the development of secondary symptoms is too long to be useful in cases where this method can be avoided. Inoculum was usually secured from diseased plants by grinding infected leaves in a mortar and straining the liquid through cheesecloth to free it from the pulp. When large amounts of inoculum were needed, the infected plants were passed through a meat grinder before extraction of the virus. In all cases, inoculations were made as soon as possible after extraction, since some of the viruses lose their virulence rapidly outside the host. The inoculations were all made by dipping a small piece of cheesecloth into the inoculum and rubbing the virus into the leaf until the same was mutilated in several places. Five plants, and sometimes ten, were used for each trial, as a rule approximately one cubic centimeter of inoculum being applied to each plant. Unless otherwise mentioned, the experiments were conducted with the Bliss Triumph variety of potato. This variety is, on the whole, judging 8 Wisconsin Research Bulletin 87 from our experience, the most susceptible to virus diseases of any of the common American varieties of potato (1). , Diagnostic Features Studied Symptom expression will, of course, always remain an important diagnostic feature with virus diseases, as it is with other diseases. Symptoms in themselves, however, cannot be regarded as a reliable index of the virus concerned in most cases where diagnosis is required on account of the similarity and overlapping of symptoms produced by different viruses, their extensive modification by environmental conditions and by other circum- stances not fully understood, together with the possibility of the existence of viruses in combination. Neither is it likely that any other single diagnostic feature will meet such a requirement. If several unrelated diagnostic features are taken into consideration, however, a process of elimination, or a "key," may be secured which will reduce the possibility of error to a minimum. While other diagnostic features than symptom ex- pression suitable for classification purposes are limited, there are some which will be found very i:seful in application. (Table I.) The most reliable diagnostic features of the "property" type are, no doubt, the thermal death-point, the longevity /;; vitrn, the effect of dilution, and the influence of certain chemicals. These, together with relative sucepti- bility, form the basis of the present paper. Several other characteristics more or less useful in classification have, however, been noted in a pre- liminary way. These are mainly : method of transmission required ; length of incubation period; comparative readiness of transmission by a single method, i. e., grafting, insects, mutilation; host range in different species or varieties ; relation of source of inoculum to infection ; influence of environmental conditions on symptom expression ; variation in cytological and histological details; and filterability. These possible diagnostic fea- tures have not been sufficiently investigated for all viruses to warrant their detailed discussion in the present paper. Experimental Results While recognizing the desirability of presenting experimental data in detail, especially where controversial matters may be involved, it seems justifiable to omit many details in the present instance since the data exist as a large number of separate trials not readily summarized for presenta- TABLE I.— A COMPARISON OF THE MORE COMMON PROPERTIES OF THE VIRUSES STUDIED AS DETERMINED ON THE POTATO Virus of Longevity in vitro Thermal death-point Tolerance to dilution 24-48 hours 43-45° C. 1-10 Rugose mosaic 24-48 hours 60-65° C. 1-100 Leaf-rolling mosaic 24-48 hours 70-75° C. 1-200 Mild mosaic 2-4 hours 40-45° C. 1-100 Virus Diseases of the Potato 9 tidii. I'^urtluTniorc, wc arc mcire cuiicenicd at this tiiiic witfi a proposed nut hod of study than we are with tlie accuracy of the present determina- tions. Improved and new technicjuc and greater accuracy of determinations will, no douht, come in future investigations if the scheme for classification and identification meets with general approval. Transmissibility by Leaf Mutilation The experiments conducted have shown that the leaf -rolling mosaic virus is the most readily transmitted to Bliss Triumph potatoes of all the viruses with which we have worked, provided that the plants are exposed to a high greenhouse temperature for approximately ten days after inoculation and subsequently placed at a lower temperature. The rugose mosaic (spot-necrosis) virus is also readily, but not as reliably, transmitted from potatoes to the Bliss Triumph variety. When transmitted from tobacco to tobacco, however, this virus rarely yields less than 100 per cent infection, although, from tobacco to potato, infection is often not secured (Table II). The "mottle" virus, which can be secured from apparently healthy as well as diseased potatoes, is readily transmitted from potato to tobacco and from tobacco to tobacco, but rarely if ever yields any symptoms on transfer to potato. Crinkle mosaic can be transferred to Bliss Triumph potatoes by the leaf mutilation method with fair certainty of securing 50 to 100 per cent in- fection, although sometimes failing to yield infection at all, for unknown reasons. While 100 per cent infection has been secured on two occasions with the mild mosaic virus, this virus is most difficult to work with because of the low percentages of infection usually secured by the leaf mutilation method, according to our experience. Incubation Period The potato may yield the first signs of disease eight to ten days after inoculation with the rugose mosaic virus if incubated at a warm temperature. Usually, however, twelve to fifteen days are required to bring out the first symptoms of disease. The mild mosaic virus appears to require the longest incubation period, twenty to twenty-five days often elapsing before even faint symptoms are evident. The crinkle mosaic virus and the leaf-rolling mosaic virus usually require fifteen to twenty days for the development of symptoms under the conditions employed. On occasional plants, a marked delay of symptom development occurs within a single series of inoculations. Repeated comparisons have shown that the incubation period can be shortened and a higher percentage of infection secured by an eight to twelve day exposure of the potato plants to a high temperature (85^-90- F.) after inoculation. This temperature is, of course, unfavorable for the normal development of the potato plant, and also for the expression of symptoms, with the exception of those of rugose mosaic. After removal to the cold house (65° -75- F.) the potatoes usually recover rapidly from the high temperature effects, an 1 the disease symptoms develop characteris- tically. 10 Wisconsin Research Bulletin 87 Aging In Vitro The dcterminatiiiii nf this property is fundamental for subsequent experi- ments on virus properties, since it is important to work witli the virus in extract in as virulent a condition as possible. The results show that tlie potato mosaic viruses as a group are very sensitive to aging /;; z'itro, and they probably begin to be inactivated quite soon after extraction. Conse- quently, it is advisable to perform the treatment to be given to the extract and to inoculate as rapidly as possible. Usually not more than one hour is required between extraction and reinoculation of the viruses in most of the experiments involving trials of thermal death-point, dilution, etc., although tw^o hours or more may be required for trials with filterability. The extracts were in all cases made from comparatively young potato foliage showing good and typical symptoms. These extracts were merely strained through cheesecloth, placed in stoppered test tubes, and stored at room temperatures for the desired period of aging. It is probably not possible to determine closely the longevity of the viruses ;';; z'ilro. as some variations due to storage conditions may be expected. The data indicate, however, that in the case of the crinkle mosaic virus a large part of the virus is inactivated at the end of six hours, and that it is all inactivated between 24 and 48 hours. In the case of the rugose mosaic virus (extracted from potato foliage) and the leaf-rolling mosaic virus, inactivation is apparently not as rapid, but again little or no infection may be expected from the virus jifter aging from 24-48 hours. The mild mosaic virus appears to be the most sensitive. Aging for even two hours in vitro appears to cause in- activation, although in one case a low percentage of infection was secured after airing for four hours. The mottle form of the virus from apparently TABLE IT.— ILLUSTRATING THE INFLUENCE OF THE SOURCE OF INOCULUM AND THE HOST INOCULATED ON RESULTS SECURED WITH AGING TESTS ON THE RUGOSE MOSAIC (SPOT NECROSIS) VIRUS' Inoculum Aging of virus From potato From tobacco To potato To tobacco To tobacco To potato None 25 8 15 10 (5) 15 15 (0) 15 12 1 clay 10 0 10 2 (6) 10 4 (t) 10 0 2 days 2.i 1 15 0 (2) 15 2 (31 10 0 4 days 15 0 15 0 (0) 10 5 (0) 10 0 6 days 15 0 10 0 (4) 10 4 (11 10 0 •lUpprr figure is nunibcr of plants inoculated; lower figure number of plants infected; figure in parenthe-sis represents number showing "mottle" form only. Virus Diseases of the Potato 11 healthy potatoes (transferred to tobacco) will resist inactivation for ten to twelve days. The rugose mosaic (spot necrosis) virus from tobacco (trans- ferred to tobacco) will also resist aging several days, consequently differ- ing from its behavior when taken from potato (Table II.). Thermal Death-Point The thermal death-points were determined by immersing about 5 cc. of the extracted virus in thin-walled test tubes in a well-agitated, constant temperature bath. The time of actual exposure in the bath was ten minutes, after which time the tubes were cooled rapidly in running water. Unheated controls of the same "age" were always used. In most cases, determinations were made only at 5° C. intervals. The crinkle mosaic virus is inactivated at a temperature of approximately 43"' C, although a temperature of even 40^ C. is apparently injurious. This conclusion is based on the inoculation of over three hundred plants with virus which had been heated at various temperatures ranging between 40° and 80° C. The thermal death-point of the leaf-rolling mosaic virus from potato was found to lie between 70- and 75^ C, whereas that of the mild mosaic virus lay between 40° and 45° C, this virus being probably more sensitive to heat, howe'^er, than that of crinkle mosaic. The thermal death-point of the rugose mosaic virus, as secured from potato and transferred to potato, lay between 60- and 65" C. The mottle form of virus present in apparently healthy potatoes when taken from tobacco is not inactivated, however, until a temperature of approximately 70° C. is reached; although the spot necrosis form (rugose mosaic) from tobacco maj' be inactivated at a lower temperature. Here again the source of the inoculum influenced the results obtained. Tolerance to Dilution The potato viruses are for the most part relatively intolerant to dilution. A dilution greater than one part of virus extract to ten parts of water usually results in rapid diminution of the percentage of infection secured. The crinkle mosaic and mild mosaic viruses are apparently the most readily inactivated by dilution, withstanding with difficulty dilutions between 1 to 10 and 1 to 100. The leaf-rolling mosaic virus may still be quite infectious at a dilution of 1 to 200. The rugose mosaic virus from potato to potato apparently will not readily stand a dilution much greater than 1 to 1(X), although infection may occasionally occur up to a dilution of 1 to 1(X)0. The mottle form of the virus from apparently healthy potatoes, taken from tobacco and transferred to tobacco, however, will stand a dilution of at least 1 to 10,000. The rugose mosaic and mottle viruses taken from tobacco will stand similar dilutions more readily than when the inoculum is taken from the potato as host (Table III.). 12 Wisconsin Research Bulletin 87 Influence of Chemicals The influeiict* of chemicals on the potato viruses has received compara- tively little attention thus far. The subject merits more special considera- tion than we have been able to devote to it. In a general way, we can TABLE III.— ILLUSTRATING THE INFLUENCE OF THE SOURCE OF INOCULUM AND THE HOST INOCULATED ON RESULTS SECURED IN DILUTION TESTS WITH THE RUGOSE MOSAIC (SPOT NECROSIS) VIRUS' Inoculum Dilution From potato From tobacco To potato To tobacco To tobacco To potato None 20 8 20 12 (8) 20 20 10 0 1-10 10 5 20 14 (6) 20 20 10 0 1-100 15 5 20 8 (11) 20 18 10 0 1-1000 20 1 20 4 (12) 20 11 (5) 10 0 1-10,000 10 0 10 0 (6) 20 4 (7) 10 0 'Upper figure is number of plants inoculated; lower figure number of plants infected; figure in parenthesis represents number showing "mottle" form only. only say that the viruses are comparatively sensitive to such chemical agents as alcohol, nitric acid, and formaldehyde. Twenty-five per cent, alcohol and 1 to 5U0i HNO., (one part nitric acid c. p. to 500 parts of inoculum) inactivate the viruses of crinkle mosaic and rugose mosaic in one hour. On tobacctj, the rugose mosaic virus is destroyed by 1 to 2(K) formaldehyde (1 part 40% formaldehyde to 200 parts inoculum) in one hoiu-, but the "mottle" form remains active. Formaldehyde 1 to 100, hduevcr, destroys the mottle form. Varietal Susceptibility A marked difference in susceptibility exists among the common American potato varieties to the crinkle mosaic virus. 'J"hc Bliss Triumph variety is the most susceptible, followed by tiie (ireen Mountain, Early Ohio, and Burbank varieties, which show an intermediate degree of resistance. We have not been able to secure definite symptoms on King, Irish Cobbler, or Rural New Yorker varieties in sinudtant'ous inoculations, under the condi- tions of the experiments. The rugose mosaic virus shows a similar behavior. In this instance, however, both Bliss Triumph and Green Mountain varieties must be placed in the su.sceptible class, Burbank, King, Early Ohio, and Irish Cobbler in Virus Diseases oe the Potato 13 tlie intermediate class, and Rural New Yorker only in the resistant class, idthough infection may tjc ticcasionally secin'ed also on this variety. Apparently all the varieties named are about equally susceptible to the leaf -rolling mosaic virus, althdugh the experiments have not been carrier! sufficiently far to determine the finer distinctions in this respect. The results are, however, sufficiently significant to suggest the possibility of using certain differential host varieties for the separation of this virus from such a virus as crinkle mosaic. The results obtained with mild mosaic are not convincing on account of the small percentage of infection secured, and the difficulty of definitely judging symptoms of the disease on varieties which are not under frequent observation. We are inclined to believe, however, from the studies and observations to date that this virus may af¥ect all of the varieties mentioned, although the symptom expression may in some cases be extremely mild and indefinite. Judging from field observation, it would appear that Green Mountain, Early Ohio, and Burbank are the most susceptible varieties to this disease. The frecjuency of occurrence of a specific virus under field conditions may evidently be due, however, to other circumstances than host susceptibility. , In this connection it is interesting to note the behavior of the common tobacco mosaic virus on the different potato varieties. This virus, when inoculated to potatoes, produces a more or less marked leaf and stem necrosis as a characteristic symptom, with some mottling and chlorosis. The Rural New Yorker variety is quite as susceptible to this virus as is the Green Mountain. Triumph, Burbank, and Early Ohio varieties are also relatively susceptible, and King and Irish Cobbler most resistant. These varietal studies taken together indicate that resistance or immunity to one virus is not necessarily associated with resistance to other viruses, although some correlation in this respect may be expected to exist (Table IV.). Symptom Expression A description of the symptoms produced l:)y the different potato mosaic viruses is in some respects the most complicated and difficult phase of the entire problem. We have previously shown experimentally the striking efYect of temperature on certain virus diseases (4) and it is now known that different temperatures affect different virus diseases in a different manner (2). Other environmental conditions, especially those occurring in the field, unquestionably influence the expression of symptoms, and we may expect that such environmental conditions may also have a variable effect depending upon the frequency and relative order in which they occur. It has, therefore, always been extremely difficult and often impossible to recognize the identity of certain virus diseases by the expression of symptoms as they occur in the field. In our own studies under controlled greenhouse conditions, we have often found such determination equally puzzling. Comparative inoculations with different viruses (as for example with crinkle mosaic and rugose mosaic viruses) on the same variety and under identical conditions frequently have yielded symptoms so similar in 16 Wisconsin Research Bulletin 87 iippearance tliat one disease might readily pass for the other. It is for this reason that any distinctions are of vahie whicli may he made on bases otlier tlian that of symptomatology. TABLE IV.— SUMMAKIZEI) RE.SULT.S OF CUKRENT INFECTION SECURED WITH VARIOUS VIRUSES INOCULATED TO DIFFERENT VARIETIES OF POTATOES' Potato variety Virus of Crinkle mosaic Rugose nio.saic Leaf-rolling mosaic Mild mosaic Tobacco mosaic Triumph 55 21 15 14 20 7 30 2 15 (5 Grpon Mountain 55 6 20 19 15 4 25 1 10 8 Burbank 40 6 20 6 20 S 20 0 15 6 Early Ohio 25 4 10 3 10 5 10 4 15 4 25 0 15 7 5 5 15 0 10 1 Irish Cobbler 25 0 15 (5 10 2 15 1 15 1 Rural New Yorker — 25 0 10 1 15 0 15 1 20 10 'Upper figure is number of plants inoculated; lower figure number of plants infected. It has fm-thermore l.een shown that potatoes normally carry a virus without any ^ymjitinns heini; evident, thonsh we are not certain whether this virus may not manifest itself under certain special environmental con- ditions. This occurrence of a specific virus in all aiiparently healthy as well as in diseased potatoes actually lueans that all uf the common potato viruses with which we are working exist only in coiuhination. Consequently, if we still regard the "mottle" virus as distinct from the potato rugose torm ( (ir spnt necrosis ),■ we are led to the assumption that the rugose mosaic virus can not as y(.-t i)e separated from the mottle form, although the inottlc form can be, and usually is, separated fri ni the rugose mosaic form as it exists in apparently healthy potatoes. It is admitted, however, that a discussidu of potato virus fliseases would 1 e (initt' impossible withdut reference to the symptoms produced. On the other hand, it is nut believed that mincir and overlapping details with respect to symiitomatology add to the descrintion, and they may result in unnecessary confusion. The following brief descrii)ti(.ns, tlurefcire, apply to the principal ctirrent symptoms as ob.served particularly on the Bliss Triumph variety, under greenhouse conditions favorable to the developmt-nt of the potato plant : -I-' O 3 p "5 ^ iJL, a; o <^, ^* ri . , , (— ' •y; c •■;; o w X ^ ^ '=; rt F -< 1^ X 18 Wisconsin Research Bulletin 87 Crinkle mosaic. Distinct mottling of foliage followed by wrinkling* or ruffling and dwarfing of leaflets and generally stunting of growth. Leaves may sometimes curl or roll to some extent. No necrosis. Symptoms masked by high temperatures (See Figure 3-A). Rugose mosaic. Symptoms commonly of two types, i. e., with and without necrosis. The characteristic and distinctive symptom is leaf and stem necrosis, often resulting in death of the entire plant. Frequently the progress of necrosis ceases before death and a dwarfed, mottled and wrinkled foliaiie continues to develop. In the absence of stem necrosis, mottling and wrinkling similar to crinkle mosaic, but often associated with distinct downward curling of leaflets and leaves ; sometimes an upward rolling or rugosity of the leaflets. Streaking on veins, petioles, and stems without serious necrosis fretiuent. Brittleness of leaves and petioles, fol- lowed by leaf -drop characteristic. Occasionally first symptoms may suddenly appear as a yellowing or chlorosis of the foliage, with one-sided leaf or stem necrosis, or streaking. The yield of tubers is very greatly reduced if not entirely prevented. Symptoms not masked by high temperatures (See Fig- ure 3-B). Leaf-rolling mosaic. The characteristic symptom is a distinct rollins;' of the young upper leaves of the plant, the disease being different in this respect from the leaf-roll disease, where the rolling more commonly occurs on the lower and older leaves. It is, of course, very different from leaf-roll in other characteristics as well. Mild and diffuse mottling usually associated with the rolling of the leaf. Leaves dwarfed in size and plant as a whole somewhat stunted. No necrosis observed. Symptoms may be partly ob- scured at high temperatures (See Figure 4-C). Mild mosaic. Mild mottling, frequently diffuse, and at other times as distinct spots. Mild wrinkling or ruffling of the leaf may or may not be present. The plant is only slightly, if at all, dwarfed. Symptoms masked by high temperatures. This disease may easily be confused with partially masked crinkle mosaic (See Figure 4-B). The Identity of the Rugose Mosaic Virus and the Spot Necrosis Virus In the previous discussion we have treated the rugose mosaic virus of Schultz and Folsom as identical with spot necrosis of tobacco and potato. This conclusion as to their identity has been reached only very gradually, and it is difficult to present convincing evidence in this direction. In our earlier paper from this laboratory it was shown that the "mottle" virus could be secured from all healthy (or diseased) potatoes tested (5). The spot necrosis form of virus was occasionally secured directly from healthy (or diseased) potatoes, but more often was secured by repeated transfer of "mottle" through tobacco. It was suspected early in the work that the virulence of the viruses concerned was variable and that mottle and spot necrosis might be one and the same virus in different degrees of virulence. Many difficulties lie in the way of conclusive proof of this contention.. 'Following' in general Sclndtz and Folsom's definition of "unit symptoms." 20 Wisconsin Research Bulletin 87 The mottle form from healthy potatoes cannot ordinarily be changed to sp<:)t necrosis at will. As far as is known at present, spot necrosis can never be secured separate from the mottle virus, although the mottle form may presumably exist separately from spot necrosis. There is some reason, therefore, for the assumption that we are dealing with two viruses in this case. Nevertheless, following repeated experiments which demonstrate changes in virulence and attenuation of these and other viruses, it was finally C(jncluded that "mottle" was only a mild form of spot necrosis, and both forms were included in one category (6). Prior to this work, Quanjer (9) and Schultz and Folsom (10) had claimed that certain of the potato viruses were transferable to tobacco and other solanaceous plants. The investigations with viruses from apparently healthy potatoes naturally led us to question those results. Repeated trials with the transference of viruses from potatoes affected with various diseases to other solanaceous species yielded on the whole results similar to those from apparently healthy potatoes. However, in the fall of 1927 ? lot of rugose mosaic Green Mountain potatoes were received from Dr. Folsom which yielded in every case infection on tobacco of the spot- necrosis type. At about the same time we were forced to the conclusion that the common mosaic occurring on Bliss Triumph in Wisconsin was not rugose mosaic but "crinkle mosaic," according to Schultz and Folsom's descriptions and specimens. We were, therefore, able to follow Schultz and Folsom's claim that the rugose mosaic virus could l)e transmitted to tobacco, and we found that it produced symptoms very similar to, if not identical with, our spot necrosis virus. It was consequently possible that Schultz and Folsom's rugose mosaic virus. Murphy's and Quanjer's crinkle virus, and our spot necrosis virus were identical. Returning now to the viruses secured from apparently healthy potatoes, it is possible to accept one of two different explanations ; namely that the "mottle" virus is a specific virus, different from any previously described virus, or that it is a mild or attenuated form of potato rugose mosaic, existing in all apparently healthy potatoes, but under certain special condi- tions increasing in virulence and causing the disease known as "rugose mosaic." The virulent form being practically self -exterminating, however, ic not commonly perpetuated, and consequently is not generally economically important in the more susceptible varieties. We have at least not seen it occurring naturally on Wisconsin Triumph ixitatoes, and it is reported as rare from other important potato-growing states as well. If the latter explanation (i. e., origin from apparently healthy potatoes) should be the correct one, it is at least interesting to speculate as to the significance of control measures for a disease of this type. In our experiments we have started on various occasions with the mottle form only from healthy potatoes, and by repeated transfer through tobacco have secured the soot necrosis (or rugose mosaic) form of virus on tobacco. (Figure 5.) We have also been able to secure various intermediate forms of mottle between the extremely mild form and the spot necrosis form, per- petuating themselves true to type (Figure 6). As previously stated, how- ever, we have on other occasions frequently failed to increase the virulence. Virus Diseases of the Potato 21 It cannot be positively stated, however, that the spot necrosis form was not accidently transmitted in some unknown manner to the experimental plants in the cases where this sudden change in virulence was noted, although we are personally satisfied that such was not the case. The spot necrosis form sometimes may be secured directly from healthy potatoes, which further- more remain healthy to all appearances for weeks following the use of portions of them for inoculum (5). Furthermore, the occurrence of spot necrosis (rugose mosaic) on tobacco following inoculation from rugose mosaic (spot necrosis) potatoes is the exception and not the rule. In one series of trials, twenty-six successive sets of inoculation from rugose mosaic potato to tobacco yielded only the "mottle" form and no typical "spot necrosis" symptoms, whereas the same virus from tobacco to tobacco always yielded "s:)ot necrosis." It is uncertain whether such differences are due merely to the source of the inoculum (Tables II., III.) or to differ- ences in the viruses. Turning now to another type of evidence, we find the rugose mosaic virus to be extremely sensitive to external conditions, and readily in- activated, attenuated or localized in the plant according to the explanation that best fits the results obtained. As previously stated, we have only observed this virus in association with the mottle form. Certain treatments of the extract, such as aging, dilution, heating, chemicals and filtration, or a change in environment of the infected growing plant may readily remove the SDot necrosis form of the virus from the "combination" and leave only the "mottle" form. A growing tobacco plant showing a virulent form of spot necrosis (rugose mosaic) on the lower leaves, placed in a different environment (usually a somewhat higher temperature) will outgrow the necrotic symntoms and the new mildly mottled or symptomless leaves will yield only the "mottle" form, whereas the older leaves from the same plant with necrotic symptoms will j'ield the spot necrosis form. If the plants infected with spot necrosis are exposed to different tempera- tures or to certain temperatures for different lengths of time, it is possible to isolate the virus in various degrees of virulence between the two ex- tremes. Several explanations may be offered. We believe we are dealing here with a virus extremely subject to attenuation as compared with tobacco mosaic, which requires quite extreme measures for the development of attenuated forms. In this respect, the spot necrosis virus would resemble certain of the human and animal virus diseases in which mild and virulent forms of the disease are said to occur commonly. If we accept the explanation that the "mottle" form of virus which is normally present in healthy potatoes is an attenuated forrn of potato rugose mosaic, the theory previously suggested that the potato protoplasm may be the cause of the disease is no longer tenable. If the "mottle" virus is regarded as distinct from the rugose mosaic virus, or if the rugose mosaic virus is regarded as a more virulent form of "mottle," this theory may still remain as a possibility. It may yet be argued that the spot necrosis virus on tobacco and potato and the rugose mosaic virus on tobacco and potato are not identical, but we have not observed sufificient evidence in our experi- ments to justify serious consideration of this possibility. 22 Wisconsin Research Bulletin 87 Discussion of Results It is hoped that the investigations reported in this paper may be of some material aid in eventually clearing ud questions of classification and nomen- clature of potato virus diseases. We are confident that these studies have at least been of value as far as Wisconsin conditions and the Bliss Triumph variety of potato are concerned. Fortunately, the investigations have tended to decrease rather than to increase the number of names to be applied to specific viruses affecting the potato, and the problem on the whole seems in some respects less complicated now than at the beginning of the studies. From an international viewpoint, much remains to be done in the way of corroboration and agreement on nomenclature, but this should not be difficult as soon as an agreement can be reached as to the specificity and description of the viruses concerned. It is hoped that the property and behavior studies such as described in this paper under controlled environ- mental conditions which can be approximately reproduced upon definite potato varieties may materially aid such mutual understanding and agree- ment. There is probably no need, for instance, of regarding rugose mosaic, Murphy and Quanjer's crinkle and spot necrosis as due to different causes, and until some one shows more convincing data than has heretofore been presented, the "streak" disease of potatoes may also well be included in this category. Particularly promising too is the possibility that the causa! agency of these diseases may be connected up with the virus commonly associated with apparently healthy potatoes. Even more important than agreements on description and nomenclature of a specific virus, is the recognition of virus combinations and their sep- aration into the constituent entities, thereby reducing confusion by avoiding the publication of new descriptions and names of uncertain and unreliable significance. An understanding of virus properties may be expected to be of particular help in this respect. By the use of certain treatments of the extracts based on their respective properties, it is certainly feasible to separate certain combined viruses from each other. In other cases, the use of the selective action of some varieties of potatoes or other host species for particular viruses may be used to separate certain combinations of viruses into their component parts. An outstanding feature of these investigations has been the recognition of the fact that although we have used the symptoms exhibited as a partial ■criterion in judging the presence or absence of a specific virus disease, the limitations of this method have become increasingly evident with the progress of the investigation. Symptoms on several plants resulting from trial inoculations with any individual case of disease are, of course, more reliable, but it is to be hoped that eventually the knowledge of the properties 2 25 46.1 6,020 8.3 32.1 99.0 3 2 4.3 7,800 1.4 0.7 9.3 4.0 17.1 4 2 10.7 16,000 28.6 S 24 19.S 4.800 13.8 57.8 98.8 6 1 l.S 6,630 0.3 1.9 4.6 * Same fields as shown in Table XIV. * The last count of this field was on suckers just before harvest, while the last counts of the others were made on stubble suckers. Overwintering Tobacco Mosaic Virus 15 TABLE XVI.— THE DISTRIBUTION OF PERCENTAOE OF MOSAIC IN SUCCEEDING ROWS ON AN ACRE OF LAND IN 1928 IN RELATION TO THE 1927 MOSAIC DISTRIBUTION COUNTS. Rows Percentage of mosaic (inclusive) 1927 1928 1- 5 26.0 7.7 6-10 16.2 16.8 H-15 16.8 15.2 16-20 6.4 4.7 2I-2S 4.0 1.0 26-30 5.1 6.8 31-35 4.1 4.8 1 36-40 9.5 7.7 41-4S 5.2 3.1 46-50 19.4 7.2 TABLE XVII. — A COMPARISON OF THE RATE OF INACTIVATION OF THE TOBACCO MOSAIC VIRUS IN MOIST, COARSE SAND UNDER AEROBIC AND ANAEROBIC CONDITIONS rEXPEWUKNT STARTED APRIL 7, 1926.) " V«-^^»^*"n » Inoculum Conditions Number of plants infected of five inoculated April 7 April 26 April 28 Infested moist sand I Aerobic 5 0 0 Infested moist sand II Aerobic 5 0 0 Infested moist sand I Anaerobic 5 Infested moist sand II Anaerobic 5 Fresh virus (control) S S 5 None (control) 9 0 0 16 Wisconsin Research Bulletin 95 would we expect it to be so on account of the number of complicating factors which determine overwintering and infection. As a matter of fact, our laboratory data would lead us to expect that in many- cases overwintering in the soil would not occur and consequently no correlation would exist. We have occasionally noted mosaic infection in a field to be more or less localized without respect to any definite factor. In such cases, localized soil infestation seemed to be the most likely explanation of the occurrence of the disease. During the summer of 1927, an opportunity presented itself for studying such a case of unequal distribution of mosaic on an experimental field one acre in size. The infection in 1927 was heaviest on the two sides of the field and comparatively light in the middle. Detailed notes on the location of the infected plants were therefore made in 1927 and likewise in 1928. An effort has been made to condense these notes in Table XVI, dividing the field into plots of five rows each, and recording the percentage of mosaic just prior to topping. It is quite evident that a considerable correlation existed in this field between the 1927 and 1928 infections, except for the two outside plots. The high count on the two outside plots (Rows 1-5 and 46-50) in 1927 was due in part to one section of each of these plots being transplanted to Burley tobacco which was known to be partly in- fected in the seed-bed. No defiinite reason, however, is known for the comparatively low percentage of infection in Rows 1-5 in 1928 as compared with the neighboring plots. Everything considered, however, the field data are convincing that the soil may harbor the virus, and that it is a source of infection which should not be overlooked in considering the epidemiology of the disease. Although we do not wish to minimize the importance of subsequent mechanical dissemination as also accounting for a high percentage of infection, we have tried to use comparable means of controlling such dissemination throughout the experimental trials. The Influence of Aeration on the Virus The relatively rapid destruction of the virus in moist soil, and the more rapid rate of destruction in ground quartz, sand, and sandy soil than in heavier soils, is of considerable interest. Several possi- bilities present themselves in accounting for the inactivation of the virus in soil, including physical or chemical absorption, toxic action, oxidation, and biological activity. In the present experiments we have been led to examine in par- ticular the relation of aeration to this phenomenon. Some evidence also exists that both physical and biological factors may play a part in the inactivation of the virus. As will be shovm, oxygen plays a considerable part, but whether through direct chemical reaction or indirect action on the biological processes of the soil is not yet clear. Overwintering Tobacco Mosaic Virus 17 When it was discovered that the virus was inactivated more ra- pidly in ground quartz and coarse sand than in field soils, aeration was naturally suspected of being a controlling factor. An experiment was therefore conducted with infested sand under anaerobic and aer- obic conditions. In one series, the oxygen was removed from desicca- tors in which the infested sand was placed, by means of the pyrogallic acid — potassium hydroxide method. The coarse sand was in this case infested with dry powdered mosaic leaves and then moistened. The results are shown in Table XVII. The rapid rate of destruction of this comparatively resistant virus under aerobic as compared with anaerobic conditions is striking. TABLE XVIII.— A COMPARISON OF THE RATE OF INACTIVATION OF THE TOBACCO MOSAIC VIRUS IN MOIST, COARSE SAND STORED UNDER AEROBIC AND ANAEROBIC CONDITIONS. (EX- PERIMENT STARTED OCT. 6, 1926.) Inoculum Storage condition Number of diseased plants of five inoculated Oct. 6 Oct. 19 Nov. 2 Nov. 9 Nov. 23 Nov. 29 Infested moist sand Aerobic Anaerobic 5 1 0 0 0 Infested moist sand 1 4 4 5 S Infested moist sand 2 Anaerobic 5 Infested moist sand 3 Anaerobic S S Infested moist sand 4 Anaerobic s Original dry inoculum (control) Aerobic 5 S 5 5 None (control) 0 1 0 0 0 0 This experiment was repeated under similar conditions as shown in Table XVIII, with similar results. It was again repeated, using extract from green mosaic plants mixed with fine sand, and inactiva- tion again occurred in about twenty days under aerobic conditions, while there was apparently no inactivation after forty days in the oxygen-reduced atmosphere. Since the tobacco mosaic virus will live for years in liquid ex- tract of mosaic tobacco plants under conditions of decomposition and fermentation, but undoubtedly under anaerobic conditions, it was thought worth while to determine the effect of air and oxygen on such an extract. In the first experiment, air was merely bubbled through the extract for about one hour daily in one case; every third day 18 Wisconsin Research Bulletin 95 in the second case; and every sixth day in the third case. The con- trol consisted of a non-aerated sample. Five inoculations were made with these differently treated extracts over a period of forty days. The first three tests during the first twenty days indicated an in- activation of the virus in the extract aerated daily, but not in the other extracts. In the last two tests the virus aerated daily seemed to have regained its infective power. During the summer months these extracts were set aside without any treatment, but were tested after six months, when they all showed normal infective power. They were again subjected to the same amount of aeration as previously, and tested ten different times during a period of two months. After forty days the virus aerated daily showed reduced infective power, but the others remained normal. However, in a dilution experiment to test the amount of virus present, the virus aerated daily gave infection of only one plant out of five at a dilution of 1-100, whereas the control, aged for the same time but not aerated, gave infection of five plants out of five inoculated with a dilution up to 1-10,000. The mosaic virus was unquestionably inactivated, though slowly, by aer- ation in this liquid extract. Since a tobacco mosaic virus extract per- mits of such high dilutions without reduction of the infective power, it was believed that the slow rate of inactivation may have been due to the escape of part of the extract from exposure to air under the particular experimental conditions used. In a subsequent trial, there- fore, g'lass tubes eighteen inches long and about one-half inch in diam- eter, filled with small glass beads, were used as containers for the vi- rus. The extract was placed in these tubes, and air was bubbled through one tube from the bottom and oxygen from a tank through a second. A third tube was untreated and served as a control. In the first trial the virus showed striking inactivation after one and two weeks in both treated tubes as compared with the control, but after three weeks the extracts seemed to regain their infectiousness. The re- TABLE XIX. — THE INFLUENCE OF DAILY EXPOSURE OF THE TOBACCO MOSAIC VIRUS IN LIQUID I.XTRACT TO AIR AND OXYGEN ON ITS INFECTIOUSNESS. (TREATMENT STARTED NOV. 4, 1926.) Inoculum Number of plants infected of five inoculated Nov. 4 Nov. 11 Nov. 18 Nov. 27 Untreated virus 5 4 3 S Air-treated virus — 2 0 4 Oxygen-treated virus - 1 0 5 None (control 0 0 0 0 Overwintering Tobacco Mosaic Virus 19 suits secured in this experiment are shown in Table XIX. This experi- ment was repeated, twelve tests for infectiousness being made over a period of three months or more. Again marked inactivation occurred after about two weeks, but after three or more weeks the virus was practically as infectious in the air and oxygen treated tubes as in the control. When, however, these extracts were tested by the dilu- tion method, it was found that the extracts treated with air and oxygen were not infectious above dilutions of 1-10, whereas the con- trol extract was infectious at dilutions of 1-1000 (Tables XX and XXI). No significant difference has been noted in these trials be- tween the inactivating power of air and of oxygen. An interesting feature of these experiments has been an attenu- ation of the virus which appeared in occasional plants Inoculated with the air and oxygen treated extracts. We have previously shown that tobacco mosaic virus may be attenuated by growing inoculated TABLE XX. — THE INFLUENCE OF DAILY EXPOSDRE OF THE TOBACCO MOSAIC VIRUS IN LIQUID EXTRACT TO AIR AND OXYGEN ON ITS INFECTIOUSNESS. (DAILY EXPOSURES FROM NOV. 11, 1926, TO FEB. 14, 1927.) Number of plants infected of five inoculated Inoculum By dilution April 7 Nov. 11 Dec. 11 Jan.22 Feb.l4 None 1 to 10 1 to 100 1 to 1,000 Untreated virus S 4 S 4 S s 5 s Air-treated virus 5 5 3 2 5 3 s 4 0 0 Oxygen-treated virus 5 4 0 5 0 0 None (control) 0 0 0 0 TABLE XXI. — THE INFLtraiNCE OF DAILY EXPOSURE OF THE TOBACCO MOSAIC VIRUS IN LIQUID EXTRACT TO AIR ON ITS INFECTIOUSNESS. (DAILY EXPOSURES FROM JANUARY 27 TO MARCH IS, 1927.) Number of plants in fected of five inoculated Jan. 27 Feb. 4 Mar. IS By dilution, April IS None 1 to 10 1 to 100 1 to 1,000 1 to 10,000 Untreated virus S 5 5 5 5 5 5 5 Air-treated virus S 3 S 3 3 0 0 0 None (control) 0 0 0 ' 0 20 Wisconsin Research Bulletin 95 plants at temperatures around 37 °C for ten or more days (7). The attenuation secured in these aeration experiments has been of a similar type, and remains stable through several generations of sub- sequent transfers. Further work needs to be done upon this subject to determine whether or not complete inactivation is a result of gradual attenuation, or whether, as appears to be the case in some instances, the virus is killed suddenly with no intermediate condition between the living and the dead virus. The tobacco mosaic virus has been kept for three and one-half years in tobacco extract in our laboratory without any apparent change in virulence of the type described as attenuation, although we have noted some small change in the type of symptoms produced as compared with freshly ex- tracted virus. We have also secured attenuated forms directly from material overwintering in the soil, showing that this phenomenon also occurs in nature and probably as a result of oxidation. The inactivating effect of aeration on the virus is no doubt con- nected in some way with biological activity, since when aseptic condi- tions are maintained in moist, virus-infested soil, marked inactiva- tion does not occur. This experiment was conducted by adding filter- sterilized virus to heat-sterilized soil. The evidence from two such trials is shown in Tables XXII and XXIII. In Table XXIII it is also shown that when soil is treated with a small excess of the virus extract, i. e., flooded instead of moistened, inactivation is slowed up or prevented. We are inclined to believe that, under field conditions, TABLE XXII. THE RATE OF INACTIVATION OF THE TOBACCO MOSAIC VIRUS IN MOIST, DRY, STERILIZED SOILS. (SOIL INFESTED VaTH VIRUS ON SEPT. 4, 1924.) Stored Number of plants infected of ten inoculated Inoculum Sept. 4, 1924 Dec. 28, 1924 Feb. 10, 1925 Mar. 10, 1925 Mosaic extract in soil Dry 10 6 4 0 Mosaic extract in soil Moist 10 2 0 0 Dry powdered mosaic leave? in soil Dry 10 9 10 10 Dry powdered mosaic leaves in soil Moist 10 0 0 0 Filter-sterile mosaic extract in sterile soil Moist 10 9 6 9 Filter-sterile mosaic extract in unsterilized soil Moist 10 3 0 2 Overwintering Tobacco Mosaic Virus 21 TABLE XXIII. THE INFLUENCE OF WATER-LOGGED AND ASEPTIC CONDITIONS ON THE RATE OF INACTIVATION OF TOBACCO MOSAIC VIRUS IN SOIL. (EXPERIMENT STARTED NOV. 23, 1927.) Inoculum Moisture Number of plants infected conditions ■ of five inoculated in soil Dec. 29, 1927 i Feb. 1, 1928 Mosaic extract in soil Moist 4 0 Mosaic extract in soil Moist 2 0 Mosaic extract in soil Water-logged 5 5 Mosaic extract in soil Water-logged 5 4 Filter-sterile mosaic extract in stesile soil Moist 5 5 Filter-sterile mosaic extract in sterile soil Moist 5 5 Filter-sterile mosaic extract in sterile soil Water-logged 5 5 Filter-sterile mosaic extract in sterile soi' ' Water-logged 5 S Original virus extract None (control) I S 5 0 0 flooding or waterlogging of the soil may result in the overwintering of more virus in some parts of the field than in others. The action of such flooding on the virus may probably be explained on the basis of interference with aeration. The evidence presented is believed to show satisfactorily that soil aeration has an influence on the persistence of the mosaic virus in the soil. It is interesting to note that twenty-eight years ago Sturgis (11) wrote as follows: "The disease occurs abundantly in some localities, notably on the close, clayey soils on the east side of the Connecticut River, sparingly in other localities, where the soil is open and porous." The variations in different soils with respect to aeration and the factors which influence it are fairly well known and in some cases too obvious to be discussed here. Whether or not any close correla- 22 Wisconsin Research Bulletin 95 tion exists under actual field conditions between the aeration of soil and the persistence of the mosaic virus in infested soil remains to be determined. Discussion of Results The overwintering of the tobacco mosaic virus under farm condi- tions in tobacco or tobacco refuse, in either the cured or the dried condition, is sufficiently evident. The part played by such material in epidemics of the disease, however, is quite uncertain. It is natur- ally dependent upon the amount of this infectious material which is transferred to the seed-bed or to the field. If it is assumed that in most cases only occasional primary infections result from such a source, subsequent dissemination in the case of epidemics must be due to a very active agency. Certain cultural operations, especially topping, of course, are responsible for much of this dissemina- tion, as is shown by the experiments of others as well as our own (Table XXIV), but this does not by any means account for heavy infection occurring before topping in cases where the seedlings were known to be healthy at the time of transplanting. That aphids are responsible for such dissemination appears very doubtful, and Dr. Hoggan has already shown that the aphid commonly suspected {Myzus persicae) does not transmit the tobacco mosaic virus (5). If, on the other hand, tobacco plants may become directly infected from virus carried to or overwintered in the soil, a disseminating TABLE XXIV. — SHOWING THE INFLUENCE ON DISSEMINATION OF TOPPING MOSAIC AND HEALTHY PLANTS IN A ROW SEPARATELY AS COMPARED WITH TOPPING BOTH AT THE SAME TIME, IN A FIELD WHERE A RELATIVELY HIGH PERCENTAGE OF MOSAIC INFECTION WAS DEVELOPING FROM SOIL INFESTATION. Mosaic and healthy plants topped Row Percentage of mosaic at time of Percentage of in- crease between time of topping and suckering Season Topping Suckering Separately 1* 24 36 SO 2 24 46 91 Together 3 4 16 76 375 1926 4 40 900 Separately 1 2 20 44 110 18 12 38 82 111 Together 4 583 20 72 12 260 Separately 1 12 0 1927 2 3 14 12 16 14 Together 74 80 516 4 6 1233 *Each row represents a total of fifty plants. Overwintering Tobacco Mosaic Virus 23 agent of the insect type is not necessary to explain the observed conditions. The investigations presented in this paper, as well as numerous confirmatory field observations in Wisconsin and certain other tobacco districts, seem to prove conclusively that infection frequently occurs from virus overvirintering in the soil. Heavy in- fections occurring in the field may then result from (a) transplant- ing of plants infected in the seed-bed; (b) infection from the field soil; and (c) spread of the original infection by one or more dis- seminating agencies, particularly cultural practices. The infection of plants in the seed-bed may result from one or more of several different sources. The soil may be infested as a consequence of the previous grovs^th of mosaic plants on the soil, or refuse from curing barns may have been used as fertilizer or accidentally carried in by other means. Tobacco seed often contains much chaff which may harbor the virus and may result in the introduction of virus into the seed-beds. On the other hand, plants may be infected in the seed-bed in more direct ways, such as by the use of tobacco refuse extract as an insecticide; the use of contaminated seed-bed frames, cloth, or sash; or the transfer of tobacco refuse by wind and animals or man to the seedbeds. According to our observations, the occurrence of heavy infections in the seed-bed is quite as likely to result from vi- rus which has overwintered in the soil as from material transferred to the seed-bed in the spring of the year. Mosaic in the field, occurring as the result of infected seedlings, is usually readily recognized on account of the early development of symptoms. Mosaic infection developing from the soil in the field appears later and develops more or less gradually as the season progresses. The tendency of the plants transplanted to virus-infested soil is to remain healthy. The continuous exposure of the plants to infectious material, together with the only occasional occurrence of circumstances favoring in- fection, such as root-wounding, naturally results in the irregular or periodic development of the disease. The virus overwintering in the soil usually comes from field refuse or roots of the preceding crop, and is consequently dependent on the preceding year's infection. That this is usually very abun- dant may be seen on the secondary growth in fields repeatedly grown to tobacco. Whether or not the virus overwinters in the soil and causes infection on the succeeding crop is apparently dependent upon a number of circumstances, some of which are indicated by our ex- perimental work, such as the influence of moisture and aeration. There is plenty of field observational evidence that infection may or may not result from the field soil, and this is to be expected according to our own interpretation of the results. If overwintering of the mosaic virus should occur regularly and in quantity, the disease would be a much greater limiting factor to tobacco production than it now is. According to our own interpretation of results, the control of to- bacco mosaic will be dependent in a considerable measure on a certain'^ 24 Wisconsin Research Bulletin 95 amount of rotation for the seed-beds and the field crop, together with some special effort to prevent seed-bed or field infection as a con- sequence of transferring crop refuse to such locations. Summary 1. The mosaic disease of tobacco occurs to some extent in prac- tically all tobacco fields. Fortunately, serious injury to the crop is less common; nevertheless the total loss from this disease is much hig'her than is generally recognized. 2. The early investigators of this disease considered soil infesta- tion as an important source of infection. More recent investigators, however, have emphasized perennial wild hosts and aphid dissemina- tion as the important factors in the occurrence of the disease. 3. The results presented in this paper point to the ability of the tobacco mosaic virus to exist for long periods of time in dead plant material as an important factor in the overwintering of the virus in the curing shed and in the soil. 4. A considerable amount of the mosaic occurring in the field is due to seed-bed infection. The seed-bed infection may result from tobacco or tobacco refuse transferred to the seed-beds in the spring or from the virus overwintering in the seed-bed soil. Field infections may also result from tobacco or tobacco refuse from curing bams or warehouses, but more commonly from field refuse, stubble and roots from the preceding infected crop. 5. The frequency and extent of overwintering of the mosaic virus in the soil is dependent upon a number of factors. Moist and well- aerated soils favor the inactivation of the virus as compared with dry, compact or waterlogged soils. 6. The tobacco mosaic virus has frequently been recovered in the spring of the year from soil and roots which have remained in the fields throughout the winter in Wisconsin. 7. Tobacco transplanted from the same seed-beds to fields knovsm to be infested with mosaic and to fields known to be practically free from the disease has repeatedly shown very heavy infection in the former case and very light infection in the latter case. 8. Tobacco plants do not ordinarily become easily or rapidly in- fected from mosaic-infested soil. Infection is rather gradual through- out the season. Much of the late infection, however, is admittedly due to field cultural operations such as topping. 0. It is believed that future considerations of control measures must take into account more generally the bearing of the overwin- tering of the tobacco mosaic virus in the soil on epidemiology, and consequently the relation of rotation to the control of the disease. Overwintering Tobacco Mosaic Virus 25 Literature Cited 1. Allard H. A. The mosaic disease of tobacco. U. S. Department lyiJ of Agriculture Bulletin 40, 33 pp., 7 pi. 2. Beijerinck, M. W. Ueber ein Contagium vivum fluidum als Ur- 1898 sache der Flecken-krankheit der Tabaksblatter. Verhandel. Kon. Akad. Wetensch. Amsterdam II, 6:1-22, 2 pi. 3. Clinton, G. P. Chlorosis of plants with special reference to calico of^to^bacco^ Con. Agr. Expt. Sta. Ann. Rpt. 1914, 4. Gardner, M. W., and Kendrick, J. B. Overwintering of tomato ■ly^^ mosaic. Bot. Gaz. 73:469-484, 1 pi. 5. Hoggan, I. A. The peach aphid (Myzus persicae Sulz.) as an agent m virus transmission. Phytopath, 19:109- 122, 2 pi. 6. Iwanowski, D. Ueber die Mosaikkrankheit der Tabakspflanze. iy03 Zeitschr. f. Pflanzenkrank. 13:2-41, 3 pi. 7. Johnson, J. The attenuation of plant viruses and the inactivating ■ly^t) influence of oxygen. Science 64:210. 8. Johnson. J. The classification of plant viruses. Wis. Agr. Expt. ^y^i Station Research Bui. 76, 15 pp., 8 pi. 9. Mayer A. Uber die Mosaikkrankheit des Tabaks. Landwirtschaft. 188b Versuchs. Sta. 32:450-467, 3 pi. iO. Raciborski, M. Verslag omtrent den staat van 'slands Planten- Aoyy tmm te Buitenzorg over het jaar 1899, 73-78; 11. Sturgis, W. G Preliminary notes on two diseases of tobacco. ^ojy Ann. Rep. Conn. Agr. Expt. Sta. 22:242-260. ResearcK Bulletin 97 December, 1929 Inheritance of Fusarium Wilt Resistance in Canning Peas B. L. WADE Agricultural Experiment Station of tKs University! of Wisconsin, Madison CONTENTS Introduction j Materials j Methods 2 Technique of the disease tests 2 Technique of the genetic studiei 6 Experimental Results 9 Wilt resistance in parental varieties 9 Wilt resistance in crosses 10 Linkage relationships 26 Rate of wilting 27 Discussion and Conclusions 2Q Summary 31 Acknowledgements: The writer gratefully acknowledges the valu- able advice and assistance of R. A. Brink, J. C. Walker, J. G. Dick- son, M. B. Linford, and E. J. Renard throughout this investigation. Without the valuable materials contributed by Mr. Renard and Mr. Linford, this study could not have been carried to completion. The InKeritance of Fusarium Wilt Resistance in Canning Peas^ THE FUSARIUM WILT of peas is a relatively new disease which has frequently been confused with Aphanomyces rootrot. Jones and Linford (6) mention it as an undescribed wilt disease occur ing in fifty fields. In some cases destruction of the crop was almost complete. Linford (8) named and described the causal organism {Fusarium orthoceras App. and Wr. var jnsi) and made a thorough pathological study of the disease. In a later paper Linford (9) considers that this disease is second in importance only to rootrot. In some sections of the country, it is the most important pea disease. It is especially important in southern Wisconsin, western Maryland and southern Pennsylvania. This disease is primarily vascular. The first and most character- istic symptom is a recurving of the margins of the younger leaves and stipules. An increase in turgor seems to be always associated with the disease. The affected plants do not become limp immediately but shrivel gradually from the top toward the base. Only in very late stages do the plants become entirely limp. Dwarfing, loss of color in the foliage, and vascular discoloration are more or less constantly associated with the disease. Resistance to diseases caused by vascular Fusaria has been found in many different crops. Resistant strains of cotton, flax, cabbage and tomato are in use in some areas. In most cases resistant strains have been secured by mass selection from standard varieties grown on infested soil. In only two cases have genetic studies been carried to a place where a factorial explanation is possible. Walker (12) and Burnham (3). Several commonly known but little used varieties of peas are known to be resistant to wilt, while most of the varieties used by canners are very susceptible (Linford 8.) The object of this work is to study the genetic relationship between susceptible and resistant varieties, in the hope that such a study may lead the way to a satisfactory breeding pro- gram in which resistance may be combined with commercially desirable characteristics. Occasional resistant plants occur in susceptible varieties but these are usually not type plants Linford (8). Materials Most of the seed used in this study was furnished by E. J. Renard, who has been working with peas for several years at the University of Wisconsin. The following compilation (Table I) shows the varieties, strains, degree of resistance and symbols used. A culture of F. orthoceras var. pisi was obtained from Dr. Linford. This culture (180C), which could be traced to a single spore isolation, 1 Papers from the Department of Genetics, Agricultural Experiment Station, University of Wis- consin, No. 104. Published with the approval of the Director of the Station. Wisconsin Research Bulletin 96 was increased to supply inoculum for all soil used in greenhouse studies throughout this experiment. The peas for the wilt study under field conditions were all grown on plots used by Linford in some of his earlier studies. These plots had been inoculated from many different cultures. The experiment of moving inoculated soil from the greenhouse to the field was also tried, but wilt occurred so late that it was impossible to take satisfactory notes on wilting. The equipment for the greenhouse tests consisted of one greenhouse section containing four benches and a device for automatic temperature control. Recording thermometers and mercury thermometers were used to check the accuracy of the temperature control and the distribution of heat. Table I. — Strains and Varieties of Peas Used and Their Reaction to Wilt Variety or Symbol Reaction Variety or Symbol Reaction strain to wilt strain to wilt Horal Ho Resistant Horsford H Susceptible Green Admiral Ad Resistant Susceptible Alaska S Susceptible Resistant Alaska R Resistant Surprise Sp Susceptible Improved Surprise Ira Resistant Alaska Rogue Ro Susceptible Fastiated Sweet Sw Resistant Alaska Rogue Rog Susceptible Acme Ac Susceptible Perfection P Susceptible Methods Technique of the Disease Tests The soil for the greenhouse tests was steam sterilized during the summer and inoculated with a single strain of the organism. The fungus was grown mostly in a nutrient solution (Tochinai, 11) composed of the following : Peptone lO.OOg Mono-potasium phosphate 0.50 Magnesium sulphate 0.25 Maltose or sucrose 20.00 Water lOOO.OOcc Some of the inoculum, however, was grown on barley, oat hulls, and corn meal sand. In these cases, however, the growth was not as great nor as rapid as in the nutrient solution. The mycelial mat formed in the nutrient solution was ground to small pieces by means of a power driven, mechanical stirrer, and sprinkled on the sterilized soil with a sprinkling can with holes somewhat enlarged. After the inoculum was mixed with the soil, it was permitted to stand for two to eight weeks with an occasional stirring to insure a thorough distri- bution of the fungus. Furthermore, some of this finely divided mycelial mat was put in each row immediately before the seeds wei'e planted in the greenhouse. Since sufficient soil was required for four greenhouse benches averag- ing thirty feet long, three feet wide, and five inches deep, it was found necessary to inoculate the soil in several lots. These different lots were carefully mixed together over a rotary screen before distributing the soil FuSARiuM Wilt Resistance in Canning Peas 3 to the greenhouse benches. About 20 per cent of sand was included with the soil. The first planting of genetic material was made September 15, 1928. Thirty seeds were planted in each 36-inch row and the rows were about eight inches apart. The depth of planting was approximately one inch. Seed from the parental stocks were planted as checks. Ten seeds from each check in each bench were considered sufficient. All the benches were alike in the material grown. The only difference was in the time of planting — one day later in each succeeding bench — and in the arrange- ment of the rows so as to secure a systematic distribution of the proge- nies. Soon after planting a trellis was built and threaded with soft twine to furnish support for the plants. Figure 1 shows the trellis and part of the equipment. During the first run in the greenhouse, notes were taken on each bench at three day intervals, but during the second run the readings were made every fourth day, i. e. one bench each day, over a period of about 36 days after the beginning of the wilting. The note-taking was begun as soon as clear cut wilt symptoms could be noticed; usually about 20 days after planting. A plant was adjudged to be sufficiently wilted for removal from the wilt bed when it showed the positive stipular curl and increased turgor. In case of the taller types some were removed when they showed unilateral stipular curl and increased turgor. The same criteria were used under field conditions. In the field, however, unilateral wilting did not occur. The difference between resistant and susceptible strains is very clearly defined. The resistant plants are entirely resistant while the sus- ceptible plants are entirely susceptible when grown on disease infested soil. It was not found necessary to provide any classes for intermediate types of wilting. There is, however, a slight difference in the time of wilting of tall and of short plants which is considered later in this paper. As a check on diagnosis, platings were made in duplicate, on acidified potato dextrose agar and read at the end of four or five days. Positive identifications were not made but the fungi isolated were compared with stock cultures of the fungus causing the disease. All contaminated plates were discarded except those in which the contaminant interfered in no way with the reading of the plate. Isolations were made during the first and second runs in the greenhouse but not under field conditions. During the first greenhouse run isolations were made at random from the wilted plants with some attempted isolations from healthy plants to serve as checks. During the second run practically all wilted plants were plated, and all the healthy survivors from one bench except those transplanted. A summation of plating results during the second run in the greenhouse is given in Table II. These results may be interpreted to mean that there is a rather high correlation between the presence of what was considered F. orthoceras var. pisi and the occurrence of the disease. They also indicate that in some cases there may be an invasion of resistant plants but not sufficient to cause wilt symptoms. Wisconsin Research Bulletin 96 It has been thought that perhaps the breaking of roots of wilt resist- ant varieties would permit the entrance of the fungus and cause the dis- ease. To test the truth of this hypothesis a wilt resistant and a wilt sus- Table //. — Results of Plating Plants on Potato Dextrose Agar to Determine the Presence of the Causal Organism Condition Strains of plants Number Number Per cent represented when plated plated Positive positive Susceptible checks Wilted 464 375 80.8 Resistant checks Healthv 127 6 4.7 Segregating progenies Wilted 621 533 85.8 Segregating progenies Healthy 302 21 7.0 Total Wilted 1085 90S 83.7 Total Healthy 429 27 6.3 ceptible variety were grown on healthy soil and transplanted to wilt infested soil with some breakage of roots. Only negative results were obtained as shown in Table III. Table III. — An Experiment to Determine the Effect of Transplanting on Resistance Jan. 5, Date Date 1929. Varietv Num- planted in transplanted Num- Num- Num- ber disease free to infected ber ber ber seeds soil soil wilted plated positive Susceptible Horsford 6 Nov. 13, 1928. Nov. 25, 1928. 5 5 S Resistant Horal 6 Nov. 13, 1928. Nov. 25, 1928. 0 4 0 Considering the possibility that the breaking of roots had not been sufficient injury a second experiment Was conducted in which severe decortication of the roots was practiced. Transplantings were made at two different stages — about the second leaf stage, and at the fourth leaf stage — to test the effect of injury of plants of different ages. Table IV shows that the results are again negative. Perhaps a more extensive series covering a wider range of maturity of the plants might give differ- ent results but the present tests do not indicate this. During the first run in the greenhouse practically all the plants in the test showed cortical lesions which are not typical of wilt. To avoid this lesioning special precautions were taken during the second run. The soil was slightly ridged at the row marks, the peas planted on top of this ridge, and covered with one-half inch of sand to a width of about two inches. Watering was done only between the ridges. Only slight cortex lesions occurred during the second run. Figure 2 shows how the peas were planted and watered during the second run in the greenhouse. Plants with severe cortical lesions are weak, seldom produce any seed and under unfavorable conditions may die or become broken off. It was thought that perhaps organic mercury disinfectants might be of value in increasing germination and in preventing cortical lesioning. To test FusARiuM Wilt Resistance in Canning Peas 1 > •rj Date first trans- plant- ing Num- ber trans- plant- ed -< ro Date of plant- ing On' On' On On oo" c c 01 « Num- ber seeds <~0 -H 1 ON_- ^-^ o o o the value of this hypothesis 60 seeds of a susceptible variety -were shaken -with organic mercury dust and compared -with the same variety without treatment. Table V presents the details of this experiment. The organic mercury interfered with the entrance of the casual organism into the plants in no way since all the treated plants -wilted. On the basis of this ex- periment all seeds planted in the field were treated with this mercury dust before sowing. Throughout the tests in the greenhouse an effort was made to keep the temperature near the optimum (21°C) for wilt as found by Linford (8). Two re- cording thermometers (air and soil) gave records in- dicating that this effort was successful. As a check upon this three ordinary mercury thermometers were placed in the soil in each bench and read thrice daily. No temperature records were kept during the field tests. Under field conditions elimination of the suscep- tible checks by the disease was almost complete. Only a few, mostly near the edges of the inoculated plot, escaped the infection. Figure 3 shows the resistant checks and the susceptible checks near the center of the plot about 50 days after planting. In the greenhouse the parental strains were used as checks and under field conditions not only the parents but also a standard resistant check (Horal) and a susceptible check (Horsford). This was done so that different parts of the wilt bed might be readily compared with each other in case of incom- plete elimination of susceptibles. Sister strains of a pure line of Horal were used for the resistant check. In some cases a very small dwarf was produced which possessed a rosette ap- pearance from the time of emergence until death; usually about four weeks later. In one check lot of ten plants there occurred seven normals and three rosette dwarfs. Figure 6 shows the dwarf in com- parison with the normal. It could not be determined whether the dwarfs were resistant to disease or not. 6 Wisconsin Research Bulletin 96 Figure 5 shows segregation for wilt resistance and susceptibility in an F2 progeny of a cross between a resistant and a susceptible variety. Technique of the Genetic Studies It seems to be generally conceded that natural crossing in peas is not of very frequent occurrence Bateson and Pellew (1), White (13), and Table V. — Effect of an Organic Mercury Compound upon Germination of Pea Seeds under Greenhouse Conditions Variety Number seeds planted Treatment Number plants produced Percent germin- ation Cortical lesion- ing Horsford Horsford 60 ISO Shaken in organic mercury dust No treatment 54 88 90 59 None Slight Sirks (10). However, Kappert (6) states that crossing is by no means rare, and Brotherton (2) found it necessary to protect his plants from natural crossing. Renard- working at Madison, Wisconsin, states that natural crossing occurs to a limited extent in canning peas. It is, of course, possible that climate and seasonal variation may have an effect upon natural crossing. If it rains for a few days during the blooming period, crossing by hymenopterous insects is almost precluded. No precautions against natural crossing were considered necessary under greenhouse conditions. Under field conditions, however, most of the flowers were covered with glassine bags before the pollen was shed. In crosses of Horal x Horsford it was found more convenient to cover the entire plant with cheesecloth during the blooming period. The seed from the plants was harvested in separate lots as "selfed" and as "not selfed" material. The material was later labeled separately when planted. The results given in Table VI indicate that this precaution was probably unnecessary, since approximately the same ratios are obtained from pro- tected as from unprotected material. In Table VI and all other tables except tables of F^ material, the unit used to express deviations and probable errors is the single plant. The unit used for the F3 material is the single progeny. In designating the strain the first number given is the row number while the second number represents the number of the plant in the row. Crossing was carried out according to the method of Giltay (5) except that pollen was transferred by means of the tweezers instead of using the de-petaled flower of the intended male parent. Most of the crosses were made in the greenhouse and no covering was used after cross- ing. All crosses made in the field were protected by glassine bags, and absorbent cotton. The Fi seeds used for growing further generations of plants were obtained by crossing the parents in the greenhouse during the winter of 2 Unpublished results. FusARiuM Wilt Resistance in Canning Peas 1.58 1.94 1.86 0.47 0.18 0.20 S3 2 4.11 4.13 3.89 3.75 2.77 3.82 Deviation from a 3:1 ratio 6.50 8.00 7.25 1.75 0.50 0.75 Total number plants wilted ■O oo 1^ ro ci r^ UO lO to -3- <^ •* Total number plants healthy § c- O Total no. of plants wilted ooooooo oooooo oo oo Total no. of plants healthy "5 Lo -r O cc t^, ro Tj. -O f^ r^ lO O- r^ CO ^ O Strain num- ber R11-I2 Rll-14 R6-1-1 Rll-19 R7-3-2 Rll- 9 Rll- 8 Rll-13 + R6-1-3 Ad 7 Ad 5 Ho 6 Ho 4 Ho 11 Ho 9 + Ho 10 Im 5 Im 2 Im 6 + Im S Sw 1 Sw 9 -f Sw 3 + Sw 7 Resistant standard checks** a 3 t/2 Id Total no. of plants wilted te< Total no. of plants healthy OOOOOOOOO— • O OOO OOO OO O — ■ rj O O — — ' 3 O Total no. of plants wilted 00>>orv^fv] l^T}-»>^QC^rvj --0»^0'3- row^vOroCMO'^fOiJ^ O'-'i-O^ Greenl Total No. of plants healthy OOOOO OoOOOO oc -^ f^ O O lOOOOOOOiotN OOO 0 e L, 'J =-^ c -f ++ + _4J -.a So"" ■■S'c = u) ° £ w °" n «) I/) '"^ 4) !C V c rs H Id t> a. •o B V % ,n V V JS ^ *" c >> rJ T1 c ^ •a cS a < Jl V +s FusARiuM Wilt Resistance in Canning Peas 9 Experimental Results Wilt Resistance in Parental Varieties Before studying the inheritance of resistance in the crosses it was essential to know the wilt reaction of the parents. The parents were grown as near the segregating progenies as possible under both green- house and field conditions. The results presented in Table VII indicate that in most cases the parents gave the expected wilt reaction. In nine cases one or more plants did not wilt in progenies expected to be homozyg- ous for susceptibility. Five of these cases occurred during the green- house tests. In each of three progenies under field conditions a single plant failed to wilt, and in another two plants failed. These plants should probably all be classified as susceptible plants which escaped, but there is, of course, the possibility that they are not. Unfortunately these progenies which gave aberrant results in the greenhouse did not yield sufficient seed for further tests. One strain (Im 5) which is supposedly homozygous for resistance had one plant out of 37 wilted under field conditions. This may be ex- Table VIII. — Occurrence of Wilt in Fi Progenies of Resisant Alaska x Sus- ceptible Alaska and Reciprocal Cross Row number Number of seeds planted Place tested Total no. of plants healthy Total no. of plants diseased R6-2-2-11 X SlO-3-3-1 700 701 702 703 704 70S 706 707 710 711 712 714 716 717 718 719 720 721 722 723 724 728 729 730 731 732 734 73 S TOTAL 709 725 727 7 S 3 7 1 2 3 3 4 4 3 S 2 3 6 6 3 1 1 3 6 5 2 2 6 1 1 1 2 8 1 Tanks 5 4 1 S 1 1 3 1 4 1 2 1 2 1 1 33 6 3 1 1 3 6 5 2 1 6 1 1 1 57 70 1 7 0 S9-2-1-3 X Rll-19-4 S9-2-1-4 X Rll-19-4 S9-2-1-8 X Rll-19-4 S7-S-2-1 X Rl-1-1-2 R4-8-2-S X S10-3-3-S R6-2-2-10 X SlO-3-3-1 R6-2-2-12 X SlO-3-3-2 R6-2-1-2 X S9-4-2-2 R6-2-1-5 X S9-4-2-1 R6-1-3-2 X SlO-1-2-3 R6-2-1-6 X S9-4-2-S R6-2-1-7 X S10-1-3-S R6-2-1-8 X S10-1-3-S R6-2-1-8 X S9-4-2-5 R6-2-1-8 X S9-4-2-4 Subtotal Field R6-2-1-9 X S10-1-3-S R6-1-3-3 X SlO-1-2-3 R6-1-3-3 X S10-1-3-S R6-1-2-9 X S9-4-2-4 R6-1-2-9 X S9-4-2-1 R-1-1-2 X S7-S-2-4 Rl 1-19-1 X S9-2-1-6 Rl 1-19-2 X S9-2-1-6 R6-1-3-4 X SlO-1-2-3 R6-2-1-10 X S10-1-3-S . S9-4-2-1 X R6-2-1-1 R6-1-3-5 X SlO-1-2-3 Subtotal Tanks Field Field Miscellaneous Fj combinations Spl-2-l-3xIm6-l R 6- 1-3 -4 X P5-4 S9-4-2-4 X Sp 13-1 1 10 Wisconsin Research Bulletin 96 plained in one of several ways — mis-classification, resistance breaking^ down or it may have been the result of seed mixture. (1.) Wilt Resistance in Crosses Crosses Between Resistant and Susceptible Strains of Alaska Fi Generation The Alaska is one of the most important early varieties of canning peas. The strain known as Alcross produces about fifty per cent resistant and about the same proportion of susceptible plants. The strains used in this test were single plant selections from Alcross and had been tested for wilt resistance by Mr. Renard, a few years before the beginning of this investigation. The Fi plants were tested under both greenhouse and under field conditions. In the greenhouse (in Wisconsin temperature tanks) ^ 33 plants were grown almost to maturity without showing any symptoms of the disease. The distribution of these 70 plants according to progenies is shown in Table VIII, Table VIII also includes the results of a few miscellaneous Fi com- binations of other strains and varieties of peas. These results indicate that resistance is dominant. 3 Wisconsin temperature tanks are devices in which the temperature is controlled automatically and very accuratelv. Dickson, J. G. Making Weather to Order for the Study of Grain Dis- eases. Wis. Agr. JExp. Sta. Bui. 379. 1926. Table IX. — Occurrence of Wilt in Backcross of F\s to the Susceptible Parent Grown in Tanks Cross Row number Number of seeds planted Place tested Total no. of plants healthy Total no. of plants diseased S9-4-1-1 X (R6-l-3xS9-4-l) S9-2-1-6 X (R6-l-lxS9-2-l) SlO-3-3-7 X (S10-3-3xR6-2-2) S12-2 X (S12xRll-8) S9-2-2-1 X (S9-2-2xR4-2-2) S9-2-2-S X (R6-l-lxS9-2-2) S9-2-2-4 X (R6-l-lxS9-2-2) S9-2-2-4 X (S9-2-2xR6-l-2) S9-2-2-3 X (S9-2-2xR6-l-2) S9-2-2-3 X (R6-1-1 x S9-2-2) S9-2-2-3 X (S9-2-2xR-2-4-2) SlO-3-3-1 X (S10-3-3xR6-2-2) SlO-3-3-3 X (R4-fi-3xS10-3-3) SlO-3-3-5 X (R4-S-2xS10-3-3) SlO-3-2-1 X (SlO-3-2 X R7-3-1) SlO-3-2-3 X (SlO-3-2 x R7-3-1) S10-l-l-lx(R6-2-lxS10-l-n SlO-l-I-2 X rR6-2-lxS10-l-l) S9-2-1-1 X (R6-l-lxS9-2-l) S9-4-1-3 X (R6-l-3xS9-4-l) S9-4-1-4 X (R6-l-3xS9-4-l) S9-4-1-2 X (R6-l-lxS9-2-l) (RII-16X.S17) X S17-1 (Rii-n.\Sin X sn-1 (R4-8-1 X SlO-3-3) X SlO-3-3-2 600 601 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 626 628 629 2 1 3 11 17 3 6 3 1 4 4 4 8 5 4 4 3 3 6 2 5 5 6 4 4 Tank 1 1 1 5 8 0 3 1 0 2 2 1 3 1 1 2 i 2 2 1 2 2 2 1 1 1 0 2 5 9 Obs. 46 45 Exp. (1:1 ratio) 45.5 45.5 D — O.SO ± 3.22 D/P.E. — 0.16 FusARiuM Wilt Resistance in Canning Peas 11 Table X. — Occurrence of Wilt in Backcrosses of Fi's to the Susceptible Parent Grown in the Field Healthy Diseased Observed 65 66 Expected (1:1 ratio) 65. S 65.5 D — 0.50 ± 3.86 D/P.E. =r 0.13 Cross Row number Number of seeds planted Place tested Total no. of plants healthy Total no. of plants diseased (SlO-3-2 X R7-3-1) X SlO-3-2-3 (SlO-1-3 X R2-4-2) X SlO-1-3-1 (SlO-1-3 X R2-4-2) X SlO-1-3-4 632 633 634 635 636 637 638 639 640 643 ,644 645 646 647 648 649 651 652 653 654 656 657 658 659 660 6 1 7 5 5 4 5 8 12 5 7 5 S 3 5 7 4 5 10 11 6 5 5 5 1 Field 2 0 2 5 2 3 1 3 7 3 4 1 2 3 0 1 2 2 5 3 3 4 3 3 1 4 1 4 (R6-1-2 X S9-4-1) X S9-4-1-1 0 (R6-1-1 X S9-2-2) X S9-2-2-1 (R6-1-1 X S9-2-2) X S9-2-2-1 0 (R6-1-1 X S9-2-2) X S9-2-2-1 3 (R6-1-1 X S9-2-1) X S9-2-1-1 4 (R6-1-1 X S9-2-1) X S9-2-1-3 S16-1 X (S16 X Rll-10) 2 S16-2 X (S16 X Rll-10) 3 S16-3 X (S16 X Rll-10) 4 (Rll-14 X Spll) X Spll-2 2 SlO-3-1-1 X (SlO-3-1 X R2-4-3) 0 SlO-3-1-6 X (SlO-3-1 X R2-4-3) 4 SlO-1-3-2 X (SlO-1-3 X R2-4-2) 5 SlO-1-2-1 X (SlO-1-2 X R6-1-3) 2 SlO-1-2-2 X (SlO-1-2 X R6-1-3) SlO-1-2-4 X (SlO-1-2 X R6-1-3 5 S9-4-2-1 X (S9-4-2 x R6-2-1) 8 S9-4-2-2 X (S9-4-2 x R6-2-1) 3 S9-4-2-4 X (S9-4-2 x R6-2-1) 1 S17-2 X (Rll-lS X S17) 0 SI7-S X (Rll-15 X S17) 2 S17-4 X (Rll-16 X S17) 0 65 66 Backcrosses to the Susceptible Parent Backcrosses of the Fi plants to the susceptible parent were likewise grown in tanks in the greenhouse, and also in the field. The individual progenies are very small, ranging in size from one to seventeen plants, but the results are consistent throughout the experiment. Of the 222 plants involved 111 were resistant to wilt and 111 were susceptible. This is a perfect fit to a 1:1 ratio. The details of the results are given in Tables IX and X. Tabli; XI. — Occurrence of Wilt in Backcrosses of Fj's to Susceptible Parent A Miscellaneous Group Cross Row number Number of seeds planted Place tested Total no. of plants healthy Total no. of plants diseased Sp 1-2-1-1 X (Sp 1-2-1 X Im5) Sp 1-1-1-7 X (Sp 1-1-1 X Im2) S7-5-2-2 X (ImS x S7-5-1) 624 625 602 627 641 642 661 6 2 6 3 6 1 3 Tank 2 1 3 2 1 1 1 1 0 3 (Sp6 X Rll-13) X Sp6-1 (Im6 X SlO-3-3) x SlO-3-3-2 1 5 Sp6-4 X (Sp6 X RU-13) Sp6-5 X (Rog 1 X Sp6) 0 1 Observed Expected (1:1 ratio) D = 0.0 ± 1.58 11 U 12 Wisconsin Research Bulletin 96 Table XII. — Occurrence of Wilt hi Backcross of F^'s to the Resistant Parent, Resistant Alaska and Susceptible Alaska, and a Miscellaneous Group Cross Row number Number of seeds planted Place tested Total no. of plants healthy Total no. of plants diseased R4.8-1-4 X (R4-8-1 X SlO-3-3) 751 752 753 754 755 756 757 758 759 760 764 765 766 767 768 769 770 771 774 775 776 777 778 779 780 782 761 762 763 750 772 773 781 2 3 5 5 5 20 2 2 2 8 5 7 7 3 3 5 6 1 3 1 3 4 6 4 4 9 5 2 1 6 5 3 5 Tanks 1 1 5 4 4 18 2 2 2 7 4 7 7 3 3 5 5 1 3 1 3 3 6 4 1 9 111 3 2 0 2 4 3 5 R2-S-2-1 X (R2-S-2 x S3-5-2) R2-S-2-6 X (R2-5-2 x S3-S-2 Field R6-2-2-7 X (SlO-3-3 x R6-2-2) R6-2-2-9 X (SlO-3-3 x R6-2-2) R2-4-3-1 X (SlO-3-1 X R2-4-3) R6-1-1-2 X (SlO-1-3 X R6-1-1) R6-1-1-5 X (SlO-1-3 X R6-1-1) R6-1-1-6 X (R6-1-1 X S9-2-2) R6-1-1-7 X (R6-1-1 X S9-2-2) R6-2-1-3 X (S9-4-2 x R6-2-1) R6-2-1-4 X (S9-4-2 x R6-2-1) R6-2-1-4 X (S9-4-2 x R6-2-1) R6-1-3-1 x (SlO-1-2 X R6-1-3) R6-1-3-1 X (R6-1-3 X S9-4-1 ) R6-2-1-S X (S9-4-2 x R6-2-1) R6-1-3-2 X (SlO-1-2 X R6-1-3) R6-1-3-3 X (SlO-1-2 X R6-1-3) (Sll X Rll-8) X Rll-8-1 (S12 X Rll-8) X Rll-8-2 (S9-2-2 X R2-4-2) x R2-4-2-1 (S10-3-lxR2-4-3) X R2-4-3-1 Rll-lS-1 X (RU-IS X S17) Rll-15-2 X (Rll-15 X S17) Rll-11-3 X (Rll-11 X Sll) Rl-1-1-2 X (Rl-1 X S9-4-3) Total Field Miscellaneous backcrosses Im5-3 X (Sp 1-2-1 X ImS) Im2-1 X (Sp 1-1-1 X Im2) Ini2-2 X (Sp 1-1-1 X Im2) IinS-8 X (Ini5 x S7-S-I) Tanks Field (Sp7xRll-10) X Rll-10-1 (Acll X Rll-19) X Rll-19-1 Im 1-1 X dm 1 X SlO-1-3) Table XI. gives the results of a few miscellaneous backcrosses to the susceptible parent involving other strains and varieties. The results are entirely in harmony w^ith those given in Tables IX. and X. In this case, 22 plants were tested of which 11 were resistant and 11 susceptible. Backcrosses to the Resistant Parent None of the 111 plants considered in this backcross of the Fi to the resistant parent showed any symptoms of the disease. Only two of these plants were tested un^der greenhouse conditions; the remainder were tested in the field wilt plot. Table XII. shows the distribution of the plants according to progenies. The results of a few miscellaneous backcrosses to the resistant parent are also included in Table XI. These, likewise, show complete resistance. F2 Generation A total of 335 plants were tested of which 264 remained healthy and 71 wilted. The deviation from a 3:1 ratio is 12.75 plants with a probable error of 5.35. This is a fairly close fit, i.e. the deviation is 2.38 times the probable error. The odds against the occurrence of a deviation as gieat or greater than the designated one due to chance alone are 8.48 to 1. FusARiuM Wilt Resistance in Canning Peas 13 aa, O 0) .2 «■- oo O* M- ^ ' O — . _ 1 oo 00 oco^o^^jr;; H 3 n i E Si c ^ o a 00 C/3 K X 06 K X X T T X 2 2" Qifti Zn a; OS It is of interest to note how the in- dividual progenies deviate from a 3:1 ratio. The poorest fit among the seven progenies tested is that of progeny 303- 4 which had twenty plants resistant and two susceptible. The best fit was that of the mixed progeny which had twenty-six resistant plants and eight susceptible. The other five progenies are well distributed between these two extreme cases. There is nothing to in- dicate that the progenies considered are significantly different from each other in their wilt reactions. The data are presented in Table XIII. The Fs Generation Thirty-four F3 progenies were test- ed for wilt resistance— 15 in the green- house and 19 in the field. A reasonably close fit to a 1:2:1 ratio was obtained — eight resistant, twenty-one segregat- ing and five susceptible. P equals 0.32 in this case. Table XIV shows the re- action and classification of each pro- geny concerned. However, the classification given in Table XIV cannot be considered as final, for according to the laws of chance it may happen that a progeny of say fourteen healthy plants (with none wilted) may in reality be segre- gating. In this case the deviation from a 3:1 ratio is 3.5 ± 1.09 or 3.2 times the probable error. The odds against it are 31.36:1. Figure 2 shows the sharply tri-modal effect obtained by plotting a curve for the amount of wilt occurring in F3 progenies. The very sharp breaks between the central mode and the other two modes indicate that it is relatively improbable that very much misclassification of progenies oc- curred. 14 Wisconsin Research Bulletin 96 I'ABLE XIV. -Occurrence of Wilt in F3 Progenies from the Cross of Susceptible Alaska X Resistant Alaska and Reciprocal. Cross Progeny number Place tested Number of seeds planted Total number plants healthy Total number plants diseased Probable genotype* R4xSl 553-1 S53-2 553-3 553-4 554-1 554-2 554-3 SSS-1 555-2 555-3 556-1 556-2 556-3 556-4 556-5 557-1 557-2 557-3 558-1 558-2 558-3 558-4 559-1 559-2 560-1 560-2 560-3 561-1 561-2 561-3 562-1 562-2 563-1 563-2 Tanks 30 34 IS 40 10 18 17 20 40 45 22 46 14 27 12 29 22 25 20 9 53 29 44 10 IS 42 43 32 67 30 12 17 0 22 14 21 S 14 15 13 21 0 13 30 8 21 5 19 17 18 12 6 49 29 38 6 13 0 30 30 18 0 49 0 5 11 22 8 0 7 3 0 0 5 7 30 5 10 3 5 3 9 6 4 0 3 0 0 0 3 5 56 11 12 14 16 16 27 0 4 fw fw Fw fw Fw Fw Fw fw Fw fw Fw Fw Fw Fw Fw fw Fw fw fw fw Fw fw Fw fw Fw fw Fw fw Fw fw Field Fw fw Fw fw Fw fw Fw Fw Fw fw Fw Fw Fw Fw RlxS3 Fw Fw Fw fw Fw fw fw fw Fw fw Fw fw Fw fw fw fw Fw fw fw fw Fw Fw Fw fw FwFw X' = 2.41 0.32 Fwfw fwfw Observed S - 24 - 5 Expected 8.50:17.00:8.50 (1:2:1 ratio) 'Fw is a factor for resistance; fw, its recessive allelomorph for susceptibility. Crosses of Surprise and Improved Surprise Surprise is a very early, wrinkled pea that is susceptible to wilt. Improved Surprise may be distinguished from Surprise by a slight dif- ference in pod shape and by being entirely resistant to wilt. In this cross so few Fi's and backcrosses were grown that the re- sults from them are of little significance. Tables VIII, XI, and XII, show a few cases in which these two strains are involved. None from this par- ticular cross were carried through to the F3. The Fo results give a close fit to a 3:1 ratio. Of the 309 plants tested 234 remained healthy and 75 wilted. This is a deviation of only 2.25 plants from the 3:1 ratio. This deviation is 0.44 times the probable error. The details are presented in Table XV. Of the eight Fa progenies tested all gave very close fits to a 3:1 ratio. The poorest fit was thirteen healthy to two wilted, in progeny 305-1. This is a deviation of only 1.75 plants FusARiuM Wilt Resistance in Canning Peas IS QCL, u^ ^ C> r— -^ 'O vc 00 ^ \i~i r.) O r>. cvi ^_ O; 00 -^^ -< -^ ooooooo 2 X. nl o PS 1.13 1.82 2.66 1.63 l.OS 1.57 2.35 1.70 5.13 §..2 .2E2 > o 1.75 2.25 0.25 1.25 0.25 0.25 2.25 l.SO 2.25 Total number plants diseased r>.i cv] -H a. f^ t^ -^ t^ »o Total number plants healthy * Place tested E E - = = - = _ Number of seeds planted 20 45 90 60 60 30 68 46 305-1 305-2 306-1 306-2 306-3 315-1 315-2 316-1 6u Sp 1-2-1 X ImS Sp 1-2-1 X ImS Sp 1-2-1 X Im2 Sp 1-2-1 X Im2 Sp 1-2-1 X Im2 Im 2 X Sp 1-1-1 Im 2 X Sp 1-1-1 Im 2 X Sp 1-2-1 o 5;w 2.49 1.07 1.14 0.65 1.79 2.33 1.58 0.07 1.94 1.86 2.08 0.47 0.19 0.20 0.67 0.96 0.76 0.47 « 2 (1, CMC^rOrOr^tNTfrOTj-POcor'^rorD^r^^H^ ;2es > o „ 6.00 2.75 4.00 2.50 5.25 4.75 6.50 0.25 8.00 7.25 8.25 1.75 0.75 0 75 2.75 2.75 1.50 6.75 Total number plants diseased Total number plants healthy ■Tj-»nOCM0C«n'^CMTj-'-<^'«0000000000io Progeny numbers O O O OIO o o o o o o o o o o o o 15 tf .S m 16 Wisconsin Research Bulletin 96 Table XVII. — Occurrence of Wilt in F3 Progenies of the Survivors of an Fi Wilt Test of the Cross Horsford x Resistant Alaska. Progeny no. No. of seeds plant- ed Place tested Total no. plants heal- thy Total no. plants dis- eased Probable genotype % Wilt Hi- X Rll-12 (200-2) 1165 1166 1167 1168 1169 1170 1171 1173 1174 1177 1178 1179 1181 1182 118.5 IISS 1188 1189 1191 1192 1193 1194 1195 1197 1199 1202 1208 1209 1210 1212 1215 1217 1218 1220 1221 1224 1227 1228 1230 1231 123 2 1233 1234 1235 1236 1238 1239 1240 1241 1242 1243 1244 1356 1357 1358 13 59 1361 1363 1364 1365 1366 1367 1369 1370 1371 6 6 7 8 6 9 6 6 14 7 6 16 U 6 17 6 13 11 8 8 5 7 5 8 7 4 8 7 9 5 10 6 6 11 7 9 8 6 7 9 11 7 8 9 10 6 5 8 5 4 9 5 8 6 13 8 17 6 S 5 6 6 7 9 10 Field 6 2 4 7 5 9 6 6 12 7 4 10 5 4 16 4 9 5 S 6 5 5 5 8 6 4 5 4 9 5 5 8 4 3 5 4 8 S 5 4 6 2 6 5 8 4 8 5 6 5 7 5 8 3 5 5 3 5 4 9 8 0 4 0 0 0 0 0 0 0 1 4 5 1 0 2 3 1 1 2 0 2 0 0 1 1 1 2 2 1 0 1 1 2 1 2 4 1 2 2 3 4 0 3 0 0 0 3 1 1 0 0 0 2 0 0 3 1 0 2 1 1 0 2 Fw Fw Fw fw Fw fw Fw Fw Fw Fw Fw Fw Fw Fw Fw Fw Fw Fw Fw Fw Fw fw Fw fw Fw fw Fw fw Fw Fw Fw fw Fw fw Fw fw Fw fw Fw fw Fw Fw Fw fw Fw Fw Fw Fw Fw fw Fw fw Fw fw Fw^ fw Fw fw Fw fw Fw Fw Fw fw Fw fw Fw fw Fw fw Fw fw Fw fw Fw fw Fw fw Fw IW Fw fw Fw fw Fw fw Fw Fw Fw fw Fw Fw Fw Fw Fw Fw Fw fw Fw fw Fw f« I'w Fw Fw Fw Fw Fw Fw fw Fw Fw Fw Fw Fw fw Fw fw Fw Fw Fw fw Fw fw Fw fw Fw Fw Fw fw 0. 66.66 33.33 0. 0. 0. 0. 0. 0 0. 20.00 28.57 50.00 20.00 0. 33.30 25.00 16.66 11.11 25.00 0. 28.57 0. 0. 14.28 20.00 16.66 40.00 40.00 20.00 0. 16.66 16.66 20 00 20.00 40.00 57.14 16.66 33.33 27.27 20.00 3 7.50 50.00 0. 60.00 0. 0 0. 50.00 20.00 11.11 0. 0. 0. 22.22 0. 0. 50.00 16.66 0 40.00 16.66 20.00 0. 20.00 Fw Fw fw fw Observed 24 : 41 : 0 Expected (1:2:0 ratio) 21.67 : 43.33 Dev. 2.33 ± 2.35 Fig. 1. — A vietv of the greenhouse during a wilt test. This shoivs re- cording thermometers, trellis, and part of the air distributing system. • 7 9i 6 o E 3 -I 1 I I 0 5 fO 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95/00 Percent oj UJilt Fig. 2. — Relation betiveen number of piogenies and percentage of u>ilt. Wisconsin Research Bulletin 96 * ' nil I. J. F/fl'. 3. — The use of sand to prevent cortex lesioniug. Row 34 shoiof; (1) resistant Moral (2) susceptible Horsford and (3) another parental strain ,?4. Row 35 contains Fj, seyrcyatimj protjenics fro)n the cross of Horsford ivith Horal. FusARiuM Wilt Resistance in Canning Peas p J ^1 A h t-< Q •i= *- c E ra I o o vo o — rt 0^ 1 „ « ^ 1 T«- vO — -^ O- ■ 0 w^ iri 00 0\ QC < c«j2^^h-j^ .2 £ E K ^ s: >>^ o c 2;oa M-2 o' C — OOOO 'x>|2 0«i: * » 2 « .Si o FusARiuM Wilt Resistance in Canning Peas 21 Inheritance of Resistance in Surprise x Resistant Alaska Crosses Surprise differs from the Alaska strain used in being wrinkled and susceptible to wilt. The total number of Fo plants considered in this case is 195 of which 155 remained healthy and 40 wilted. This is a very fair fit to a 3:1 ratio since the deviation of 8.75 plants is 2.14 times the probable error. The odds against the occurrence of a deviation as great or greater due to chance alone is 5.38 to 1^ The small number of Fo plants considered is due to the relative unproductiveness of the Fi plants. Of the three progenies tested, 240-2 gave the largest deviation from a 3:1 ratio, with 59 healthy and 11 diseased plants, while 239-2 gave the smallest deviation, having 42 healthy and 16 diseased plants. The deviations with probable errors are: 6.50 ± 2.44 and 1.50 ± 2.22, respectively. The data are pre- sented in Tabid XX. Although the number of F3 progenies is small, an excellent fit to a 1:2:1 ratio is obtained. Twenty progenies were considered of which six were found to be resistant, nine segregating, and five susceptible. The value of P in this case is 0.88. The number of plants per progeny is again small, ranging from five to nine. The possibility should be borne in mind, therefore, that some of the families classified as resistant or susceptible may belong in reality to the segregating class. Inheritance of Resistance in Stcsceptible Alaska and Horal Unfortunately this cross did not furnish sufficient seed for both greenhouse and field tests. A greenhouse test was conducted with a total of 74 plants. Of these 50 were classified as resistant and 24 as susceptible. This is a deviation from a 3:1 ratio of 5.50 plants which is 2.19 times the probable error. Ordinarily this would be considered a very fair fit but in view of the fact that there is an excess of susceptible segregates when the checks in the adjoining rows showed less than complete elimination, this may in reality indicate a poor fit. Table XXI shows the results in detail. Fifty-nine F3 progenies containing from five to twenty-five plants each, were tested. Eighteen families were classified as resistant, 33 as segregating, and 8 as susceptible. These results indicate only a fair fit to a 1:2:1 ratio. The value of X^ is 4.23 and P equals 0.12. There is a large deficiency in the recessive class and ani excess of segregating proge- nies, suggesting perhaps, that elimination of the susceptible plants in F3 was not complete enough to satisfactorily differentiate these two classes. Weight is given to this suggestion by the fact that these F3 progenies were grown next to the Perfection x Horal F3 progenies in which a close check of the soil revealed that inoculation was not complete. Inheritance of Resistance in Perfection and Horal Crosses Perfection is one of the most important of the wrinkled canning peas. It is a slightly taller plant than Horal and is completely susceptible to wilt. Wisconsin Research Bulletin 96 S'w o\ ^^ tM V J3 M ! nj O &< c _ o .. o do'" i/i Qi; fe «-n , 13^- !" Tot num plar disea '*• otal mbe nts Ithy o H 3— ? «°-^ UT3 V « "t; '> ' £ o c V O o-s-s 'a E S = 1 ° ^^"i 1 >>.^ 1 c c ^E r5 Oc C s Q o .ss ■^ 'C ^ -^ ■^ o O IS IN c -'^ J * rs) >0 O Tj- ^ rv) Tj- -o r^ fo O O d O O O O <^ 0\ »-H ro O -H O r^ rs) -H ^ T}- lO •|e1 So'" H 3 Q,.2 --- OJ y^ >» o S S — ^2 O B « S ^o «• OJ Xi u, 3 V0g0"l=0<-gr^0'>«-0r0 c — • .2 ■■ o •^E« > 5 *-. _ C V, JJ J d i-^ .^' d - "3x1 -2 ja o E S- -H « rt _ „ rt 00 ►^„„„>> X3 s:s S Si a 3 J2 ;5oO. —<-<—. o 0>U: y. »; X y, y; i^. H ^■3 ^ o ^ 3 si p o 5 FusARiuM Wilt Resistance in CANNDsra Peas 23 Under both greenhouse and field conditions a very close fit to a 3:1 ratio in F2 was obtained. The total number of plants tested was 287 of which 217 were classified as healthy and 70 as wilted. This represents a deviation of 1.75 plants and it is only 0.31 times the probable error. Only two progenies are considered of which 212-2 was tested in three different places. The details are given in Table XXII. Fifty-nine progenies from this cross, comprising from five to twenty- two plants each, were tested in F3. Of these 26 were classified as resist- ant, 25 as segregating, and,' 8 as susceptible. The expectation on a 1:2:1 basis is 14.75:29.50:14.75, respectively. The value of X^ is 12.35 and P is equal to 0.0021. In only one trial in about 500 should a fit as poor or worse than this occur due to random sampling alone. It was noticed during the wilt test that a part of the F3 progenies from this cross were very much retarded in development of wilt symp- toms. To test the distribution of the inoculum, check plants were planted approximately one to the foot throughout the rows upon which these progenies were planted. Of the 122 special check plants considered, only 72 wilted. The other fifty plants remained healthy. To make sure that it was not the effect of season on the wilting of these special checks, a similar lot of twenty seeds was planted at a place in which complete elimination had occurred. In this case, all the plants wilted promptly. The results with these special checks indicate that the inoculum was dis- tributed in a rather irregular manner in this part of the field. Inheritance of Resistance in Fasciated Sweet and Acme Fasciated Sweet is a taller pea than Acme, and is resistant, to wilt. Fasciated Sweet is a wrinkled segregate from a cross of Arthur field pea with Perfection. Acme is a wrinkled segregate from a cross of Horsford with a segregate from a previous cross of Horsford with French June. Considering the total number of plants worked with (1141) a very good fit to a 3:1 ratio is obtained. Eight hundred fifty-seven (857) plants remained healthy and 284 wilted. The deviation of 1.25 plants is only 0.13 times the probable error. In one case, however, the deviation is 4.30 times the probable error based on a total of 43 plants. This result may be due to the accidental destruction of some of the healthy survivors since this progeny was damaged in the cultivation of nearby plots. On the other hand, it may be due to random sampling, but the odds against this occur- rence due to chance alone are 267.2 to 1. In the case of progeny 233-1 a perfect fit to a 3:1 ratio was obtained — 12 resistant and 4 susceptible. The data are presented in Table XXIII. Inheritance of Resistance in Fasciated Sweet and Susceptible Alaska Crosses In this case there occurred segregates which seemed to mature some- what earlier than Alaska. Since maturity masks wilt symptoms to a certain extent it is possible that a part of the deficiency of recessives shown here may be due to mistakes in classification as a result of this condition. On the other hand, it is possible that modifying factors affected the result. 24 Wisconsin Research Bulletin 96 Dev. P. E. 0.91 0.29 2.68 2.22 1.03 0.40 2.83 2.76 0.54 1.69 0.87 0.25 0.53 3.70 O (U 0\ro*-*r^T:t-vOsO-HOtoOOON*^l fNr-]fNirr)^rO<^fOfO*^<^'^'^^ C — ' .2^2 1 — « rf ++ 1 1 i + 1 1 1 1 1 1 + 1 Total number plants diseased ■O — to Tt 0> so I--1 O VO t^ (-1 « ^ O 1 ^ r-j --^ fvj r.4 ^ f/^ ro rr, (N» .^ Tj- T^ 1 Total number plants healthy i E 2 Greenhouse Field la lb Ila lib III IV V VI VII VIII Total S 3i c 1 O'^ro'T'i-'O-OOOOOO c v |E a. c 270-2 270-2 270-2 236-3 236-2 237-3 237-2 237-2 237-1 237-1 237-1 236-4 236-4 "(5 2i o 1 :o c/: •>: "T •rj C« t/3 X X K X X ov a> o> -o S S S "s Dev. P. E. OO — OO — O-hOOO—OO fviw^OOOOO*OOror^fOO\OO^fO^ Deviation from 3:1 ratio ^^Of^l^OroOfO.-H-H^fO^'^ Total number plants diseased I Total number plants healthy T^lr)00 progeny, but there is no indication of any mode of inheritance other than the simple monohybrid relationship found in other strains and varieties. In Table XXV may be found the data covering the wilt reactions of this group of progenies. These crosses produced such a small number of seeds that they were not deemed worthy of separate consideration. The great- est significance of this group of progenies lies in the fact that they extend the range of the investigation to additional types of peas. Results of Crossing Suscejjtible Strains and Vai'ieties of Peas The crosses of Susceptible Rogue Alaska x Perfection and of Sus- ceptible Rogue Alaska x Susceptible Alaska gave Fo progenies which were entirely susceptible, under both greenhouse and field conditions. Under field conditions the F2's of Horsford x Susceptible Rogue Alaska and of Susceptible Rogue Alaska x Susceptible Alaska were en- tirely susceptible. Nevertheless, under greenhouse conditions some ap- parently resistant plants occurred. However, these plants all occurred in the same bench and a progeny test showed three of the twelve (205-1) to be homozygous for susceptibility. The other nine plants failed to pro- duce seed. In progeny 251-4 no seed was produced. Susceptible Rogue Alaska x Surprise gave four plants out of 70 with no symptoms of wilt when this Fo progeny was tested in the greenhouse. Table XXVI. — Occurrences of Wilt in F2 Progenies From the Crosses of Sus- ceptible Varieties Total Total Original Progeny Number Place number number crosses number of seeds tested plants plants planted healthy diseased Ro 7 X P 7 230-3 90 Greenhouse 0 71 Ro 7 X P 7 230-2 51 Field 0 41 Ro 7 X P 7 230-1 35 '■ 0 36 Rog 6 X S 16 245-4 90 Greenhouse 0 71 Rog 6 X S 16 245-S 50 Field 0 48 Rog 6 X S 16 245-3 50 " 0 47 H 3 \ Ro 5 205-1 90 Greenhouse 12 * 47 H 3 X Ro 5 205-1 50 Field 0 48 H 3 X Ro 5 205-1 50 ■' 0 46 Ro 3 X S 16 251-4 90 Greenhouse 8 *' 65 Ro 3 X S 16 251-5 33 Field 0 30 Ro 3 X S 16 2S1-S 33 0 32 Rog 6 X Sp 8 244-2 90 Greenhouse 4 *** 66 *These 12 plants all occurred in one bench. A progeny test of three plants showed all three tn be susceptible. **These all occurred in one bench. * ''These all occurred in one bench. Causal organism was isolated from two. 26 Wisconsin Research Bulletin 96 As in the other two cases, these plants all occurred in one bench. Unfor- tunately, no seed was available for a field test. It seems probable that the resistant plants found in these crosses are not genetically resistant but are escapes. The data are given in detail in Table XXVI. ' Results of Crossing Resistant Strains and Varieties of Peas Out of 96 F2 plants from the cross of Horal x Admiral, raised almost to maturity on wilt infested soil, none showed any indications of wilt. This material was tested in the greenhouse on soil selected from places in the benches at which very rapid elimination of susceptible plants had occurred. The susceptible checks for this material wilted completely. The results of the cross of Resistant Alaska x Horal are entirely in harmony with those of the first mentioned cross. The number of plants involved is again small (107) but the tests were made under very severe conditions so that it is quite probable that if any susceptible plants were present they could not have escaped the disease. The results are shown in Table XXVII. Table XXVII. — Occurrence of Wilt in F2 Progenies from Crosses of Resistant Varieties Original crosses Progeny number Number of seeds planted Place tested Total number plants healthy Total number plants diseased Ho 4 X Ad 6 R 11-lS X Ho 6 319-1 320-1 103 116 Greenhouse 96 107 0 0 Linkage Relationships Relationship Between Sugary and Resistance Two Fo progenies were tested to determine if any linkage existed "between the gene for sugary (a simple recessive to non-sugary) and the gene for wilt resistance. These two progenies were from the cross of Horsford x Resistant Alaska. The first (200-1) showed white seedlings but the presence of these apparently does not influence in any way the segregation for wilt resistance. The X^ value of 1.88 with a value for P of 0.60 indicates that there is no linkage to disturb the 9:3:3:1 ratio, as shown in Table XXVIII. Table XXVlll.^Linkage Studies of Sugary and Resistance in F2 Progenies of the Cross Horsford x Resistant Alaska Orisinal cross Progeny numbers Type of seed Number of seeds planted Number of plants healthy Number of plants wilted H 2 X R 11-12 200-1 200-1 200-2 sugary non-sugary sugary non-sugary 75 152 43 103 31 97 31 73 8 38 10 30 X» (9 : 3 : 3 : 1 ratio) — 1.S8 P = 0.60 FusARiuM Wilt Resistance in Canning Peas 27 It is possible that the gene for sugary may modify the expression of resistance but the results given here indicate that it does not. A com- parison of rate of wilting for sugary and non-sugary plants would give a better opportunity to study the effect of the sugary gene on resistance. Relationship Between Height and Resistance After making a few preliminary observations under greenhouse con- ditions, it seemed evident that there existed some linkage between tallness and resistance. F2 progenies from the cross of medium-tall Fasciated Sweet X dwarf Acme were tested under field conditions to determine the amount of this linkage. The results of this study are shown in Table XXIX. Table XXIX. — Linkage Studies of Height and Resistance in F2 Progenies of the Cross Fasciated Sweet x Acme ■ Original cross Progeny numbers Type of plant Number of plants healthy Number of plants wilted Type of plant Number of plants healthy Number of plants wilted Sw4 X Ac3 234-4 234-3 268-1 269-1 269-1 269-3 269-3 269-2 Tall 22 30 108 129 78 106 98 100 671 12 4 32 18 22 17 17 24 Short 2 4 21 20 24 21 17 23 7 8 Sw7 X Acl3 14 S\v8 X Ac9 24 11 16 15 18 Totals 146 132 113 X» (9:3:3:1 ratio) = 78.62 P = .000000 + In using the X^ test, the deviations are so large that P has no positive value to at least six decimal places, and hence the data support the idea of a linkage relation existing between the characters involved. These figures, nevertheless, should be regarded as only approximate since the distinction between tall and short is rather arbitrary under some conditions. The calculated gametic ratio based on F2 results (Emerson, 4) is 2.23:1.00:1.00:2.23. This is equivalent to a crossover value of about 31 per cent between the factor for tall and the factor for resistance. The actual number of plants considered was 817 tall of which 671 remained healthy and 146 wilted, and 245 short of which 132 remained healthy and 113 wilted. On the basis of the calculated gametic ratio the zygotic ex- pectation is 657.5:139:139:126.5, respectively. This is a close fit with X- equal to 2.42 and P equal to 0.47. Rate of Wilting It has been observed that tall plants wilt more slowly than short plants and some varieties more quickly than others. However, there are no grades of resistance and when the total number of plants wilted is con- sidered at the end of a test it is found that elimination of a susceptible variety or strain is always complete, provided that the plants have been 28 Wisconsin Research Bulletin 96 42 44 47 50 53 56 59 62 65 68 Daijs after planting m ur'ilt injected soil fairly exposed to infection. That differences in rate of wilting may be more apparent than real is evidenced by the fact that most plants that are going to wilt during a test can be picked out a few days before wilting actually begins, and that these preliminary symptoms seem to occur in all varieties and strains at almost the same time. The preliminary symptoms referred to are a slight loss of color, with a very slight shrivelling of the basal leaves. It also requires approximately the same time for complete elimination in one variety as in another. The wilt notes under field conditions were taken in such a manner that rate of wilting can be studied] for most of the progenies under obser- vation. The cross selected for consideration of this matter is Horsford x Resistant Alaska. This cross is considered for two reasons. First, the number of plants involved is relatively large, and secondly, a very good fit to the expected 3:1 ratio of resistant to susceptible plants was obtained. The chart above shows the curves plotted for the data at hand. The curves have the same general form with the most rapid wilting from the 44th to 47th day after planting. The short segregates wilt much more rapid- ly from the 42nd to the 44th day after planting than do the tall segregates, and slightly faster than the Horsford checks. The greater initial rate of the short segregates is about three times that of the tall i.e. 26.14 per cent and 8.52 per cent respectively from the 42nd to 44th day after plant- ing. Since all three types are completely wilted at about the same time — 68 days after planting — it is necessary that at sometime during the wilting season that the rate of wilting be reversed with the tall segregates wilting at a more rapid rate than the shorter. This change is shown to occur from the 44th to the 47th day after planting. The percentages are based on 159 short susceptible segregates, 223 tall susceptible segregates, and 122 Horsford checks. It is possible, of course, that the period selected (every third day, except first reading) may be too gross a measure for an accurate study of Fus.\RiUM Wilt Resistance in Canning Peas 29 rate of wilting. The plants were examined in detail on the 42nd day after planting and it was decided that none showed sufficiently advanced symptoms to justify removal from the wilt bed. Since the expression of wilt in a tall plant is somewhat different from that in a short plant, it is possible thatr the two kinds of plants may not really have been in comparable stages of wilt when removal occurred. On a wilt bed in which elimination of susceptibles does not take place so quickly during the first few days of wilting, there may be a better oppor- tunity to compare rate of wilting in the tall and in the short plants. Nevertheless, from the data at hand, it appears probable that the tall plants do really have a somewhat lower initial rate of wilting than do the short plants. This may be attributed to a modifying factor or fac- tors or to the modifying effect of the tall gene itself. Although each variety of peas may differ from other varieties of peas in its expression of wilt, the final results are the same. A study of F3 progenies homo- zygous for the characters under consideration would give much more reliable information than that obtained from F2 results. Discussion and Conclusions The results of the wilt tests cover a wide range of material, includ- ing some of the peas most commonly grown by canners and other strains of importance on account of their wilt resistance. The results clearly indicate that, in the varieties concerned, only a single factor* difference is involved in resistance to F. orthoceras var. jnsi. Most of the material considered is F2 material from crosses of resistant x susceptible strains or reciprocal. It made no difference from the standpoint of ratios obtained whether the material was closely or distantly related. In six crosses some of the material was carried to the F3. These results give further support to the monofactorial explanation. Crosses of susceptible x susceptible, resistant x resistant, backcrosses of Fi to the susceptible parent, backcrosses to the resistant parent and Fi produced results indicative of a monohybrid difference, with resistance dominant to susceptibility. It is of interest to note that of the three Fusarium vascular diseases which have been studied genetically two give monofactorial segregation for resistance (cabbage wilt and pea wilt) while one (flax wilt) shows a more complicated relation. In the case of cabbage the situation is very similar to that in peas, with resistance dominant (Walker, 12) but in flax (Burnham, 3) more than one factor is concerned and susceptibility is dominant. Linkage studies of resistance with sugary and of resistance with height have been conducted. The data indicate that there is no linkage between sugary and resistance. The linkage data for height and resis- tance show that these two genes are rather loosely linked, with a crossover value of above 30 per cent. This crossover value is probably only an approximation since it is frequently difficult to distinguish between genetic short and environmental short plants. 30 Wisconsin Research Bulletin 96 The linkage of height with resistance indicates a linkage between the factors "Le" (White, 13) and "Fw". The factor for height is considered to be "Le" since the two varieties concerned have approximately the same number of internodes but with a difference in length. Due to the destruc- tive nature of the wilt disease this factor may be of but little value in linkage studies. Since linkage is so loose it should not be difficult to in- corporate resistance in plants of any height. A rate of wilting test showed that in the case of Horsford x Resis- tant Alaska the shorter type of peas wilted at a slightly more rapid initial rate than did the tall segiegates. This result may be due to modifying factors or to the modifying effect of the gene for tallness itself. Further tests of rate of wilting are desirable. While not stated, it is implied that there is in reality but one strain of the fungus or that the hosts react to all the strains in the same man- ner. Under greenhouse conditions a single strain was used, but the field plot had been inoculated with a mixture of many strains, and in both cases approximately the same type of reaction occurred. However, since no effort was made to canvass all sources of the fungus, it is possible that other strains of the fungus may be encountered. In a few cases the expression of the disease has indicated that there may be modifying factors concerned. However, modifying factors must necessarily remain speculative until more satisfactory methods of con- trolling some of I the pathological phases have been worked out. Since the environment is so variable under field conditions, it seems that further work should preferably be done in the greenhouse. Under field conditions, we have to consider that confusion may arise from sudden changes in temperature, heavy rains causing erosion and elimination of some plants, and complications arising from other diseases. Under greenhouse con- ditions there are also many difficulties but temperature variations and losses of plants from causes other than wilt can probably be better con- trolled than in the field. From the results obtained in this study there are indications that resistance to Fusarium wilt can be readily combined with other desirable characteristics and that it should be possible in the course of a few years to incorporate resistance in the valuable commercial varieties without sacrificing either yield or quality. It is, of course, possible! that in other cases diffei'ent factors for resistance may be involved; but since such a wide range of varieties and strains was tested for genetic relationships, it seems highly probable that the information gained in a study of the peas mentioned in this paper should be readily applicable to other varie- ties of peas. Since resistance to wilt has proved to be a simple Mendelian dominant the method of obtaining wilt resistant peas of a desirable type is cor- respondingly simplified. The first step in breeding disease resistant peas should, of course, be to ascertain if the desired resistance does not already exist in the chosen variety. This can be quickly determined by planting the variety on severely infested soil and examining the survivors for type. In case no resistant type plants' ai'e secured, it will be necessary FusARiuM Wilt Resistance in Canning Peas 31 to resort to crossing with a resistant variety. In hybridizing, it is desir- able to make crosses between varieties as nearly alike as possible in order to simplify selection. The Fi plants should be well spaced so as to secure a large yield of seed. The Fi may be grown on either infested or disease free soil. The F2 plants should be grown on infested soil where the sus- ceptible plants will be eliminated. The seed should be harvested separ- ately from each seemingly desirable Fo plant. This seed should be planted separately according to parent plant, on infested soil where the F3 fami- lies that are homozygous for resistance may be readily detected. If desir- able plants are found in any homozygous resistant progenies, the segre- gating progenies may be discarded. It is usually advisable to bulk the seed from the segregating progenies and hold foil a few years. In case it should be necessary to start over again, this seed should be at least equivalent to F2 seed. After securing the desired homozygosity for resistance the next step is to fix the type of the new strain. Since type is a much more complex thing than resistance, it will probably require many more generations to secure exactly the type desired. This may be done by making single plant selections from each generation or by bulking the seed and making selec- tions after several generations. The peas by this time should tend to be homozygous for most characters since they are a self-fertilized crop. Probably the most important thing to observe about the foregoing plan is the fact that it is not necessary to consider wilt resistance after the third generation if the work has been carried along properly. After this time, the attention should be given to other characters. This plan is, of course, subject to many modifications. For instance, the Fi plants may be backcrossed to the type parent, and the process carried through as above, considering the backcross generation as equiva- lent to an Fi. This procedure may hasten somewhat the attainment of the desired type. Summary Data have been presented to show that resistance to Fusarium wilt in the peas studied is due to a single dominant factor. For the most part, the data strongly support the single factor hypothe- sis. There are some exceptions which are explained in the text as possible masking of wilt sjrmptoms by early maturity and escape of susceptible plants due to an irregular distribution of inoculum. The new factor for resistance is designated "Fw" and its recessive allelomorph for susceptibility as "fw." Linkage studies of resistance with two other factor pairs were carried out. These indicated no linkage between resistance and sugary, but a loose linkage of about 31 per cent between resistance and tallness. A rate of wilting study indicated that tall F2 segregates from the cross of Horsford x Resistant Alaska wilted at a somewhat lower rate than did the short segregates. The greatest difference was found in the initial stages. 32 Wisconsin Research Bulletin 96 Literature Cited ( 1) Bateson, W., and Pellew, C. On the genetics of "rogues" among culinary peas. Jour. Gen. 5:13-36. 1915. ( 2) Brotherton, W. Further studies on the inheritance of "rogue"' type in garden peas. Jour. Agr. Res. 24:815-852. 1923. ( 3) Burnham, C. R. Studies on the inheritance of wilt resistance in fiax. Doctor's thesis on file, University of Wisconsin Library. 1929. ( 4) Emerson, R. A. The calculation of linkage intensities. Amer. Nat. 50:411-420. 1916. ( 5) Giltay, E. Plantenleven, 2:147-150. Groningen, 1918. ( 6) Jones, F. R., and Linford, M. B. Pea disease survey in Wisconsin. Wis. Agr. Exp. Sta. Res. Bui. 64, 30 p. 1925. ( 7) Kappert, H. 1st das Alter der zu Kreuzungen verwandten Individuen auf die Auspragung der elterlichen Merkmale bei den Nachkommen von Einfluss? Biol. Zentralblatt 42:223-231. 1922. ( 8) Linford, M. B. A Fusarium wilt of peas in Wisconsin. Wis. Agr. Exp. Sta. Res. Bui. 85, 44p. 1928. ( 9) Pea diseases in the United States in 1928. U. S. D. A. Plant Disease Rpt. Suppl. 67, 14 p. 1929. (Mimeographed), (10) Sirks, M. J. Die Verschiebung genotypischer Verhaltniszahlen innerhalb Populationen laut mathematischer Berechnung und experimenteller Priifung. Med. Landbouwhoogeschool Wageningen 26, verh. 4, 40 p. 1923. (11) Tochinai, Y. Comparative studies on the physioloy of Fusarium lini and Col- letotrichum lini. Jour. College Agr. Hokkaido, Imperial University 14:171-236. 1926. (12) Walker, J. C. Inheritance of Fusarium resistance in cabbage. Jour. Agr. Res. Vol. 40, No. 7. April 1, 1930. Studies of inheritance in Pisum II: The present state of knowledge of heredity and variation in peas. Proc. Amer. Phil. Soc. 56: 487-589. 1917. (14) Inheritance studies in Pisum III: The inheritance of height in peas. Mem. Torrey Bot. Club 17:316-322. 1918. Research Bulletin 107 February, 1931 Resistance to Fusarium Wilt in sarden^ canning and Field peas J. C. Walker Agricultural Experiment Station of the University of Wisconsin, Madison Contents Introduction 1 Sources of Materials 2 Method of Experimentation 2 Experimental Results 3 Studies in the Alaska Variety 3 Studies in the Perfection Variety 6 Studies in Other Garden and Canning Varieties 7 Studies in Field Peas 11 Discussion 13 Sununary 14 Literature Cited 15 Acknowledgments J5 Resistance to Fusarium Wilt in garden, canning and field peas THE FIRST RECORD of the Fusarium wilt of peas was made from material collected in Wisconsin by Jones and Linford (2) in 1924. A full description of the disease was later published by Lin- ford (3) in which the causal organism was established as Fusarium ortho- ceras App. and Wr. var. pisi. A comprehensive survey by Linford (4) in 1928 definitely established the disease as present not only in Wisconsin but also in Maryland, Pennsylvania, Ohio, Michigan, Indiana, Illinois, Mon- tana, and in the upper Snake River Valley of Idaho. It has recently (1930) been found by B. L. Wade^ and by L. K. Jones^ to be a serious disease in the Palouse section of eastern Washington and northwestern Idaho. Linford (3) pointed out that a number of the standard varieties of can- ning peas were highly resistant to the disease while others were very sus- ceptible. Wade (6) studied the inheritance of wilt resistance in a number of pea varieties and concluded that it is due to a single dominant factor. Thus he found that pea plants may be classified into two discontinuous classes, resistant and susceptible, and that intermediate grades of resistance to wilt did not occur in the materials he studied. He pointed out the im- portant possibility of combining wilt resistance with desirable type through hybridization and selection. Since Fusarium wilt is increasing in distribution and severity in the pea growing areas of Wisconsin its control through disease resistance is be- coming more important each year. Certain of the canning varieties now in use have been tested as to their resistance or susceptibility. Some have not been tested while in others the observations have been confined to a rela- tively few stocks. Furthermore, many varieties used commonly for fresh peas have never been studied as to their reaction to this disease. The pur- pose of the present investigation has been to secure a wider knowledge of the proportion of susceptible and resistant plants in a larger number of standard varieties. The inquiry has been directed with the three following objects in mind: (1) To test a large number of stocks of the canning va- rieties most widely used in Wisconsin, especially Alaska and Perfection, both of which have been listed as susceptible, and to ascertain whether or not certain stocks contain resistant desirable-type individuals in percentages sufficient to warrant improvement through selection. (2) To determine which, if any, of the standard garden varieties are resistant, since it may be reasonably expected that this disease may soon become of importance to the home and market gardener. (3) To inquire into the possibility of the existence of resistant varieties which may prove of value to the canning industry either in their present form or as parents for hybridization with other more desirable susceptible plants. ^Correspondence 2 "Wisconsin Research Bulletin 107 Sources of Materials The materials used in the trials reported herewith were from three gen- eral sources. (1) Several seed growers submitted samples of the numerous varieties listed by each. This list, therefore, represents a cross section of the major varieties grown at the present time for the canner and for the retail seedsman. (2) Some five hundred samples were secured through Dr. E. J. Renard from various canning companies of Wisconsin and from seeds- men supplying the canning trade. The majority of these was about equal- ly divided between the two varieties, Alaska and Perfection. This list gave a comprehensive representation of the various lots of seed being used by the canners of the state and gave an opportunity to determine the variation in resistance among stocks of canning varieties. (3) Several hundred samples were submitted by Dr. D. N. Shoemaker of the United States Department of Agriculture. This group contained a much wider range of varieties than the first two groups and included many which are little used in America or are entirely foreign stocks. Method of Experimentation In order to carry out reliable comparative tests with such a large num- ber of samples it was necessary to use a field which was uniformly and thoroughly infested with the wilt organism and relatively free from other pea diseases which might complicate the readings as to the wilt disease. The field selected was one on which peas were a complete failure because of wilt in 1928 according to the observations of Linford. A preliminary study of the extent and severity of infestation was made in 1929. One acre was selected and planted with the Perfection variety. Four bushels of seed were drilled in, as for the canning crop, which yielded aprpoximately 500,000 plants. Wilt appeared with equal promptness throughout the en- tire plot and slightly less than 500 widely scattered plants remained healthy. Each of these was more or less off-type. The progenies from thirty of them, when tested upon wilt soil in 1930, all showed complete or relatively high resistance to wilt, indicating that most of the survivors were probably resistant individuals and not escapes. In any case the number of survivors was so low in the 1929 plantings that the value of this field as a testing ground for wilt resistance was clearly established. In the 1930 trials duplicated plantings were made of each sample in groups 1 and 3 noted above, and single plantings of those in group 2. A double four foot row was made in each planting, approximately fifty seeds being used in each test. Check plots of Perfection were scattered through- out the plot and, as in 1929, there was complete elimination of this variety by wilt except for the survival of an occasional off-type plant. In both 1929 and 1930 root rot (Aphanomyces eniciches Drechsler) was rarely encountered and tlicn only in very mild form so that the readings for wilt were not impaired. Other pea diseases were negligible in amount. Resistance to Fusarium Wilt 3 Experimental Results The 1930 plot was planted on May 6 and 7. On May 29 a few very slight indications of wilt were apparent, but the symptoms of the disease increased rapidly after that date. On June 19 a large percentage of the susceptible plants were dead. As the season progressed the clear differentiation of plants into two discontinuous classes, resistant and susceptible, was striking- ly evident throughout this wide range of stocks and varieties and coincid- ed with the observations of Wade (6) in a more limited group. A com- parison of the duplicate plantings of given samples in different positions in the plot showed very uniform reactions as might have been expected from the general infestation found in 1929. In order to secure definite ratings of the various samples, the actual number of diseased and healthy plants in one plot of each lot was recorded, with certain exceptions noted later. It would have added to the value of the data if both plots of each sample had been counted where duplicated plantings were made. Since time did not permit of this, the second planting of each sample was examined but not actually counted. In no case 'was any appreciable difference in the reaction of the counted sample and its duplicate noted. The first disease count was made on June 19 and the days immediately following. A second and final count was made on July 1 and 2. On the basis of these counts the percentage of plants remaining healthy (resistant) was calculated. In the tabulation of results (Tables II, III, IV, and V) the varieties are grouped according to the scheme of classification used by Hedrick (1). To simplify their examination they are arbitrarily divided into two classes, resistant and susceptible. Those showing 50 per cent or more survivors are placed in the resistant column, those showing less than 50 per cent survivors are placed in the susceptible column. The rating, i.e., the percent- age of resistant plants of each sample, is placed in parentheses following the name of the variety. In many varieties more than one sample was tested ; this is indicated by the number of rating figures following each variety. There were a few instances where the various samples received under a given name varied so widely in percentage of resistant plants, that it became necessary to place a variety in both the resistant and the susceptible classes. In such cases the ratings are placed at opposite points in the two columns. Studies in the Alaska Variety Of the early varieties Alaska is the most widely used for canning in Wisconsin and at present comprises about forty per cent of the total acre- age. Linford (3) classed Alaska as a susceptible variety although he did find some resistant individuals in the lots he tested. All progeny tests from survivors showed complete resistance on infested soil. With one exception, however, these selections differed distinctly in type from the parental variety. On the other hand he noted that Alcross, a very uniform variety developed by Delwiche from a cross between two individual Alaska plants contained about 50 per cent resistant plants. One sample trial of this 4 Wisconsin Research Bulletin 107 variety furnished by a seedsman showed 44 per cent resistant plants in the 1930 trials. The practicability of developing resistant pure lines from Al- cross has been demonstrated by Renard whose resistant selections from this variety served as parental stocks for a considerable portion of Wade's (6) genetic studies. Linford (3) found a strain of Alaska selected by Prof. C. E. Temple of the Maryland Agricultural Experiment Station to be resistant. What is probably a continuation of this same strain, submit- ted to us by a seedsman as Maryland Alaska, was included in the 1930 trials and was found to contain 97 per cent resistant individuals (Table II). Insofar as could be judged from gross morphological characters it is a good uniform Alaska. In a variety as popular as Alaska with the canning industry which re- quires uniform stock of desirable quality, considerable effort has been di- rected by seed growers in the improvement and maintenance of stocks. Even though such improvement work may not have been done with par- ticular reference to resistance to wilt it seemed not at all unlikely that much variation might occur in the percentage of resistant individuals in various stocks. In order to secure a cross section of the Alaska stocks now in use in Wis- consin, samples of seed lots delivered to about fifty canners by various seedsmen for 1930 planting were secured through Dr. Renard. This list in- cluded 217 samples. It may be assumed, of course, that many of the samples came originally from one stock, but no attempt was made to trace the history of the stock represented by each sample. In addition to these, 26 samples were secured from other sources, chiefly seedsmen. Each sample was given a four- foot plot test in the 1930 trials. There was not sufficient time to make complete counts of every sample in this group. Seventeen representative samples were counted and the data secured are to be found in Table II under the Alaska group. It is to be seen that the proportion of resistant plants ranged from none in some samples to 100 per cent in another sample. All of the samples were exam- ined and, without actual counting of individual plants, they were divided into five classes according to the percentage of resistant survivors. The classes and the number of samples which fell in each are given in Table I. Even though this is a rough classification it is important in showing that an appreciable number of the samples showed relatively high percentages of resistant plants. In only one sample were all plants resistant and unfor- tunately it contained a number of distinctly off-type plants. In a large percentage of the samples in classes 3 and 4 the survivors appeared to be good type Alaskas. The samples which fell into these two classes were quite generally distributed throughout the various sources. At the end of the 1930 season the seed from eleven of the Alaska stocks containing relatively high percentages of survivors in the Waupun plot was collected and mixed. Some of the soil from the plot was removed to the greenhouse and seed from this lot planted in it along with a known homo- zygous susceptible stock and a known homozygous resistant stock of the Resistance to Fusarium Wilt 5 same variety. After a period of six weeks, 100 per cent of the 49 plants in the susceptible lot had wilted and died. In the known resistant lot all of the 49 plants were healthy. From the seed saved from the survivors on the Waupun plot 203 plants were grown and none of them became affected with wilt. This shows that, if not all, at least a high percentage of the survivors in the lots tested were homozygous for resistance since no seg- regation into resistant and susceptible classes took place. Table I — Relative Resistance to Wilt in Various Samples of Alaska Variety Tested Class number Approximate percentage of resistant plants Number of samples Percentage of samples 1 2 3 4 S 0 1-25% 26-50% above 50% 100% 40 112 58 32 1 16 46 24 13 1— Total 243 100 The present situation with regard to the Alaska variety may thus be briefly summarized as follows. A great variation prevails among various stocks as to the proportion of plants resistant to Fusarium wilt. Certain stocks contain practically 100 per cent resistant plants, as the sample of Maryland Alaska tested, for instance (Table II). The sample of Del- wiche's Alaska No. 19 contained about 76 per cent resistant plants while the sample of Alcross contained about 44 per cent, (Table II). Other stocks vary from 0 to 75 per cent or more. With the basis of inheritance of resistance established by Wade's (6) investigations, the opportunity and method are clearly indicated whereby seedsmen may and should bring their stocks up to contain a high percentage of resistant individuals. This may be done by selection from survivors of good type on thoroughly and uni- formly wilt-infested soil. Single-plant selection from such survivors is, of course, advisable but it naturally involves a protracted period of increase. On the other hand where a stock is already very uniform in type and shows fifty per cent or more resistant individuals the procedure may be shortened by mass selection. This should be done by planting the seed stock in larger quantity on thoroughly and uniformly infested soil for one or more generations until most or all of the susceptible individuals have been elim- inated. The practicability of this latter method is strengthened by the fact that the partially resistant stocks appear to consist largely of homozygous resistant and homozygous susceptible individuals and very few hetero- 6 Wisconsin Research Bulletin 107 zygous individuals. In this connection the limitations of mass selection in a rogue-infested stock should be emphasized. The work of Renard (5) shows that after the rogue content of a stock reaches a certain point elim- ination of off-type becomes almost impossible because of the fact that the "rogue germplasm" is being constantly transferred to type plants by cross- pollination before the rogues are removed. Where mass selection is used it should not be looked upon as a substi- tution for the improvement of stocks through the more refined methods of the developing pure lines through single-plant selection, but merely as a temporary expedient to serve during the interval required to bring up stocks by the latter method, improvement by which requires five to ten years to secure sufficient seed for commercial use. When it is used, certain pre- cautions are advised. The stock should be of high grade, i.e. of desirable type, uniform, and free from off-types. The fact that it has a relatively high percentage of homozygous resistant individuals should be determined by a preliminary test on wilt-infested soil. The field to be used should be known by previous observation or testing to be thoroughly and uniformly infested with the wilt organism, and located in a region where the climate and soil are such as to favor the prompt and severe development of wilt. These conditions are absolutely essential in order to secure the development of wilt in most or all of the susceptible individuals with sufficient rapidity to insure their elimination before any seed is set. Complications due to the prevalence of root rot and other diseases of pea in the soil should be avoided. Obviously this plan will best succeed with the help and advice of someone sufficiently familiar with pea diseases to give reliable diag- nosis of wilt and differentiation from other diseases. Studies in the Perfection Variety Perfection at present is the major wrinkle-seeded variety used by Wis- consin canners. It has already been stated that the entire trial field com- prising one acre in 1929 was planted with this variety. The seed in this case did not come from a single stock since it consisted of remnants of a large number of samples submitted by canners and seedsmen to Dr. Renard for type studies. Less than one-tenth of one per cent of the plants were resistant and none of the survivors were true Perfection plants. In 1930, 199 samples from canners and seedsmen were included in the trials on the wilt plot. All proved to be very susceptible. Out of a total population of around 10,000 plants, only 23 survived and none of these were of true Per- fection type. It thus appears that all the stocks of Perfection represented in this wide range of samples are practically homozygous for susceptibility to wilt. The possibilities for selection of resistant lines within the variety are small. A few of the survivors of the 1929 plot were preserved for progeny tests. Most of those tested so far are completely resistant but on the other hand most of them are also sufficiently different from Perfection to be of doubtful value for the canner. Certain of these lines are being increased Kesistance to Fusarium Wilt 7 for further study as to their adaptability to canning. At the present time it appears that the development of a resistant Perfectipn type suitable for the present needs must apparently depend in a large measure upon h3bridi- zation of standard Perfection with other resistant varieties and reselection for a combination of Perfection characters and wilt resistance. Progress in the development of a new variety by this method is necessarily slow and requires a period of several years. Studies in Other Garden and Canning Varieties As already stated, one of the major purposes of this investigation was to gain more information as to the occurrence of resistant plants in a wider range of varieties than has hitherto been tested. The results of the trials with garden and canning varieties are compiled in Tables II, III, and IV. As these varieties are grouped together according to the Hedrick system (1) it may be seen at a glance that with the exception of the Ad- vancer and Champion of England groups the samples in every group range from those in which all plants were susceptible to those in which all were resistant. One of the limitations of these results, of course, is the fact that in manj' varieties the tests are confined to a single sample. This is particularly true of relatively new or little used varieties. Their ratings are included because of their possible interest in the future to seedsmen and growers and to those searching for resistant parent stocks for hybridization. The number of samples of a given variety tested is a fair index to its present popularity, since several specialists in pea seed were asked to sub- mit all samples of varieties which they were listing for sale. From what has already been shown from tests with Alaska it is to be expected that different stocks listed under a given name may vary in per- centage of resistant individuals. These differences are further increased by the variation in the conceptions of seedsmen as to the ideal of a given variety. Furthermore it is not uncommon for seedsmen to substitute under one name stock of another closely similar variety. In certain varieties variation from a high percentage of susceptible plants to high percentage of resistant individuals was found. This is to be seen in addition to Alaska, already mentioned, in Extra Early, Early Bird, and Ameer (Table II), in Hundredfold, English Wonder, and World's Record (Table III) and in Sherwood, Admiral Beatty, Quite Content, and Prizewinner (Table IV). In contrast to these, certain widely used varieties consistently showed high percentages of susceptible plants such as : Large White Marrowfat (Table II) ; Gradus, Duchess of York, American Wonder, Premium Gem, Notts' Excelsior, Surprise, Little Marvel, World's Record, Laxtonian, Laxton's Progress, Onward, Horsford, Sutton's Excel- sior, Advancer, and Perfection (Table III) ; Thomas Laxton, and Rice's 13 (Table IV). Among the most noteworthy of the varieties consistently high in percentage of resistant plants are Black eyed Marrowfat, Rice's 330, Horal, Harrison's Glory (Table II) ; Allan's Canner, Green Admiral, Rogers K, Improved Surprise, Prince of Wales, Everbearing, Dwarf Tele- 8 Wisconsin Research Bulletin 107 phone, Yorkshire Hero, and Dwarf Champion (Table III) ; Stratagem, Dwarf Defiance, Champion of England, Telephone, Alderman, Prince Ed- ward, and most of the edible-podded varieties (Table IV). From this list the varieties now used by Wisconsin canners, in addition to Alaska and Perfection discussed earlier, may be classified as follows : Susceptible Resistant Canners' Gem Green Admiral Horsford Yellow Admiral Ashford Roger's K Surprise Prince of Wales Badger Senator Rice's 13 Bruce Winner Horal Laxtonian Rice's 330 Onward Improved Surprise Thomas Laxton ^ c^ P-l cMk A. BADGKR, ALL PLANTS KILLKD WITH WILT; B. ROGERS K, ALL PLANTS RESISTANT TO WILT. iy . *^%-^ u: ':jr. B Plate III,-A. SENATOR, A CANNING VARIETY, ALL PLANTS RESISTANT TO WILT; B. f*IONEER, A GARDEN VARIETY, ALL PLANTS KILLED WITH WILT; C. SUR- PRISE. A CANNING VARIETY, 96 PER CENT KILLED WITH WILT. WAUPUN WILT PLOT, 1930. ^*^-> ^ < '-^5 'A Resistance to Fusarium Wilt 9 Table II. — Resistance and Susceptibility to Fusarium Wilt in the Extra Early, Tom Thumb, Alaska and Dimple-Seeded Groups Susceptible Resistant Extra Early Group Extra Early (0)» (0) Extra Early (98) (100) Mammoth Pod Extra Early (0) Prince Albert (100) (90) Sutton's Early Champion (2) First of All (96) Ringleader (0) Caractacus (29) Giant Lightning (0) First on Market (27') First and Best (0) (24) Saxonia (0) Early Dwarf or Prince Arthur (0) ' Tom Thumb Group Bishop Long-pod (0) Tom Thumb (100) Marrowfat Group Large White Marrowfat (7) (29) (0) Improved Sugar Marrowfat (100) (2) (0) ' Very Dwarf White Marrowfat (100) Kmver Marrowfat (0) Early Greeir- Seeded Marrowfat (100) Melting Marrowfat (0) Webb's Advancer Marrowfat (100) Springtide (0) Webb's Stourbridge Marrowfat (97) Daniel's Matchless Marrowfat (93) Favorite Marrowfat (100) Wonder Marrowfat (83) Blackeyed Marrowfat (100) (100) Alaska Group Alaska (0) (0) (0) (0) (2) (7) (29) Alaska (55) (57) (64) (65) (67) (71) (100) (42) (47) (48) Alcross (44) Kentish Invicta (0) (0) Maryland Alaska (97) Earliest of All (12) Alaska No. 19 (76) Eclipse (2) (4) Rice's 330 (100) Winner (0) (0) (0) Earliest Blue (68) Kentish (0) Rapide (52) Express (0) Velocity (0) Dimple-Seeded Group Early Bird (0) (0) Early Bird (55) Ameer (0) (0) (0) (3) Ameer (100) Radio (0) (0) Charlton's Radio (88) Pr™o Pilot (0) Harrison's Glory (100) (100) (100) Pilot (0) (0) Johnson's New Glory (100) Eight Weeks (0) (0) Old England (94) Leader (0) (0) Britisher (100) Blue Peter (0) Councillor (100) Fillbasket (0) Pride of the Market (0) Bountiful (0) Claudit (0) Acquisition (S) (0) Superb (0) » Eldorado (0) Talisman (0) Benefactor (0) "The numbers in parentheses after each variety represent the percentage of re- sistant plants in the individual samples tested on the Waupun wilt plot. 10 Wisconsin Research Bulletin 107 Table III. — Resistance and Susceptibility to Fusarium Wilt in the Wrinkle Cream-Seeded, Gem, Large-Podded Dwarf, and Advancer Groups Susceptible Resistant Wrinkled Cream-Seeded Group Hundredfold (0)» (0) (0) Reading Wonder (6) Marvellous (0) Daisy (0) Gradus (0) (0) (0) (0) (26) (32) Duchess of York (0) (0) Prestige (0) (11) Edward VII (0) Hundredfold (96) Sutton's Harbinger (100) Lincoln (92) Allan's Canner (100) Green Admiral (100) (100) (100) (100) (98) Yellow Admiral (98) (100) (100) Rogers K (100) (100) Horal (88) (100) (100) Profusion (100) Champion of Scotland (100) Prince of Wales (100) (100) (98) (100) Langport (100) Early Giant (100) Sweet Market (100) Gem Group English Wonder (yellow seed) (7) Surprise (0) (0) (3) (4) (4) McLean's Little Gem (0) American Wonder (0) (0) (0) (0) (0) (29) Premium Ciem (0) (0) (0) (0) (4) (12) Xott's Excelsior (0) (0) (0) (2) (2) (2) (41) Little Marvel (0) (0) (0) (0) (4) (23) (26) Canner's Gem (0) Delicious (0) (0) (0) Prince Arthur (0) Melbourne Market (0) iMiglish Wonder (green seed) (95) Surprise (Improved) (100) Everbearing (1»0 (100) (100) (100) (100) (100) (97) (97) William Hurst (97) Large-Podded Dwarf Group World's Record (0) (0) (2) Laxtonian (0) (O) (0) Pioneer (0) (0) Peter Pan (0) (0) Blue Bantam (0) Marchioness (0) (0) Laxton's Progress (0) (0) (0) (0) (0) Onward (0) (0) Horsford (0) (0) (0) (0) (0) (0) Ashford (0) Sutton's Excelsior (0) (0) (0) (0) (0) (0) (2) (6) Reading Gem (0) Buttercup (0) Sutton's Supreme (0) Sensation (0) Renown (0) (31) Top O' Th" Morn (0) (0) Mayflower (0) Daffodil (0) Green Gem (0) Early Duke (16) Worlds Record (58) (70) Discovery (100) Dwarf Telephone (100) (100) (97) (96) (86) Early Morn (73) Yorkshire Hero (100) (97) Dwarf Champion (100) (100) (98) British Wonder (100) Matchless (100) Sutton's Perfection (100) Giant Stride (100) Arcadian (100) Dwarf Prolific (90) Ideal (61) (69) (96) Victor (lOOJ (93) Lancashire Lad (74) Bruce (100) Commander (100) Advancer Group Advancer (0) (5) (13) Perfection (0) (0) (0) (2) Abundance (0) Veitch's Exonian (2) Delicacy (0) •The luimbers in parentheses after each variety represent the percentage of re- sistant plants in the individual samples tested on the Waupun wilt plot. Resistance to Fusarium Wilt 11 Table IV. — Resistance and Susceptibility to Fusarium Wilt in the Stratagem, Champion of England, Ne Plus Ultra, Telephone, Senator, and Edible-podded Groups Susceptible Improved Stratagem Sherwood (0) Resistant Stratagem Group (0> Stratagem (100) (100) Sherwood (100) (100) Dwarf Defiance (100) (100) (87) Potlatch (100) Battleship (76) Majestic (81) Champion of England Group Champion of England (100) (100) (97) (96) (95) (93) Ne Plus Ultra Group Thomas Laxton (0) (0) (0) (0) (0) (0) (0) (0) Magnum Bonum (0) G. F. Wilson (4) Snowdrop (0) Captain Cuttle (0) Prince of Peas (9) Dreadnought (0) Ne Plus Ultra (92) (89) Juno (100) (100) King Edward (92) Liberty (83) Omega (98) Telephone Group Admiral Beatty (0) Carter's Quite Content (0) Sutton's Prizewinner (0) Up to Date (0) Market Gardener (34) Good Indeed (5) Lord Kitchener (7) Premier (0) Exhibition (26) Centenary (12) Victory (14) Admiral Beatty (100) Quite Content (100) (100) Prizewinner (95) (100) Telephone (100) (100) (95) Duke of Albany (100) (89) Admiral Dewey (100) Alderman (100) (100) (100) (100) (100) (100) (98) (95) Duke of York (11) Duchess (100) American Champion (100) Harvestman (94) Wm. Richardson (100) Lord Leicester (97) Amateur Pride (93) Maincrop (100) Bell (100) Prince Edward (100) (100) (100) (94) Light -Podded Telephone (100) (97) Standard (100) Senator Group Gladstone (42) Delicatesse (0) Unique (0) Rearguard (0) Union Jack (0) (0) (0) (30) Rice's 13 (0) (0) (23) (3) New Giant Butter (0) Senator (100) (100) (100) 100) Shropshire Hero (100) Heroine (55) Eureka (smooth seeds) (100) Eureka (wrinkled seeds) (53) Maincrop (100) Glory of Devon (100) Advance Guard (100) Sutton's Best of All (68) Edible-Podded Group Giant Luscious Sugar (100) Melting Sugar (100) (100) (97) (93) Dwarf Grey Sugar (100) (100) (100) (100) Giant Butter (100) French Sugar (100) • The numbers in parentheses after each variety represent the percentage of re- sistant plants in the individual samples tested on the Waupun wilt plot. 12 Wisconsin Research Bulletin 107 Studies in Field Peas A large number of field pea samples were included in the Waupun trials of 1930. The results with a portion of the better known varieties are listed in Table V. The range in percentage of resistant plants is quite as wide as in garden varieties. In Canada Field two samples showed 25 and 95 per cent resistant individuals, respectively, while Victoria varied from all sus- ceptible in one sample to all resistant in two other samples. It therefore should not be assumed that field peas are generally resistant. The possi- bilities for improvement in resistance are as great here as in garden varie- ties when and where the needs justify the attention of the plant breeder. Table V. — Resistance and Susceptibility to Fusarium Wilt in Field Peas Susceptible Resistant Cream (White) — Seeded Section Canada Field (25) a Canada Field (95) Canada Beauty (0) White Canada (100) Gregory (90) Arthur (98) Golden Marrow (100) Colorado Field (76) Victoria (0) (22) Victoria (100) (100) Cream-seeded, Black-eyed Section Canada (97) Green- Seeded Section Dwarf Blue Imperial (100) Bangalia (100) Blue Prussian (100) Dark- Seeded Section Kaiser (0) Delano (100) Carleton (15) Solo (98) (100) Pcluschka (12) French Grey (100) Killarney (30) Large White Capucijner (100) (98) Austrian Winter (21) Alaplc Partridge Brittany (7) "The numbers in parentheses after each variety represent the percentage of re- sistant plants in the individual samples tested on the Waupun wilt plot. Resistance to Fi'sarium Wilt 13 Discussion Because of the increasing importance of Fusarium wilt as a limiting factor in pea production in Wisconsin the matter of wilt resistance must be considered in any well-balanced program of pea improvement. The pri- mary purpose of this paper has been to bring together, for the use of the plant breeder as well as the seedsman and the canner, such information as has been obtained regarding the proportion of plants resistant to this dis- ease in a wide range of stocks and varieties. The writer wishes to make it clear that the data presented should be prop- erly interpreted as indicating percentages of resistant plants only in those samples tested. It is hardly possible to give a final rating of any variety as a whole. The wide variation in the stocks of such a variety as Alaska, where a large number of samples were studied, naturally raises the ques- tion as to whether in such a consistently resistant variety as Alderman, for example, further sampling from a larger number of sources would not show a greater range in the proportion of resistant and susceptible plants. For this reason stocks of varieties to be used in improvement work should be carefully tested beforehand on wilt-infested soil. The data presented refer to resistance to a single disease of pea, the Fusarium wilt, as found in a representative field of southern Wisconsin. Similar studies may well be extended to the disease as found in other re- gions. Resistance to wilt in a given variety does not imply similar resist- ance to any of the other numerous diseases of pea. In the discussion already given with regard to Alaska and varieties derived from it, such as Alcross and Alaska No. 19, the opportunity to build up highly resistant satisfactory stocks through selection has been pointed out. The survey of Perfection stocks shows that resistant plants true to type in this variety are very rare. The possibility of developing a resistant Per- fection by selection alone is therefore quite remote and improvement in resistance in this varietj- will therefore naturally depend in a large measure upon hybridization with other resistant varieties and reselection along the lines already outlined by Wade (6) and others. The same may be said for those garden and canning varieties which are shown to be consistently susceptible since in these as well as in Perfection the occasional resistant survivor usually has been found to be an off -type plant. 14 Wisconsin Research Bulletin 107 Summary The proportion of plants resistant or susceptible to Fusarium wilt was studied with a large number of varieties of garden, canning, and field peas by means of field trials on thoroughly infested soil. Variation from 100 per cent susceptible plants to 100 per cent resistant plants was found in a survey of 243 samples of Alaska variety. Trials of 199 samples of Perfection showed all to contain very high percentages of susceptible plants. In addition to these two popular canning varieties numerous garden, canning, and field peas were studied. Many varieties were consistently re- sistant while others were quite uniformly susceptible. A few varieties showed wide variation in the reaction of individual samples. The facts presented are of importance to the plant breeder and seedsman in connection with pea improvement work, and to the canner whose choice of stocks and varieties may be influenced by the occurrence of Fusarium wilt in his territory. Resistaxce to Fusarium Wilt 15 Literature Cited (1) Hedrick, U. P., Hall, F. H., Hawthorn, L. R., and Berger, Alwhi. Peas of New York. In Vegetables of New York Vol. 1, pt. 1. 132 p. 1928. (2) Jones, F. R. and Linford, M. B. Pea disease survey in Wisconsin. Wis. Agr. Exp. Sta. Res. Bui. 64, 31 p. 1925. (3) Linford, M. B. A Fusarium wilt of peas in Wisconsin. Wis. Agr. Exp. Sta. Res. Bui. 85, 44 p. 1928. (4) Pea diseases in the United States in 1928 U.S.D.A. Plant Disease Rpt. Supplement 67, 14 p. 1929. (Mimeo- graphed). (5) Renard, E. J. Origin and nature of rogues in canning peas. V.'iS. Exp. Sta. Bui. 101, 56 p. 1930. (6) Wade, B. L. Inheritance of Fusarium wilt resistance in canning peas. Wis. Agr. Exp. Sta. Res. Bui. 97, 32 p. 1929. Acknowledgments The writer is indebted to Mr. W. C. Snyder for assistance in the execu- tion of the field trials ; to Mrs. Anita R. Sammet for assistance in the assembling and checking of the data; to Dr. E. J. Renard for advice and the supply of many samples ; to Dr. D. N. Shoemaker for samples of his valuable and extensive collection of pea varieties ; and to the various seeds- man who kindly furnished samples of peas for this investigation. These studies were supported in part by financial assistance from the University Research Fund. Research Bulletin 111 September, 1931 Wisconsin Studies on Aster Diseases and Their Control Agricoltaral Experiment Station of the University of Wisconsin, Madison CONTENTS Importance of Aster Diseases 1 Symptoms of Aster Yellows and of Aster Wilt 4 Aster Yellows 4 Aster Wilt 4 Aster Wilt: Its Cause and Control 16 The Cause of Aster Yellows 5 The Control of Aster Yellows 8 Aster Wilt: Its Causes and Control 16 The Cause of Aster Wilt 16 The Control of Aster Wilt 17 Summary 34 Literature Cited 36 Wisconsin Studies on Aster Diseases and Their Control L. R. Jones and Regina S. Riker TWO DISEASES in particular threaten the culture of the China aster in the United States. These are the yellows, due to a virus, and the wilt or stem rot, caused by a Fusarium. Most aster plantings are troubled to a greater or less degree by one or both of these diseases. The two diseases are sometimes confused. They are, however, maladies caused by different agencies, and each requires specific precautions for control. This bulletin points out cer- tain well established facts as to the occurrence, symptoms, and causes of these diseases, and discusses especially results with specific con- trol measures. The China aster is one of the choicest annual flowers of late summer and early autumn, both for the home gardener and for the commercial florist. When the two diseases are clearly disting- uished, and the mode of introduction and spread of each is under- stood, ordinary garden culture of the China aster may, with proper attention to seed, soil, and rotation, be carried on with satisfaction. For the commercial florist, however, the situation is moi-e difficult. Therefore, the present studies upon control of aster yellows and aster wilt have been concerned especially with the conditions and problems of the commercial florists, particularly of Wisconsin and the adjacent territory. Brief reports upon progress in this work have already appeared (Jones and Riker 1928, 1929, 1930, Weiss 1929, Ball 1930). IMPORTANCE OF ASTER DISEASES BOTH ASTER yellows and aster wilt are widespread in the United States. A generation or even a decade ago the China aster was considered as one of the most popular and reliable of the annuals. Year by year these two plagues have spread and in- creased until few entirely healthy aster plantings can be found. In many instances both diseases are present. Figure 1 indicates the dis- tribution of the two maladies in this country, as far as it has been reported. The severity of the two diseases varies with different localities. Either may be present to a sufficient degree to practically destroy an aster plantation. The two occurring together, as they commonly do. are therefore doubly discouraging. This is evidenced in Table I, which analyzes the damage from these two maladies in the horticul- tural^ garden at Madison, Wis., in 1927. It will be noted that although ^ The authors are indebted to their colleagues in the Department of Horticulture for co- operation in this and other ways during the progress of these investigations. Wisconsin Research Bulletin HI there were four types of China aster, including 15 varieties, the loss with most ranged from 95 to 100 per cent and the best of them yielded only 10 per cent of healthy plants at the end of the season. Figure 2B illustrates the destructiveness of wilt in a susceptible variety. More- over, in the strain relatively resistant to wilt (Figure 2A), many of the plants were infected with yellows, although this does not show ^^ Yellows ^ Wilt and Yellows I I Nciiher disease reported FIG. 1.— WIDESPREAD DISTRIBUTION OF ASTER YELLOWS AND ASTER WILT The distribution shown is recorded mainly in the Yearbook of the- United States De- partment of Agriculture and in the Plant Disease Bulletin and Reporter. in the photo^aph. As a result, this commercial aster field at Ran- dolph, Wis.,2 which had been planted for both cut flower and seed pur- poses, was considered by the grower a total loss commercially. Such conditions are typical of many aster plantings throughout the country whether in the home garden or in the commercial cut flower or seed field. Indeed, the aster seed industry in places in the eastern United States has been practically destroyed by the combination of yellows and wilt. It is especially significant, therefore, that both diseases have appeared in the chief aster seed growing centers of the Pacific re- gions (Figure 1) and seem destined to increase unless control meas- ures are taken. These two diseases, aster yellows and aster wilt, are thus wide- spread, and seriously menace aster culture in the United States. ' In this connection the authors wish to acknowledge the generous cooperation during the last three years of the J. W. Jung Seed Co. of Randolph, Wis., in granting the free use of 'aster-sick' soil and in other ways. During this period courteous advice or as- sistance with seed has been given by other commercial aster growers, including G. J. Ball, West Chicago, 111.; Bodger Seeds, Ltd., El Monte, Cal.: H. L. Cady. Fox Lake, Wis.; Vaughan's Seed Store, Chicago, 111.; and H. B. Williams, Baldwinsville, N. Y. Aster Diseases and Their Control Table I. — Aster Diseases in the Horticultural Garden at Madison, Wisconsin, in 1927 Total Per cent Per cent Per cent Variety number of healthy of wilted of yellowed of plants plants plants plants Ostrich Feather soft pink 59 7 22 71 vermilion carmine 51 2 71 27 light blue 31 0 48 52 Average 3 47 50 American Branching white 31 0 100 flesh pink 26 4 42 54 crimson 28 4 39 57 lavender 112 1 56 43 dark violet 32 3 47 SO Average 2 S7 41 Cbego rose pink 21 10 38 52 cnmson S3 2 79 19 lavender 46 3 IS 82 purple 19 0 47 S3 Average 4 4S 51 King rose 27 3 41 56 lavender 62 S S5 40 violet 41 0 SI 49 Average 3 40 48 FIG. 2.— COMMERCIAL ASTER PLANTING AT RANDOLPH, WISCONSIN, OCTOBER 1927. The soil is heavily infested with the aster wilt Fusarium, i.e., it is aster-sick. The destructiveness of wilt with a susceptible variety is shown in the Royal lavender pink in the right foreground (B). Individual plant resistance to wilt appears within this variety. Differences in resistance to the wilt in aster strains is apparent especially between the sus- ceptible one mentioned above and the relatively wilt-resistant Comet-like selection from Peerless shell pink in the left foreground (A). However, many plants in the latter selection which escaped the wilt were attacked by the yellows so that the field was considered a total loss commercially by the grower. From this relatively wilt-resistant strain, individual plants were saved whose progeny, by ref)eated selection, have attained considerable wilt resistance. (See later details and Figures 12 and 13.) Wisconsin Research Bulletin 111 SYMPTOMS OF ASTER YELLOWS AND OF ASTER WILT ASTER YELLOWS and aster wilt which, as mentioned earlier, are so widespread as to occur either singly or together in al- most every large aster plantation, are often confusedly re- garded as a single malady. Yet they differ in cause, in dissemination, and in response to control measures. Therefore, it is essential to recognize the characteristic symptoms of each. ASTER YELLOWS Aster yellows has been described in American literature from the early writings of Smith (1902) to the recent publications of Kunkel (1926b, 1927) and others. Therefore, only a summary of the essen- tial points is given in this bulletin. The symptoms of aster yellows are in part illustrated in Figures 3 and 4. The most important changes in the diseased plants are in habit and color. Often the changes are more or less one-sided because of the location of the infection from which the disease started. There may be little change in the older mature leaves or other parts of an infected plant but the subsequent growth shows disease symptoms. In general habit the infected plant becomes more constricted and straight (Figure 3D), the upper leaves especially tending toward an upright position, and remains dwarfed (Figure 4C). In the young leaves a slight yellowing or "clearing" appears along the veins. In all the new growths there is a tendency to abnormal increase in the number of branches, giving a "bunchy" or "rosetted" appearance (Figure 40). These later shoots are yellowish, with the leaf petioles elongated and leaf blades narrowed. Follovdng infection the flowers do not develop normal color or size. The flower heads tend to be dwarfed, greenish in color, and often show one-sided deformity. Al- though the diseased plants never recover, most of them live through- out the season, unless the disgusted grower pulls and destroys them. This should, indeed, be done at the first appeax'ance of disease since each such sick plant is a center of continued disease dissemination. ASTER WILT Aster wilt in America was probably first i-ecorded by Galloway (1896). However, within the last decade it has received especial at- tention, particularly from Beach (1918), Jackson (1927), and Weiss (1925, 1929) who have included descriptions of this disease. It, therefore, again suffices to give a brief summary of the distinguish- ing characters. The symptoms of aster wilt are illustrated in Figures 3, 4, 5, and 6. Plants may be destroyed by the wilt disease at any period from the seedling stage to full bloom. In very young seedlings the symptoms are similar to damping off. In plants with only a few leaves the Aster Diseases and Their Control entire seedling will wither quickly and die. With plants which are attacked by the wilt at a somewhat older age one finds two types of symptoms. The one, which seems associated with a one-sided and slow invasion of the parasite, gives a stunted growth with a one-sided development of the plant and leaves, and a decided yellowing of the leaves or parts of leaves most stunted (Figures 3B, and 5). These symptoms suggest those of the closely related cabbage yellows dis- ease. When the stem of such an aster plant is cut the vascular ring is found to be brown, particularly on the side most affected. Eventu- ally such plants wither. The other type of aster wilt seems associ- ated with a more complete invasion of the fungus. In this type, the lower leaves first show signs of wilting. This is followed more or less rapidly by the collapse of the entire plant which then withers and dries. The stem in these plants is usually externally blackened at the base and for some distance up, thus giving rise to the name stem rot (Figures 3C, and 6). If a cross section of the stem is taken a general browning of the vascular elements is evident. In some cases, vdlt may not become apparent until the plants are coming into full bloom when they suddenly collapse and wither due to the demand at that time for a larger supply of moisture (Figure 4B). The con- trast between the wilt and the yellows disease thus becomes striking at blooming time (Figure 4B and C). With wilt there may be a sud- den collapse and death of plants even when full of normal blossoms. With yellows, although the symptoms of one-sided or complete dis- coloration of flowers as well as foliage may be serious, the plant re- mains stiff and unright and often lives to the end of the season. The above discussions and illustrations give the essential charac- teristics of aster yellows and aster wilt and enable one to distinguish between the two diseases. ASTER YELLOWS: ITS CAUSE AND CONTROL THE CAUSE OF ASTER YELLOWS THE CAUSE of aster yellows must be understood in order to insure intelligent control. Aster yellows was first described by Smith (1902) who, in discussing its nature says, "Caused by no fungus, insect or other organism, not due to any apparent effect of treatment or environment, it is notwithstanding a sharply defined, widespread and destructive disease of this plant." Smith noted simi- lar symptoms on marguerite, Calendula, African marigold, and rag- weed or Roman wormwood. He also suggested the similarity of this disease to peach yellows and to "calico" of tobacco. It remained, how- ever, for the recent painstaking researches of Kunkel (1924, 1925, 1926a, 1926b) to demonstrate certain virus characteristics of the aster malady. Kunkel found that, although asters are attacked in nature by various insects such as aphids, tarnished plant bugs, and several species of leaf hoppers, the dissemination of this virus from sick to Wisconsin Research Bulletin 111 Aster Diseases and Their Control m o ■" 5 I < -r Wisconsin Research Bulletin 111 ^ \ f ,^ 4! ^ FIG. S.— ASTERS SHOWING WILT SYMPTOMS, "ONE-SIDED" TYPE, FROM experimental PLOTS AT MADISON, WISCONSIN, JULY, 1928. Three plants are illustrated which were arrested at slightly different stages of normal development by the invasion of the aster wilt Fusarium. The type of symptoms shown here seems to be associated with an unequal and slow invasion of the fungus. Eventually such plants wither and resemble those in Figure 6. healthy plants was accomplished experimentally only by a single species of leaf hopper, Cicadula sexnotata Fall., of the insects tried. Apparently this virus is not carried over winter in the aster seed nor in the eggs of the infectious leaf hoppers. Probably all aster seedlings early in the season are free from yellows infection, and all leaf hoppers when hatched in the spring are free from the virus. The yellows virus may, however, infect any one of many plants be- sides aster, including Roman wormwood, horseweed, lettuce, plantain, and dandelion. Some of the many hosts are biennials or perennials common about fields or garden borders and in these the disease may persist from year to year. By feeding upon a diseased biennial or perennial the leaf hopper may become infected in early summer and then, after an incubation period of at least ten days, if such an insect feeds upon an aster plant, the disease may be transmitted. To keep aster plants free from the yellows it is necessary to keep them free from the attacks of such leaf hoppers. THE CONTROL OF ASTER YELLOWS The essential factor in the control of aster yellows is, therefore, to keep the viruliferous leaf hoppers from the aster plants. Theo- retically, at least, several methods might be considered. (1) Plants capable of carrying the virus over winter might be eliminated from the vicinity of aster plantings. (2) Repulsion or destruction of the Aster Diseases and Their Control leaf hoppers might be accomplished. (3) The aster plants themselves might be shielded from the attacks of the viruliferous leaf hoppers. The first two methods have proven difficult in practice (Weiss 1925, 1929, White 1931). The third method has been found successful both by Kunkel and by the present writers. Kunkel (1929a) protected the aster plants by the use of fences. He found that a screen of 18 wires to the inch prevented the passage of the leaf hoppers. He, therefore, surrounded his experimental aster plantations (10 by 25 feet) with wire screen fences of this mesh but of various heights (4, 5, 6, and 8 feet). When careful and persistent roguing of all yellowed plants was practiced, loss from the disease within the enclosures was 20 per cent, whereas 80 per cent of the unfenced check plants was diseased. Weiss (1929) suggests a fence five or six feet high as most desirable. Williams (1930) reports a re- duction of yellows, by the use of 18 mesh wire screens five feet high, from 60 per cent outside to 16.9 per cent inside the enclosures. In the Wisconsin studies, control of aster yellows was undertaken simultaneously with that of aster wilt in 1925. In view of the success- ful control of various Fusarium diseases, particularly cabbage yellows, FIG. 6.— ASTERS SHOWING WILT SYMPTOMS, "WILT" OR "STEM-ROT" TYPE, FROM EXPERIMENTAL PLOTS AT MADISON, WISCONSIN, JULY, 1928. Three plantj are illustrated which were arrested at slightly different stages of normal development by the invasion of the aster wilt Fusarium. The type of symptoms shown here seems to be associated with a more complete invasion of the fungus than that pro ducing the symptoms in Figure 5. No unequal development of parts appears but a more or less sudden wiltinK of the plant occurs, frequently accompanied by a blackening of the lower portions of the stem. 10 Wisconsin Research Bulletin 111 by wilt-resistant varieties, this method of approach was taken with the aster, as will be discussed later. However, the losses in out-door plots in the horticultural garden in previous years because of the yellows were so disturbing that, when the control of the wilt was at- tempted, the control of the yellows was also tried. Control of aster yellows by small cages Aster yellows was controlled successfully by enclosing the ex- perimental plants in small cheesecloth-covered cages. These were of the simple type commonly used for insect exclusions in virus studies, about two and one-half feet high by four by six feet on the ground (Figure 7). Such cages proved fully effective in preventing the yel- lows in 1925 and the two succeeding seasons. Every year twelve cages were prepared, each covering 42 plants. In addition an uncovered adjacent plot was planted with 200 plants. Not a plant was lost from yellows under such cages in the three years' trials, although the vir- uliferous leaf hoppers were so abundant outside that a large ma- jority of the unprotected aster plants were stricken by the yellows each year. By the summer of 1928, the Fusarium-resistant strains discussed later had reached such numbers that the use of the smaller cages for their trials was no longer practicable and a new means of protecting the plants from the viruliferous leaf hoppers was needed. Two gen- eral methods were tried. The one was the use of fencing as tried by FIG. 7.— TYPE OF SMALL CAGE USED TO EXCLUDE THE VIRULIFEROUS LEAF HOPPERS IN EXPERIMENTS ON THE CONTROL OF ASTER YELLOWS FROM 1925 TO 1929. During the first three years coverings of cheesecloth (38 x 44 threads per inch) were employed, and perfect control of yellows was obtained. In 1928 and 1929 like cages were constructed with coverings of cloth of various meshes. Any mesh coarser than 22 x 22 threads per inch proved ineffective as a control for yellows (Tables 2 and 3). Madison, Wisconsin, 1927. Aster Diseases and Their Control 11 FIG. S.— EXPERIMENTAL ASTER PLANTATION AT MADISON, WISCONSIN, 1928. The prominent feature is the large cage or house (6+ by 30 by 98 feet) completely covered with cloth of 22 x 22 threads per inch. To the left and rear are the small cages (2^ by 4 by 6 feet) covered with cloths of different mesh, and adjacent open plots for controls. To the side and right of the large cage is the enclosure (6+ by 10 by 49 feet) with open top, and sides covered with cloth of 22 x 22 threads per inch. Around! the lat- ter, asters were also planted in the open for controls. Kunkel (1929a). The other was the employment of the caging method on a large scale. Partial control of aster yellows by fencing The fencing method was tested in 1928. Rectangular enclosures were constructed with cloth-covered side walls somewhat over six feet high, and open tops. These were adjacent to the large completely cloth-covered cages which will be discussed later. In choosing the cloth the advice of tobacco-shade cloth experts was sought and a trial use of a cloth having 22 x 22 threads per inch was decided upon partly because of its availability and partly because of the assurance courte- ously given by Kunkel that in his experience a wire screen of 18 meshes to the inch served to exclude the hoppers although they passed through coarser screens. In an enclosure 10 by 49 feet in size (Figure 8) with open top and sidewalls of cloth of 22 x 22 threads per inch at Madison, Wis., yellows appeared in 44 per cent of the plants which survived the wilt as against 90 per cent of yellows in an unfenced control plot. In a similar enclosure 45 by 90 feet, at Randolph, Wis., 82 per cent of the non-wilt plants showed evidence of yellows while 95 per cent of the non-wilt unfenced plants developed yellows. In these enclosures roguing was practiced every week. It may be noted that there was a considerably greater percentage of yellows in the larger enclosure. However, too much significance must not be at- tached to the comparative figures since the enclosures were not in 12 Wisconsin Research Bulletin 111 the same locality. Kunkel's (1929a) greater success with the fenc- ing method may be in part attributed to more prompt roguing of infected plants and possibly in part to his use of smaller enclosures. Allowance must also be made in any such comparisons for differences in the relative prevalence of leaf hoppers in a locality and possibly in the character of the adjacent vegetation. Control of aster yellows by large houses The use of the caging method on a large scale in 1928 (Figure 8)' proved more successful under experimental conditions in Wisconsin than the fencing method. The effectiveness of these enclosures was proven by the results in two large cages or houses, one at Madison and the other at Randolph (6+ by 30 by 98 feet and 6+ by 16 by 98 feet respectively), each completely covered with cloth 22 x 22 threads per inch. Not a single case of yellows developed within them until after midsummer. In August a night storm tore openings of considerable size in the tops of both cages. Although the damage was repaired within a few hours, early symptoms of yellows showed about two weeks later on three out of 2,160 plants in one of the houses and on two out of 1,080 plants in the other house. These were promptly re- moved and no other yellowed plants appeared. Evidently a few vir- uliferous hoppers entered through the torn top during the brief ex- posure. The side walls, some six feet high, had not been torn by this storm. Hence, if the above interpretation is correct, the hoppers found their way very quickly over the tops of such side walls but were elim- inated by prompt roguing. In 1929 similar results were obtained with cloth 22 x 22 threads per inch in two large cages (6+ by 30 by 98 feet at Madison, and 6+ by 60 by 98 feet at Randolph). Cages of like size were used in 1930. In that year no yellows de- veloped within the enclosure at Randolph. At Madison two plants showed early symptoms of yellows within three weeks after planting. They were promptly removed but in spite of roguing, 4 per cent of the plants showed some symptoms of yellows by the end of the sea- son. The plantings in previous years and at Randolph that year were made some days after the cloth house was constructed. At Madison, however, in 1930 the planting immediately followed the completion of the cloth house. Whether or not a time interval after construction serves to starve out any viruliferous leaf hoppers which might be enclosed during construction remains to be seen. In this connection it should be noted that in several smaller houses at Madison, even with coarser top coverings, where planting was not done for two weeks after construction, no cases of yellows developed until relatively late in the season. The above experiments indicate that large houses covered com- ' A brief description of the construction of these first large cages is given by Weiss (1929). Modifications in detail, based on experience, may be found in an article by Ball (1930). John E. Luddy, The Windsor Company, Windsor, Connecticut, who has had much experience in the construction of similar rloth shades for tobacco culture, will gladly furnish suggestions or specifications for aster houses of desired sizes. Aster Diseases and Their Control 13 pletely with cloth of 22 x 22 threads per inch are generally effective in excluding the viruliferous leaf hoppers and thus in controlling aster yellows. In practice it seems safest to erect and cover the house sev- eral days in advance of the transplanting. Mesh of cloth in control of aster yellows In order to secure information as to the effectiveness in excluding the leaf hoppers of the various other grades of screening cloths in the market an experimental series of the smaller cages ( 2 ^/^ by 4 by 6 feet, Figure 7) was set up in 1928, each cage covered with a different cloth. Asters were grown under these on Fusarium-free soil along- side the larger cloth-covered house at Madison (Figure 8). Table II shows the results with these smaller cages, each completely enclosed with the cloth of mesh indicated. Cheesecloth, as in the preceding three years, and cloth of 22 x 22 threads per inch, as in the large houses, were effective in excluding the leaf hoppers. Cloth of the next degree of fineness tried, 12 x 12 threads per inch, and all coarser cloths admitted the viruliferous leaf hoppers. In 1929 a similar trial was made using cloths of grades intermediate between 22 x 22 and 12 x 12 threads per inch, since these were the grades respectively effective and ineffective in the 1928 trials (Table II). The results are given in Table III. Results were similar to those of the 1928 trials with the same grades of cloth, and the new intermediate grades proved ineffective in fully excluding the leaf hoppers. Simultaneously one of these intermediate grades, 22 x 16 threads per inch, was used for top and sides in a house 6+ by 10 by 49 feet. Within this enclosure 22 per cent of yellows appeared in the plants Table II. — Yellows Control in Small (Jloth-Covered Cages, 1928 Mesh of cloth Plants with yellows Threads per square inch Per cent 38 X 44* 0 22 X 22 0 12 X 12 30 10 X 10 S3 10 X 8 67 8x8 47 8x8 67 Not caged 73 ^Cheesecloth Table III.— Yellows Control in Small Cloth-Covered Cages, 1929 Mesh of cloth Plants with yellows Threads per square inch Per cent 22 X 22 0 22 X 16 30 16 X 12 44 12 X 12 54 Not caged 71 14 Wisconsin Research Bulletin 111 escaping the wilt. The entrance of the viruliferous leaf hoppers into this enclosure corroborates the results in the small cages with cloth of the same mesh. It would seem that any coarser cloth than 22 x 22 threads per inch is relatively ineffective when used for both top and sides. The possibility was also tried of employing a coarser cloth for the top while using cloth 22 x 22 threads per inch for the sides. In 1929 a cage 6+ by 10 by 49 feet was constructed with top of cloth 16 x 12 threads per inch and sides 22 x 22 threads per inch. Within this enclosure, 6 per cent of yellows developed in the plants not affected by wilt. Further study of the modification of top covering, coupled with roguing, was made in 1930. Four cages 6+ by 10 by 24% feet were constructed at Madison with sides of cloth 22 x 22 threads per inch and tops of cloth of the following meshes: 8 x 8, 12 x 12, 12 x 16, and 16 x 22 threads per inch. Within these enclosures the per cent of yellows developing in the plants not affected by wilt were 14, 8, 6, and 3.6 per cent respectively. In the open, 96 per cent of plants not af- fected by wilt showed symptoms of yellows. Thus some of the coarser top coverings, while permitting the entrance of leaf hoppers, gave relatively satisfactory control. Commercial control of aster yellows by fencing and caging Concerning the practical aspects of the control of aster yellows, the experience of one commercial firm, the J. W. Jung Seed Co.i Randolph, Wis., may be cited. This firm was interested in trying out the use of screening cloths on a practical scale in 1928, 1929, and 1930. The first year, 1928, all the cloth used was 22 x 22 threads per FIG. 9.— commercial ASTER CAGE OR HOUSE SUCCESSFUL IN CONTROLLING THE YELLOWS This house was completely covered with cloth of 22 x 22 threads per inch. (J. W. Jung Seed Co., Randolph, Wisconsin, 1929.) A profitable commercial crop was obtained. Aster Diseases and Their Control 15 inch. Part of their plantation was enclosed by six foot side walls only, i.e., without top cover; part was covered completely like the experi- mental houses. Where the top was open the yellows disease appeared so abundantly that they despaired of roguing it out, and very little benefit resulted. In the plants which were grown under the complete covering of cloth no yellows occurred. Almost equally significant was the fact that the flowers were of superior quality. This same company in 1929 grew a considerably larger proportion of their asters under cloth (Figure 9). Again they lost practically all those unpro- tected, whereas those under full protection showed practically no yellows. Again the flowers grown under cloth were of superior quality for cutting purposes — topping the price in the Chicago cut flower mar- ket toward the end of the season. In 1930 no yellows developed in a large house completely covered with cloth 22 x 22 threads per inch up to August first. At that time a severe storm removed the top. This was not replaced and subsequently a considerable percentage of yellows appeared. This was late enough not seriously to affect the cutting of flowers except in the very late varieties. In two other cages with sides of 22 x 22 threads per inch and tops of 8 x 8 and 12 x 16 threads per inch there developed, respectively, 27 and 16 per cent of yellows. Thus the results secured commercially by the J. W. Jung Seed Co., corroborate those obtained experimentally at Madison and Randolph. Corroborative evidence is also found in the experiences of another commercial firm near Chicago (Ball 1930). DiscTission and conclvsions on the control of aster yellows The conclusions to date as to the control of aster yellows may be briefly summarized. The control of the disease is dependent upon- shielding the plants from viruliferous leaf hoppers. Effective shield- ing materials are cloth not coarser than 22 x 22 threads per inch, as shown in Wisconsin trials, and wire screens of 18 meshes to the inch, as shown by Kunkel (1929a). The difference in mesh in the two materials may be easily explained by the fact that with cloth the threads often spread irregularly, thus evidently requiring a closer original mesh than suffices with the more stable wire screen. Two types of shields have been tried, i.e., fences and cages. Kunkel (1929a) reports encouraging results with the use of open top and screened sidewalls coupled with careful and persistant roguing. Wil- liams (1930) likewise records fair success with fences. Under con- ditions of Wisconsin experimental and commercial trials the degree of control by this method was not commercially satisfactory. However, in Wisconsin experimental and commercial trials aster yellows was practically controlled by the use of cages or houses cov- ered, top as well as sides, with cloth not coarser than 22 x 22 threads per inch. This was the most satisfactory control tried. However, the use of a cloth top as coarse as 12 x 12 or 16 x 12 threads per inch, with cloth walls of 22 x 22 threads per inch has shown promise. Such 15 Wisconsin Research Bulletin HI modification of the top covering, with roguing if disease appears, seems worthy of further attention. Based on the data above the following recommendation for com- mercial culture of asters in the north central section of the United States is made: 1. The construction of complete cloth-covered enclosures rather than open top ones. 2. The employment of cloth not coarser than 22 x 22 threads per inch for the side walls of such enclosures. 3. The use of the above mesh cloth for the top cover unless further tests show somewhat coarser cloths to be effective and more desirable. ASTER WILT: ITS CAUSE AND CONTROL THE CAUSE OF ASTER WILT ASTER WILT, as has been stated earlier in this paper, is due to an entirely different cause than aster yellows. As previously noted, aster wilt was probably first mentioned in America by Galloway (1896). He recognized that it was produced by a fungus similar to the one attacking cotton, watermelon, cabbage, and other plants. Woods (1899) and others ascribed the disease to a Fusarium. It was left for Beach (1918) to make and report the first detailed study of the causal organism which he named Fusarium conglutinans Woll. var. callistephi because he found it so closely related to the cabbage yellows organism F. conglutinans Woll. Jackson (1927) agreed with Beach as to the pathogenicity of F. conglutinans v. cal- listephi but also listed four other Fusarium strains pathogenic to asters. One of these he identified as F. angustum Sherb. while the .other three were not named. In the Wisconsin studies on aster wilt the original experimental plots at Madison were inoculated partly with Beach's Michigan strain of F. conglutinaTis v. callistephi, and partly with Jackson's On- tario strain of the same fungus. In recent years other experimental plots have been located in an aster-sick field at Randolph, Wis. Fusaria have been isolated from wilting plants both from this field and from the horticultural garden at Madison which in culture (on potato dextrose agar and steamed rice) and in pathogenicity re- semble Beach's original strain. The senior author has also seen the aster wilt disease as it occurs in commercial aster plantings in various localities from New England and New York to California. All the observations made indicate that wilt or stem rot of the China aster in the United States is in general one and the same disease. In this connection it should be noted that there are many refer- ences in foreign literature to a wilt or stem rot disease of aster which seems from descriptions to resemble the American disease. The causal agent has frequently been considered a Fusarium whose identity has not been definitely determined. Several species have been listed as Aster Diseases and Their Control 17 associated with diseased asters but the pathogenicity of these species in general has not been clearly established.* Because of the difficulty in comparing the evidence it has seemed best to restrict the scope of the present bulletin to the aster wilt disease as it occurs in the United States. It is, however, probable that this same Fusarium wilt may occur in foreign countries wherever the China aster has been long cultivated. The fungus of aster wilt may easily be introduced into new local- ities, since it may be seed-borne (Beach 1918, "Marketman" 1921, Gloyer 1924, 1931a, 1931b, Rose 1925, Weiss 1925, 1929, Jackson 1927, Gregory 1929, White 1931, Williams 1931). This probably accounts for the widespread distribution of the disease in the United States (Figure l).^ The Fusarium of aster, following its introduction, seems to per- sist indefinitely in the soil much like its close relative the cabbage Fusarium. Consequently, where asters are grown in short rotation, the disease rapidly increases until it is calamitous. Control of aster wilt must, therefore, be directed against a seed- borne, soil-inhabiting Fusarium of wide distribution. THE CONTROL OF ASTER WILT The control of aster wilt depends upon the clear recognition of a few simple facts which have already been suggested. These are (1) that it is a specific disease caused by a parasitic fungus, a vascular Fusarium; (2) that this parasite may be seed-borne, therefore it is widely distributed and liable to be introduced into virgin soils; (3) that once introduced it persists indefinitely and may show rapidly increasing severity if asters are replanted on old soils; and (4) that, as with the closely related vascular Fusaria of cabbage and other plants, there may be a natural variation in susceptibility or relative disease resistance as between host individuals or varieties. ^ * The writers are studying the pathogenicity of a number of the organisms isolated from diseased asters in Europe and furnished through the courtesy of Dr. H. W. Wollenweber. These include several strains of Fusarium, Verticillium, and Cephalosporium. Tests on seed germination and very young transplants indicate that several strains besides F. conglutinans V. callistephi are pathogenic to asters at the above stages. Further trials are under way. = Since the aster wilt Fusarium may be carried on the seed and commercial seedsmen are distributing seed promiscuously from country to country, this disease may doubtless soon occur wherever the China aster is grown, subject only to the limitations imposed by en- vironment. In this connection the junior author has shown that high temperatures favor the disease and that continued low temperatures inhibit the disease even though the fun?u» is present. It is probable, therefore, that in certain regions this wilt may never become serious. _ Such a limitation apparently occurs for the similar Fusarium disease of cabbage. Aster wilt, however, the junior author has found, is a somewhat lower temperature disease than that of the cabfeage. 'The literature on aster wilt has numerous discussions of control measures including: (1) seed disinfection (Beach 1918, "Marketman" 1921, Gloyer 1924, 1931a, 1931b, Rose 1925, Weiss 1925, 1929, Jackson 1927, Gregory 1929, Rager 1930, White 1931, Williams 1931)-' (2) rotation (Paddock 1902. Findlay 1917, Weaver 1917, 1924, Beach 1918, Beal 1919* "Marketman" 1921, Muller 1922, Rose 1925, Weiss 1925, 1929, Mumford 1926, White 1931), soil disinfection (Weaver 1917, Beach 1918, "Marketman" 1921 Mumford 1925 1926. 1928, Rose 1925, Weiss 1925, Jackson 1927, Gregory 1929, White 1931); (4) direct p'anting (Beach 1918, "Marketman" 1921, Mumford 1925, 1926, White 1931); (5) harden- ing or outdoor sowing (Smith 1902, Stone and Smith 1902, 1903, Beal 1919, Weiss 1925 Newton 1929, Ball 1930); and (6) sanitation (Paddock 1902, Findlay 1917, Beach 1918 Beal 1919, Weiss 1925, Jackson 1927, Rager 1930). 18 Wisconsin Research Bulletin 111 It is evident, therefore, that the simplest way to escape the dis- ease is by the use of clean seed and clean soil. Home gardeners may continue to succeed in cultivating the aster by careful attention to these matters. They may well save their own seed from strictly healthy plants and then by rotation keep their asters on clean soil. Where asters are started in special seed beds and later transplanted as is so frequently done it is especially important to use clean soil for seed bed purposes. Since commercial seed may carry more or less of the wilt fungus, seed disinfection is a wise precautionary measure.''^ Whenever direct planting of the seed in its permanent site is practic- able it is preferable, since it lessens the danger of seed bed infec- tion, avoids root mutilation and growth check associated with trans- planting, and insures early deep rooting in cool soil. Where the seed bed method is preferred, then, in addition to attention to clean soil, or soil disinfection,^ care should be taken so to harden off the plants as to insure full vigor and consequently quick deep rooting follow- ing transplanting. Finally, if the disease appears, soil contamination and further spread may be reduced by attention to sanitary meas- ures, i.e., the prompt pulling and burning of diseased plants as they appear. Attention to these methods may suflfice to give the home gardener or small grower reasonable freedom from aster wilt. With the large scale commercial grower, however, the hazards remain great, and even with the small grower the difficulties are such as increasingly to discourage aster culture. Obviously the one way to avoid all these troublesome expedients and restore confidence in aster culture is to seek to obtain and to use wilt-resistant strains. The results of the Wisconsin work with the closely related Fusarium disease of cabbage indicated the probability that this would prove successful. This possibility is especially sig- nificant for the florists' use since the avoidance of wilt by rotation is more difficult with florists operating on large areas than with small gardeners. Moreover, this difficulty will be further increased if the practice is adopted of culture under cloth to avoid the yellows dis- ease, as advocated earlier in this bulletin. It is to be noted that the problem of securing disease resistant strains for the florists' trade is simplified because but few types and colors are needed, as will be discussed in detail later. To obtain wilt-resistant strains of China aster has therefore been ^The specific measures for seed disinfection which have been recommended are: (1) mercuric chloride, one-tenth per cent, for one-half hour (Jackson 1927, Weiss 1929, White 1931); (2) same for ten minutes or one-half hour depending upon the presoaking (Weiss 1925, 1929, Gregory 1929, Gloyer 1931a. 1931b): (31 Uspulun or Semcsan, one-fourth per cent, for one-half hour (Jackson 1927, Weiss 1929, Williams 1931); (4) hydrogen peroxide, three per cent, for one and one-half hours (Beach 1918, Rose 1925, Weiss 1925. Gregory 1929); and (5) Dipdust, or Semesan Bel. applied as a dry coating (Weiss 1929). "The specific measures for soil disinfection which Jiave been recommended are: (I) steaming (Weaver 1917, Beach 1918, "Marketman" 1921, Rose 1925. Jackson 1927, Gregory 1929); (2) treatment with mercuric chloride, one-tenth per cent (Jackson 1927); (3) treatment with formaldehyde, the details of application varying slightly (Beach 1918, "Marketman" 1921, Mumford 1925, 1926, 1928, Rose 1925, Weiss 1925, Jackson 1927, Gregory 1929, WTiite 1931). Aster Diseases and Their Control 19 the chief aim of the Wisconsin experiments on the control of aster wilt. Experience with other vascular Fusarium diseases indicated two ways of approach as hopeful. These were (1) the search for out- standingly resistant varieties of asters among the established com- mercial types, (2) the search for more or less resistant individuals in commercial strains that were in general susceptible, followed by repeated reselections from the progeny of these in the hope of de- veloping and establishing wilt-resistant strains. Both of these meth- ods have been in mind and in use simultaneously. For clarity of dis- cussion, however, they will be considered separately so far as prac- ticable. Since in the early years no outstandingly resistant strains of asters were found, studies were begun with the second method, and the historical developments of the work will therefore be made clearer by discussing first the progress year by year through repeated selec- tions starting with relatively susceptible strains. The development of wilt-resistant strains of China asters by selection from susceptible coTnmercial varieties Although few commercial strains of asters are uniformly re- sistant, resistance to the wilt disease among individuals is met with in most of them (Figure 2B). That by conserving the seed from such plants and by selecting from the progeny, wilt-resistant strains of asters might eventually be developed, has been suggested by sev- eral writers.9 In the Wisconsin studies attempts have been made to develop such resistant strains by the selection of individually resistant plants in most part from susceptible commercial strains. In doing so the emphasis has been placed upon what was most desirable to the com- mercial florists. Fortunately, the greater part of the needs in the florists' field may be met by a relatively limited range of types and of colors. The general reliance at present in the Northern Mississippi Valley seems to center about the American Branching type, although the American Beauty with its chrysanthemum-like flower and non-lat- eral habit is meriting increasing attention. In color the white is the favorite, with pink, rose, lavender and purple following in succession. The first selections in the Wisconsin trials were made in the Ameri- can Branching type and the Heart of France. Later attention was paid to the American Beauty. In the East the Royal, in the same range of colors, is in great demand. Therefore the 1930 trials gave increased attention to this type. ° Beach in 1918 said "The selection of resistant plants is probably to be the ultimate means of controlling the disease." Arnold (1919) also suggested selection for resistance. Curtiss (1926) reported that he was saving seed from healthy plants. Jackson (1927) said "doubtless this means must finally be depended upon to control wilt." He secured seed from some resistant plants and found that it stood up better than commercial seed when planted in inoculated and in aster-sick soil. Adams (1928) believes the solution of the problem to be the development of wilt-resistant varieties. Kunkel (1929b) reported high resistance in four strains grown from seed selected by Milbrath and also the selection of seed from the most resistant plants in these promising strains in the hope of still further improving thetri. Gregory (1929). Ball (1930), and White (1931) advise sowing seed from resistant individuals. Skinner (1930) records that, whereas one plant of several hundred Queen of the Market proved resistant in 1927, the progeny of this plant showed 88 per cent resistance in 1928. 20 Wisconsin Research Bulletin 111 Results in 1925 In 1925 in Wisconsin trials at Madison in artificially inoculated soil the commercial aster strains used proved very susceptible to the wilt. American Branching mixed showed 16 per cent resistance to the disease and Heart of France 29 per cent resistance. From twent; of the individual plants which resisted the wilt, enough seed was obtained for experimental purposes in 1926. Fifteen of these were of the American Branching type in a variety of colors and five of the Heart of France type. Results in 1926 In 1926 certain selections showed greater resistance to wilt than did the parent stock planted in 1925. With the American Branching selections, grown in sick soil, stands ranging from 7 to 60 per cent were obtained in 1926. From ten of these fifteen strains, twenty-three reselections were made for trial in 1927. With the Heart of France selections stands ranging from 28 to 50 per cent were obtained in 1926. From three of these five strains, five reselections were made for trial in 1927. Results in 1927 The behavior on sick soil of the selections mentioned above as compared with that of commercial varieties during the season of 1927 suggested that progress was being made in the development of wilt-resistant strains. With twenty-three American Barnching se- lections, stands were 0 to 100 per cent when flowering began and of these, twelve selections showed stands of 50 per cent or more. For convenience 50 per cent has been chosen as an arbitrary division be- tween undesirable and desirable wilt-resistant strains. By the end of the flowering season, following a period of high temperature and excessive drought, more of the plants had succumbed so that only eight selections had stands of 50 per cent or greater at this time. It was observed in this and following years that many of the selections would produce a fine crop of blossoms suitable for cut flower purposes but before the end of the season would show some symptoms of wilt. The latter plants were considered as undesirables for experimental or commercial seed purposes. From eleven of the twenty-three selec- tions tried in 1927, twenty-eight reselections were made for trial in 1928. Three plantings of commercial American Branching gave 3, 11, and 22 per cent stands, respectively, at the end of the season. With five Heart of France selections stands ranged from 14 to 94 per cent when flowering began, four of the five being over 50 per cent. At the end of the blooming season two selections had stands over 50 per cent. Twelve reselections for trial in 1928 were made from four of these five strains. In sharp contrast with the above se- lections, a planting of commercial Heart of France gave a 7 per cent stand at the end of the season. Aster Diseases and Their Control 21 Results in 1928 In 1928 the work was much extended, as explained earlier, and the small cages were supplanted by large cloth-covered enclosures. Plantings were made not only in the artificially inoculated soil at Maaison but in naturally infested soil at Randolph, Wis., both in 1928, 1929, and 1930. It was possible in these houses to use much larger numbers of plants. In 1927 the largest number of plants of a strain tested was 17. In a majority of cases seven or less were used. Chance therefore played a large part in the results obtained. In 1928, 1929, and 1930 the smallest number used of any one strain was 30 plants. When possible 50 or 60, and with the most promising strains, 100 or 120 plants were tested. As a consequence the results in these years are considered of much greater significance than those of previous years. The seed from the most promising strains in 1928 was collected for the first time in bulk as well as by individual plants, thus allowing more extensive trials of these strains in 1929. Bulk se- lections were also made in 1929 for trials in 1930. Since the results at Madison and at Randolph always showed close agreement the data given is an average for the two plantings. The results obtained with certain of the selections were even more promising in 1928 than in preceding years. In this year with twenty-eight selections of American Branching, stands were 29 to 100 per cent when flowering began, with twenty-five selections 50 per cent or over. Late in the season, again after a period of high tempera- tures and drought, ten selections had stands of 50 per cent or over. Reselections for trial in 1929 were made from these ten strains either by individual plants, bulk, or both. Four plantings of commercial American Branching gave 3, 6, 8, and 8 per cent stands, respectively, at the end of the flowering season. New individual plant selections were made from the commercial strains in the experimental plots in this year. In Heart of France, with twelve selections, stands were 48 to 98 per cent when flowering began with eleven selections over 50 per cent. At the end of the blooming season, nine selections had stands over 50 per cent. Five of these strains were collected in bulk for trial in 1929 and an individual plant selection made from the best strain. A planting of commercial Heart of France gave a 7 per cent stand at this time. The plantings in 1928 included not only the first series of strains but also certain selections made in 1927 from commercial plantings on naturally infested aster-sick soil. Of those collected at Randolph, stands of the Comet-like selection from Peerless pink (Figfure 2A), and of Royal purple were promising. Of those collected at Fox Lake, a strain of white American Branching did well. In 1928 a beginning was made with the American Beauty variety. Commercial seed of this variety gave the following results at the end of the season: white 3 per cent, pink 9 per cent, rose 1 per cent, lavender 15 per cent, and purple 5 per cent resistance to wilt. The 22 Wisconsin Research Bulletin 111 seed from the plants resisting the disease was collected for trial in 1929. Results in 1929 In 1929 the larger cloth-covered enclosures both at Madison and Randolph were again employed. Figure 10 shows the interior of the aster cage or house at Randolph. For the first time certain of the wilt-resistant strains which had been collected in bulk were tested, not only in Wisconsin but also in other localities, notably California. The results obtained in this year and 1930 will be given in greater detail than those of the preceding years in order to show clearly the present status of the work. The data from Wisconsin will be con- sidered first. In the case of the American Branching type (Table IV) twenty- two out of twenty-seven selections gave at least a 50 per cent stand when flowering began and sixteen maintained this degree of re- sistance at the end of the blooming season. Of these sixteen strains two were of white. Of the pink there were two shades in which selec- tions showed 76 per cent resistance. A commercial strain comparable to one of them showed a 25 per cent resistance to wilt. Of rose, the next most desirable color from the florists' point of view, one selec- FIG. 10.— EXPERIMENTAL ASTER PLANTING ON ASTER-SICK SOIL AT RANDOLPH, WISCONSIN, 1929 The interior of the large cage or house (.6+ by 60 by 98 feet) completely covered with cloth of 22 x 22 threads per inch is shown. Differences in susceptibility to the wilt disease is apparent between the two commercial strains in the right foreground (B) and the comparable selected strains to their left (A). (Similar contrasts appear in Figures 12 and 13.) Aster Diseases and Their Control 23 Table IV. — Wilt Resistance at the End of the Flowering Season in the American Branching Type in 1929 and 1930 1 1929 1930 Wisconsin California Wisconsin California Color and strain per cent per cent* per cent per cent" White 40 78 70 79 90 40-2 92" 64 70 40-2-2 77b 40-2-S 90" 45 86" commercial 4 Pale Pink 13-1-1-5 76" 71 90 13-1-1-5-2 58" Pink 39-4 76»> 43-1 76" K. (Kunkel) 89 80 77 70 .commercial 25 30 Light Rose 21-1-2-1 83" 37 60 21-1-2-1-1 66b Deep Rose 20-2-1 48 0 + ' 60 20-2-1-1-1 SS" 20-2-1-1-2 15b 20-2-2 39 10 60 20-2-2-4 5 7b 20-2-2-5 32b commercial 14 14 Red 26-S-4-a 83 70 90-f 26-5-4-b 85 80 904- 26-5-4-1 75'> 26-5-4-2 98b 84 90 26-S-4-2-1 87b 26-5-6 71 80 76 90 + Pale Lavender 25-1-3 71 10 60 25-1-5 47 0 + 70 25-1-5-3 58 Deep Lavender 30-1-1 53 40 70 30-1-1-6 61" 30-1-1-7 1 61" 42 si 90 commercial 62 45 Purple 37-1 840 37-1-1 S9b 37-1-2 68b 37-1-4 78b commercial 30 a These figures are estimates provided by Miss Bodger and Dr. Jagger. b Individual plant selection. tion showed 83 per cent resistance. A commercial strain of crimson showed 14 per cent resistance. Of red in the same shade as the Heart of France but with the more desirable American Branching plant habit (less branching and less of yellow-centered flowers) the five strains tried showed over 50 per cent resistance. Of a pale lavender, two strains showed promise. Of a deeper lavender, three selections 24 Wisconsin Research Bulletin 111 gave 53, 61, and 61 per cent stands. A commercial strain of this shade proved equally promising. Of purple, one plant selection proved relatively resistant. Twelve of the above sixteen strains originated from five of the selections of 1925. Thus there were at hand in the American Branching type promising Wisconsin selected wilt-resistant aster strains in the range of colors (white, pink, rose, lavender, and purple) desired by the florists' trade. These strains, in general, were much more resistant than the comparable commercial stock. Two other promising strains of American Branching appeared in the 1929 trials. One of these was courteously supplied by Kunkel (1929b) who, in 1928, had found a commercial strain of Semple pink of considerable promise. This strain proved excellent in resistance (Table IV). The other promising strain was the commercial Ameri- can Branching lavender mentioned above which gave 62 per cent resistance. In Heart of France in 1929 (Table V), seven out of eight selected strains gave over fifty per cent resistance both when flowering began and later in the season, as compared with a commercial strain of 6 per cent resistance. All of the selected strains owe their origin to one of the original selections of 1925. Thus, desirable wilt-resistant selec- tions have been developed in the Heart of France type. Figure 11 shows both the 1929 status and the development through several years of the wilt-resistant selections in the red American Branching and in the Heart of France types and serves to illustrate graphically the possible development of wilt-resistant strains by annually repeated selection. In American Beauty, in which first selections were made in 1928, some progress appeared in 1929. In white, selections showed 40, 30, and 12 per cent disease resistance late in the season as compared with 2 per cent resistance in a commercial strain. The type of flower in these selections, however, was not so desirable f i*om a commercial florists' viewpoint as that of the American Branching white men- Table V. — Wilt Resistance at the End of the Flowering Season in the Red Heart of France Type in 1929 and 1930 1929 1930 Strain Wisconsin California Wisconsin California per cent per cent* per cent per cent" 2-1-1-a 79 70 90 2-1-1-b 58 70 90 -f- 2-1-1-1 92'> SO 80 2-2-1-a 62 80 90 -f- 2-2-1-b 72 70 90 2-2-5-a 66 70 90 2-2-5-b 71 70 90 -j- S-1-1 16 20 90 commercial 6. a These figures are estimates provided by Miss Bodger and Dr. Jagger. b Individual plant selection. Aster Diseases and Their Control 25 1925 f ^ 1926 T^\ 1927 K '• T^^ # 1928 1 U 'A -^^m 1929 G rri Qt9 QC A RED AMERICAK BRANCHING 1925 i ^ ^ 1926 1 i 192? M\ ^ 1928 fi ^%l^ 1929 7 :iZ ft^ <;&&# B. HEART OF f RANGE FIG. 11.— DISEASE RESISTANCE IN ANNUALLY REPEATED SELECTIONS IN RED AMERICAN BRANCHING (A) AND HEART OF FRANCE (B). IVkite Sector — Per cent healthy plants. Black Sector — Per cent plants with wilt. Hexagon — Results from use of commercial seed. The commercial stock used for the American Branching (A) in 1925 was of mixed colors. In 1937, 1928, and 1929 a commercial crimson was employed for comparison. Heart of France (B) commercial stock was bought under that nanie. Smooth Circle — Results from use of individual plant selected seed. Toothed Circle — Results from use of bulk selected seed. This shows that progress was made by re- peated annual selections in both of these varieties. tioned earlier. In shell pink one selection showed 57 per cent re- sistance. No 1928 selections of rose produced seed for the 1929 trials. In lavender no progress was made. In purple, however, a selection showed 22 per cent wilt resistance as compared with a commercial of only 2 per cent resistance. Thus in American Beauty some progress has been made in selections of certain desirable florists' colors but no selection showed a commercially profitable degree of resistance in 1929. Of other aster types the Comet-like selection from Peerless pink mentioned before gave excellent results. The three bulk and two in- 26 Wisconsin Research Bulletin 111 FIG. 12.— EARLY SEASON APPEARANCE OF WILT RESISTANT (A) AND COM- MERCIAL (B) STRAINS OF THE COMET-LIKE SELECTION FROM PEERLE.SS PINK ON ASTER-SICK SOIL AT RANDOLPH, WISCONSIN, JULY 1929. (SEE FIGURE 13.) dividual plant selections tried showed respectively 75, 62, 92, 82, and 53 per cent wilt resistance at the end of the flowering season. At the same time plants from the commercial stock of the same origin all died of the wilt. The behavior of one of these selected strains in contrast with the commercial strain in early July and in mid-August are shown in Figures 12 and 13 respectively. This is typical of the contrast between selected and comparable commercial strains in many instances (Figure 10). An individual plant selection from Royal purple also proved excellent. This showed 95 per cent resistance late in the season. Aster Diseases and Their Control 27 FIG. 13.— LATE SEASON APPEARANCE OF WILT RESISTANT (A) AND COM- MERCIAL (B) STRAINS OF THE COMET-LIKE SELECTION FROM PEERLESS PINK ON ASTER-SICK SOIL AT RANDOLPH, WISCONSIN, AUGUST 1929. (SEE FIGURE 12.) The contrast in behavior between the commercially satisfactory stand of the selected resistant strain and the complete loss of the commercial strain is marked both early and late in the season. A similar contrast appears in Figure 10. As mentioned above, certain of the wilt-resistant strains were tested in 1929 not only in Wisconsin but also in other localities. The most extensive trials were made in an aster-sick field of Bodger Seeds, Ltd., El Monte, Cal.io The results obtained there are given ^^ In planning and securing data from these and tests of commercial varieties the authors are indebted to the continued courtesy of the Bodger Seeds, Ltd., and especially to the personal attention of Miss Elizabeth Bodger. Helpful advice was received at the outset from Professors R. E. Smith, W. T. Home, and D. G. Milbrath of the California in- stitutions. During the progress of the trials Dr. I. C. Jagger, as representative of the United States Department of Agriculture, cooperated with Miss Bodger both in 1929 and 1930 in the details of planting arrangements and in the taking of final notes upon the re- sults. In addition profitable advice was received from Dr. W. J. Zaumeyer and Dr. Free- man Weiss of the United States Department of Agriculture following visits to these plots. All such aids were especially appreciated since neither of the writers visited El Monte between 1928 and 1931. 28 Wisconsin Research Bulletin 111 in Tables IV and V and agreed for the most part with those obtained in Wisconsin. Figure 14 shows the behavior of certain selections of the Heart of France type in California as compared with commercial strains. Kunkel's Semple pink proved resistant in Wisconsin and Cal- ifornia as it had in New York. Some of the Wisconsin strains were also tested by Kunkel in New York and by Weiss in Washington, D. C. According to their reports strains which were wilt-resistant in Wisconsin and California were also resistant in the above localities. This agreement in the behavior of the strains serves as evidence for the statement made earlier that aster vdlt may be due to essentially the same cause in many localities from coast to coast. If so, strains resistant in one locality would probably prove resistant in other re- gions. Results in 1930 Trials in 1930 followed in general plan those of 1929 except in the following particulars. Certain of the promising wilt-resistant strains of 1929 were not continued in Wisconsin in 1930 either because they were essentially duplicate types, or because the shade or type of blos- soms was not the most desirable from the commercial florists' point of view. Such reduction in the number of strains permitted an exten- sion of the work to include certain colors of the Royal type, a variety much in demand among the eastern florists. Trials of certain of the promising Wisconsin selections were made by florists in a larger FIG. 14.— ASTER PLANTING ON ASTER-SICK SOIL OF BODGER SEEDS, LTD. AT EL MONTE, CALIFORNIA, 1929. HEART OF FRANCE TYPE. (A.) Wisconsin selection 2-1-1-b (2 rows). (B.) Commercial (1 row). (C.) Wisconsin selection 2-2-1-a (3 rows). (D.) Commercial (2 rows). (E.) Wisconsin selection 2-2-1-b (2 rows). The commercial strains (B and D) proved highly susceptible and the selected strains (A, C, and E) relatively resistant to wilt. The approximate stands are indicated in Table V. Aster Diseases and Their Control 29 number of widely different localities in order to determine further the wilt-resistant qualities of the strains and to increase the desirable strains for commercial distribution. In the American Branching type nineteen out of twenty-two se- lections in Wisconsin gave at least a 50 per cent stand when flower- ing began and eighteen maintained this degree of resistance at the end of the blooming period (Table IV). These were of essentially the same colors as the commercially desirable promising selections of 1929, i.e., white, two shades of pink, rose, red, deep lavender, and purple. The five commercial strains tried showed 4, 30, 14, 45, and 30 per cent resistance late in the season. In the Heart of France only one selection was tried in Wisconsin in 1930 and this had an 80 per cent resistance to wilt (Table V). Since the American Beauty aster is rather late for the north- central state conditions, no extension of the work with that type was made. However, the shell-pink and purple selections of promise in 1929 were continued in 1930. Three selections of the shell pink gave 75, 77, and 82 per cent resistance at the end of the flowering season, and two purple selections 66 and 77 per cent resistance. The other aster types of promise in 1929 again proved satisfactory. Two strains of the Comet-like selection from Peerless pink showed 63 and 83 per cent resistance late in the season. At the same time three strains of Royal purple were of 85, 80, and 76 per cent resist- ance. A commercial strain of Royal purple showed only 2 per cent wilt resistance. As mentioned above, the trials of 1930 in Wisconsin were ex- tended to include other colors of the Royal type. The results with commercial stock were as follows : rose 2 per cent, lavender 1 per cent, and pink 49 per cent wilt resistance. Preliminary selections were made from the resistant plants. As in 1929 the behavior of the strains in other localities agreed with that in Wisconsin. Tables IV and V show the stands in Cali- fornia both for selections sent in 1929 and grown again in 1930, and for additional selections. The results from annually repeated selection seem, therefore, con- clusive in the major points of interest. They may be summarized as follows. In all the commercial varieties of China aster that have been tested, there is a considerable range of variation as to individual re- sistance to the wilt disease. With certain individuals at least this character is inheritable. By selecting such individuals, even from strains showing relatively low general resistance, and by annually repeating the ti'ials upon aster-sick soil and then reselecting the most promising progeny, strains may be developed of a relatively high and stable degree of resistance. More thorough trials have been made at Wisconsin in the Ameri- can Branching type, this being the one of chief commercial interest in the Wisconsin-Chicago area. As a result, strains of American 30 Wisconsin Research Bulletin HI Branching of a promising degree of resistance in the colors espe- cially sought by commercial growers (white, pink, rose, lavender and purple) have been obtained. In other types early attention was given to the Heart of France. This is a red aster in which a high degree of resistance was easily secured from a susceptible commercial strain. This was likevnse true with the corresponding red in the more de- sirable American Branching type. Less has been done with the some- what earlier Royal type, the Comet, and the later American Beauty. In all of these, so far as tested, however, progress has been made. The occurrence of wilt-resistant varieties of China asters among established commercial types The occurrence of outstanding differences in varietal susceptibil- ity to wilt has been noted from time to time, beginning with the very early reports upon the disease. ^^ Search for evidence of this has been made as opportunity permitted from the beginning of the Wis- consin studies in 1925. The unfortunate necessity of limiting ex- perimental plots in this state to areas which could be protected by cloth made advance by this means slow. As will be indicated later, developments in California of recent years have favored much more rapid and encouraging progress in this direction. Tests of commercial varieties in the experimental plots at Madi- son and Randolph on a small scale have shown few to be wilt resist- ant. These tests have been in progress from 1925 to date. Table VI shows the results in detail. In all, thirty-two commercial strains have been tried. Of these only three have approached or exceeded 50 per cent wilt resistance. The remainder have produced stands vary- ing from 0 to 30 per cent. Observations were extended to other plantings in Wisconsin. Extensive surveys were made in 1927 among available aster planta- tions growing upon presumably aster-sick soils in the search for significant varietal resistance. These included notes upon the rela- tive amount of wilt occurring in the various strains of aster in the horticultural garden at Madison (Table I). Although there were some differences as to wilt resistance between types there were even greater differences between colors of the same type. No very promising com- mercial strain was found here. This same year an unusually good opportunity for comparison of commercial varieties was afforded by the plantings of the J. W. Jung ^^ As early as 1897 Stone and Smith quoted a letter to the effect that the Giant White Comet seemed free from the disease. In 1923 in Ohio an inclination toward resistance in the Comet variety was again noted (Martin 1925). Arnold (1917, 1919) observed that there were decided differences in varietal susceptibility. Weaver (1917) recommended that the Queen of the Market variety be planted on new soil each season because of the stem rot, whereas Vick's Royal might be planted several seasons on the same soil without having stem rot. Curtiss (1926) reported from Iowa that certain varieties showed a higher degree of resistance than others. WTiile Jackson (1927) did not find special resistance in any variety at St. Catherines, he noted a report by Mr. F. L. Drayton that Heart of France .showed con- siderable resistance at Ottawa. Kunkel (1929b) found one very resistant strain of a commercial variety. Ball (1930) recorded that Royal shell pink proved highly resistant, as well as Heart of France. White (1931) suggested Semple's shell pink, Heart of France, Express, and Royal varieties as having some resistance. Aster Diseases and Their Control 31 Table VI. — Wilt Resistance at the End of the Flowering Season in the Commercial Varieties Tested in the Experimental Plots at Madison and Randolph, Wisconsin, from 1925 to 1930 Strain 1925 1927 1928 1929 1930 American Branching mixed 16 white "4 pink 3 6 'is 30 rose 14 crimson ill "s 14 lavender 22 8 62 45 purple 3 30 Heart of France 29 7 7 6 American Beauty white 3 2 pink 9 rose 1 lavender IS purple S "2 Comet pink 0 Royal pink 49 rose 2 lavender 1 purple 2 Seed Co. at Randolph. This concern had replanted to asters a con- siderable area that was infested with the aster wilt Fusarium. In this field some well marked differences in varietal susceptibility ap- peared (Figure 2). A Comet-like strain of Peerless shell pink proved especially promising (Figui-e 2A). Selections from the resistant plants were made and have been continued as discussed earlier. This was the first resistant commercial variety found in the Wis- consin studies. A greater revelation of the possibilities offered by inspection of commercial seedsmen's plantings came to the senior author in the season of 1928. Visits were then made to a number of commercial seed growing fields in California, including the very extensive aster plantations of Bodger Seeds, Ltd. at El Monte. Through the courtesy of Mr. John Bodger and other members of the firm the plantings of that year were observed and the earlier experiences of the company were noted. In 1928 the asters were being grown on "new" soil. Al- though some wilt was scattered through most of the fields it was not serious. The asters had been planted on "new" soil because in the previous summer, 1927, some varieties had suffered very badly from wilt. These varieties had been placed in an ""'old" field which had unexpectedly proven to be infested with the aster wilt Fusarium. Examination of the 1927 records showed that wilt may be as highly destructive in California as in Wisconsin. It also revealed that there may be a wide range in loss from wilt among different varieties on soils similarly aster-sick. In general less than 25 per cent of the 32 Wisconsin Research Bulletin 111 normal yield of seed was obtained. The greatest extremes, as shown by the 1927 records, occurred in the American Branching type be- tween the light flesh pink variety Mary Semple, apparently pe- culiarly susceptible as shown by a loss of over 99 per cent, and the apparently resistant lavendar variety showing a loss of less than 10 per cent. Two things seemed obvious. The first was that there al- ready existed among these aster strains wide differences in relative susceptibility, including some presumably resistant types like the lavender. The second was that, where a susceptible strain like the Semple pink had been grown on sick soil, the seed from the sur- viving plants might carry a higher degree of resistance than prev- iously. The experiences in Wisconsin substantiated these suppositions. In 1929 Bodger Seeds, Ltd. made comparative trials on their aster-sick soil of a number of their commercial strains and cer- tain selected strains sent from Wisconsin and elsewhere. These were repeated in 1930. The behavior of the selected strains from Wisconsin is shown in Tables IV and V and Figure 14. Of the fifteen commercial strains of Bodger seed planted in 1929 three showed a high degree of resistance. These results were so encouraging that further and more extensive trials were made on the same sick soil in 1930. The trials then included 148 commercial strains of Bodger seed of which 127 showed a rather uniform susceptibility whereas 21 showed promise of a commercial degree of resistance. Thus, it appears that although certain commercial strains of aster may show a high resistance to wilt, most of them are at present very susceptible to this disease. On the other hand it is evident that where it is possible to test a large series of commercial varieties in generous numbers on Fusarium-infested soil, valuable wilt-resistant strains may be discovered. It has been practicable to do this only in a small way in Wisconsin because the necessity of screening the plants against yellows has restricted the size of the trial plantations. The results on the Bodger grounds in California in the three years as reported above have already led to the discovery of a number of desirable commercial strains and indicate the value of continued and even more extensive trials of existing commercial strains. Discussion and conclusions on the control of aster wilt The studies upon the possibilities of the control of aster wilt through the use of disease-resistant strains have led to certain conclusions. It seems that with this vascular Fusarium disease as with other similar diseases studied at Wisconsin, e.g., that of cabbage yellows and pea wilt, there are differences in susceptibility between varieties or strains and more especially between individuals within the variety or strain of the host. These differences, in many cases at least, are largely inheritable. Therefore through selection followed by repeated trials upon infested soil and annual reselection, encouraging progress has been made in the Aster Diseases and Their Control 33 development of wilt-resistant strains. In general, annual selections through three to five generations have been sufficient to secure a satisfactory degree of resistance for purposes of successful floricul- ture, even upon heavily infested soil. There are numerous types of China aster in cultivation varying in season of blooming, floral type and color, and vegetative habit. So far the American Branching, Heart of France, American Beauty, Comet, and Royal types have received attention in the Wisconsin studies. These include the types most frequently used by the local commercial aster growers. All of these, however, have shown similar general susceptibility to wilt and all have yielded wilt-resistant strains by selection. Essentially like results may probably be ex- pected with the other established commercial types. There is no clear evidence of a correlation of color with resistance although it has been somewhat easier to establish resistant strains of red in the American Branching and Heart of France types. It may be noted that these red asters have high vegetative vigor. As far as color is concerned but slight evidence has been secured of cross pollination of asters in the experimental plantings in Wis- consin. Flowers of different colors have developed immediately ad- jacent with infrequent admixture of color, although this may some- times occur as Fleming (1929) has pointed out. The progeny of individual plants frequently vary in flower type and habit of growth. Consequently in selecting for resistance other desirable characters must be kept constantly in mind. Although resistance to wilt seems a fairly stable character, in no case have resistant strains shown complete immunity. That is, upon infested soil under conditions which favor disease development, such as high temperature, some wilt may be expected in even the most resistant strains. It appears that for the best results with resistant asters, as with cabbages, the mother seed at least should be produced continuously upon Fusarium-infested soil. In this way the degree of resistance may be not only maintained but presumably gradually increased. An indication of the stability of the resistant character and of the similarity of the wilt disease in widely separated places is afforded by the behavior of certain of the resistant strains developed in Wisconsin upon aster-sick soil in California, New York, Wash- ington, and other places. In all cases their resistance has been es- sentially the same in the other localities as in Wisconsin. 34 Wisconsin Research Bulletin 111 T SUMMARY wo DISEASES, aster yellows and aster wilt, seriously menace the culture of the China aster throughout the United States. 2. The symptoms of these two diseases are so similar at certain stages that aster growers often confuse them as one malady. They have, however, distinguishing characters which are herein described by which each may be recognized with certainty. 3. Both are infectious diseases, but they are introduced and spread in such different ways that they require entirely different methods of control. 4. Aster yellows is a virus disease, transmitted by the leaf hopper Cicadula sexnotata. 5. The virus overwinters in certain biennial and perennial host plants, including some common garden weeds. From these it is car- ried to the young asters in early summer by leaf hoppers. Thereafter, it is spread among the aster plants with increasing rapidity. 6. Control of aster yellows is dependent, therefore, upon shield- ing the aster plants from viruliferous leaf hoppers. 7. An effective shielding material was found in cloth not coarser than 22 X 22 threads per inch. 8. Two types of shield have been tried comparatively: a. Fences: The use of cloth-covered side walls or "fences" six feet high (tops uncovered), combined with roguing, reduced the yel- lows somewhat but was not commercially satisfactory under Wis- consin conditions. b. Houses: In Wisconsin trials for six years, aster yellows has been controlled by the use of cloth-covered cages or houses. The tops and sides of the enclosures were completely covered with cloth not coarser than 22 x 22 threads per inch. Trials using a somewhat coarser cloth for top cover showed some promise for commercial houses. 9. Aster wilt is caused by a parasitic fungus {Fusarium conglut- inans v. callistephi) . This may be carried on aster seed and, once introduced, persists indefinitely in the soil, making it aster-sick. 10. Use of clean seed and disease-free soil is, therefore, the simplest way to escape the wilt. This may be accomplished with reasonable success by painstaking home gardeners or even small com- Aster Diseases and Their Control 35 mercial growers, especially if they can use home grown or disinfected seed and plant on new soil each year. But with most commercial growers the hazards are unavoidably great. 11. The use of wilt-resistant strains of aster seems to be the most promising way to avoid incrasing difficulty from this wilt for commercial growers and for many home gardeners. 12. Wilt-resistant strains of aster have been developed at Wis- consin by annually repeated selections during the last six years. Those that are now stabilized include the several basic colors and a fairly wide range in flower type and growth habit, as represented by Ameri- can Branching, Heart of France, Comet, Royal, and American Beauty. This seems to justify the opinion that resistance can thus be developed in the standard types and colors of asters. 13. Special attention has been given to the development of wilt resistance in a series of types suitable for the commercial cut flower trade of the middle west. This has been most satisfactorily accomp- lished through securing American Branching types in the important florists' colors, viz., white, pink, rose, lavender, and purple. 14. In additon to limited trials of miscellaneous commercial strains at Wisconsin, data have been secured from larger field trials by a commercial seed firm in California which justify confidence that further large scale trials on suitable aster-sick soil will discover numerous highly vdlt-resistant strains already in existence among standard commercial asters. 15. The probability of this is increased by the known facts that at times aster wilt has occurred during recent years in highly de- structive degree in some seed growing areas, both eastern and west- ern. The '''survival of the fittest" has doubtless operated to bring about increased resistance in commercial strains even though the nature of the trouble was not clearly recognized at the time. 16. Trials of resistant Wisconsin strains on aster-sick soils in other sections extending from New York and Washington in the east to California in Ihe west indicate the stability of the resistant char- acter. It appears that there is essentially one type of pathogen and that disease expression is not so affected by environment as to over- come the resistance of well-established strains. 36 Wisconsin Research Bulletin 111 LITERATURE CITED Adams, J. F. 1928. Report of the plant pathologist 1927. Del. Quart. Bui. State Bd. Agr. 18: 3-29. Arnold, G. 1917. Growing the China aster; summary of the usual troubles. Rural New Yorker 76 : 273. 1919. Stem rot of the aster. Florist's Exchange 48 : 349. Ball, G. J. 1930. Asters under tent safe from hopper. Florists' Rev. 66 (1718) : 23-26, illus. Beach, W. S. 1918. The Fusarium wilt of China aster. Mich. Acad. Sci. Rpt. 20: 281-308, illus. Beal, a. C. 1919. China asters. N. Y. Agr. Col. (Cornell) Reading course for the farm 152: 175-212. CXJBTISS, C. F. 1926. Aster wilt studies. Iowa Agr. Exp. Sta. Ann. Rpt. (1925- 26) : 30-32. Findlay, H. 1917. The aster; a good flower to raise for the home and market. Country Gent. 82 (8) : 24-25, illus. Fleming, W. M. 1929. Inheritance of colour in asters. Sci. Agr. 10: 272-275. Galloway, B. T. 1896. Disease of China asters. Amer. Gard. 17: 518. Gloyer, W. D. 1924. Fungous diseases of the China aster. (Abstract) Phyto- pathology 14: 64. 1931a. Aster seed treatment. N. Y. Geneva Agr. Exp. Sta. Cir. 112, 6 p., illus. 1931b. China aster seed treatment and storage. N. Y. Geneva Agr. Exp. Sta. Tech. Bui. 117, 41 p., illus. Gregory, C. T. 1929. Avoiding trouble among the asters. Country Life [Garden City, N. Y.] 55 (April) : 67-68, illus. Aster Diseases and Their Control 37 Jackson, A. B. 1927. The Fusarium wilt of China asters. Sci. Agr. 7: 223-247, illus. Jones, L. R. and Riker, Regina S. 1928. Studies upon the Fusarium wilt of China aster. (Abstract) Phytopathology 18: 150. 1929. Progress with the control of aster wilt and yellows. (Ab- stract Phytopathology 19: 101. 1930. Further progress with the control of aster wilt and yel- lows. (Abstract) Phytopathology 20: 129. Kunkel, L. 0. 1924. Insect transmission of aster yellows. (Abstract) Phyto- pathology 14: 54. 1925. Insect transmission and host range of aster yellows. (Ab- stract) Science 62: 524. 1926a. Incubation period of aster yellows in its insect host. (Ab- stract) Phytopathology 16: 67. 1926b. Studies on aster yellows. Amer. Jour. Bot. 13: 646-705, illus. 1927. Sterility caused by the aster yellows disease. Mem. N. Y. Hort. Soc. 3: 243-244, illus. 1929a. Wire-screen fences for the control of aster yellows. (Ab- stract) Phytopathology 19: 100. 1929b. Wilt-resistant asters. (Abstract) Phytopathology 19: 100-101. "Marketman" 1921. Aster diseases. Amer. Florist 56: 854. Martin, G. H. 1925. Diseases of forest and shade trees, ornamental and miscel- laneous plants in the United States in 1923. U. S. Dept. Agr., Bur. Plant Indus. Plant Dis. Rpt. Sup. 37: 349-446, illus. [Mimeographed]. MULLER, R. T. 1922. China asters for the home flower garden. Mass. Agr. Col. Exp. Leaflet 54, 4 p., illus. 38 Wisconsin Research Bulletin 111 mumford, h. w. 1925. Recent progress in solving some farm problems of Illinois. 111. Agr. Exp. Sta. Ann. Rpt. (1923-24) 37, 196 p. 1926. A year's progress in solving some farm problems of Illin- ois. 111. Agr. Exp. Sta. Ann. Rpt. (1924-25) 38, 191 p. 1928. A year's progress in solving farm problems of Illinois. 111. Agr. Exp. Sta. Ann. Rpt. (1927-1928) 41, 321 p. Newton, G. A. 1929. Department of plant pathology. W. Wash. Exp. Sta. Bui. 14-W (n.s.) : 28-31, illus. PADDOCK, W. 1902. Plant diseases of 1901. Colo. Agr. Exp. Sta. Bui. 69, 23 p., illus. Rager, E. R. 1930. Treatment of aster troubles. Horticulture 8: 389-390. Rose, R. C. 1925. Suggestions on control of aster wilt. Minn. Hort. 53: 149-150. Skinner, J. H. 1930. Report of the director. Ind. Agr. Exp. Sta. Ann. Rpt. (1928-29) 42, 95 p., illus. SMITH, R. E. 1902. Growing China asters. Mass. Hatch Agr. Exp. Sta. Bui. 79, 26 p., illus. Stone, G. E. and Smith, R. E. 1897. Report of the botanists. Mass. Hatch Agr. Exp. Sta. Ann. Rpt. (1896) 9: 57-84. 1902. Report of the botanists. Mass. Hatch Agr. Exp. Sta, Ann. Rpt. (1901) 14: 57-58. 1903. Report of the botanists. Mass. Hatch Agr. Exp. Sta. Ann. Rpt. (1902) 15: 27-42. Weaver, E. J. 1917. Trying to grow the China aster; a bunch of truth about it. Rural New Yorker 76: 67-68. 1924. Brutal truth about aster growing. Rural New Yorker 83: 499. Aster Diseases and Their Control 39 Weiss, F. 1925. Diseases of the China aster. Amer. Florists 64: 269-270. 1929. Research revives hope for asters. Florists' Rev. 63 (1635) : 27-30, illus. White, R. P. 1931. Diseases of China asters. N. J. Agr. Exp. Sta. Cir. 232, 4 p., illus. William, C. G. 1930. Forty-eighth annual report for 1928-29. Ohio Agr. Exp. Sta. Bui. 446, 216 p., illus. 1931. Forty-ninth annual report for 1929-30. Ohio Agr. Exp. Sta. Bui. 470, 269 p., illus. Woods, A. F. 1899. The aster disease. Gardening [Chicago] 7: 277. Research Bulletin 116 September, 1933 Stripe Resistance and Yield of Smooth-Awned Barley Hybrids Agricultural Experiment Station of the University of Wisconsin, Madison Contents Page Introduction 1 Materials and Methods 3 Agronomic Characters and Yield 4 Kernel weight 5 Hull percentage 6 Chemical composition of kernel 6 Yield 7 Stripe Resistance Studies 8 Experimental data and results 12 Preliminary artificial inoculations 19 Discussion 20 Literature Cited 21 Stripe Resistance and Yield of Smooth- Av?ned Barley H>?brids R. G. Shands, B. D. Leith, J. G. Dickson, H. L. Shands' THE EARLY DEVELOPMENT of barley in Wisconsin was cen- tered around the introduction and improvement of the varieties Oderbrucker and "Manshury" (Manchurian types) by the Agronomy Department. Of several pure lines from these varieties, Pedigrees 5 and 6 from the Oderbrucker were selected for high yield and excellent malting quality. These barleys had two objectionable characteristics however, barbed awns and susceptibility to the stripe disease caused by Helminthosporiwm gra/mineum Rabh. In 1917, the Wisconsin Experiment Station started on a new line of attack in barley breeding, the production of a white, six-rowed, smooth-awned barley by hybridization and selection. The rough-iawned Oder- brucker, Hordeum vulgare var. pallidum typica^ Ser. which was the standard variety in the state was crossed with a small, black, smooth- awned barley, H. vidgare var. nigrmn leiorrhynchum. Kcke., primarily to combine the smooth-awned condition with the desirable characters of the Oderbrucker barley. Selection within the segregating hybrid lines had proceeded far enough by 1925 to place several of the more desirable hybrid strains in preliminary yield trials. The selection, X39-5, which outyielded Oderbrucker for three consecutive years (see Table III) was increased and distributed as Wisconsin Pedigree No. 37. Later, the selection, X39-9-3, which in two yeai's' plot trials had been superior to Pedigree 37 in yield and stripe resistance, was increased and distributed as Wisconsin Pedigree No. 38 and was named "Wisconsin Barbless". Preliminary field counts in 1926 showed certain of the hybrid selections, notably the X39 group, relatively free from the stripe disease. Other selections, such as certain of the X105 group, had more diseased plants than the Oderbrucker parent variety. The re- sults indicated, therefore, a range in reaction to the stripe disease from susceptible to practically immune selections within the progen- ies of the crosses. The question, therefore, naturally ai'ose whether these hybrid selections were more resistant to stripe or were escap- * Cooperative investigations between the Departments of Agronomy and Plant Pathology, University of Wisconsin and the Division of Cereal Crops and Diseases, Bureau of Plant Industry, U. S. Department of Agriculture. 2 Citation by number to literature cited, Harlan. (2) Wisconsin Research Bulletin 116 ing the disease. This was a question of practical sig- nificance since the stripe disease was becoming of increasing importance and heretofore there had been no report of commercial varieties or selec t i o n s showing resistance to this disease. The stripe disease had increased in the Upper Mississippi Valley spring barley section to where it had become one of the most important diseases in barley production. Field surveys showed an aver- age estimated loss of 2.1, 2.5, and 1.9 per cent' of the crop in this area for the years 1924 to 1926 in- clusive. Seed treatments for the control of the seed borne infection were un- satisfactory as well as difficult and expensive to use with a crop like barley, where most growers produced their own seed. The development of stripe resistant barley varieties was becoming increasingly important therefore, if this loss to Wisconsin barley growers was to be prevented. The logical way to proceed was to combine this stripe study with the intensive agronomic study being made upon the new smooth- awned hybrid selections. A parallel study was made with the scab disease caused by Gibberella sanbinetii (Mont.) Sacc. A number of the smooth-awned hybrid selections and Oderbrucker have been tested for i-esistance to the scab disease by inoculating under muslin cages kept at a high humidity by sprinkling with water. The percentages of scabbed kernels developing under these very favorable conditions for disease production follow: Oderbrucker, Pedgree 5-1, 74.9 per cent; Oder- brucker, Pedigree 6, 75.8 per cent; Pedigree 37, 73.5 per cent; and Wisconsin Barbless, Pedigree 38, 73.4 per cent. From the averages of a three years' cage test and from field observations where the per- centage of scab is very much less, there seems to be no significant FIG. 1.*— STRIPED PLANTS OF ODERBRUCKER PEDIGREE 6. ^ Data taken from the Plant Disease Bulletin Supplement, U.S.D.A.. 1925-1927. * Photographs and graphs were made by Eugene H. Herrling, Department of Plant Pathology. Stripe Resistance and Barley Hybrids difference in the resistance of these varieties. In regard to loose smut reaction, field observations indicate that Pedigree 37 and Pedi- gree 38 show the same susceptible tendency found in most of the other smooth-awned barleys. Materials and Methods THE HYBRIDS used in these studies were obtained by crossing pedigreed strains of Oderbrucker, Hordeum vulgare var. palli- dum typica Ser., with Leiorrhynchum, Hordeum vidgare var. nigrum leiorrhynchum Kcke, a six-rowed, black, smooth-awned variety. In field plots, Oderbrucker, Wisconsin Pedigree 6, was rather susceptible to stripe, whereas Oderbrucker, Wisconsin Pedigree 5, selected from the same original lot, was more resistant. Both of these lines were used in makng crosses with Leiorrhynchum before their reaction to stripe had been ascertained. Table I. — Parentage of Various Barley Crosses and Dates When the Hybrids Were Made Pistillate Staminate Year cross Cross parent parent was made X39» Ped. S Leiorrhynchum 1917 X57 Leiorrhynchum Ped. 6 1920 X60 Leiorrhynchum Ped. 6b 1922 X66 X39-8'^ Ped. 6 1923 X67 X39-11 Ped. 6 1923 X69 XS7-10-4 Ped. SI' 1923 X102 X60-1 Ped. 6 1924 X104 X39-9-10 Ped. 6 1924 XIOS X39-3-9 Ped. 6 1924 X106 XS7-5-2 Ped. 6 1924 » The parentage of X39 was incorrectly published in the Journal of the American Society of Agronomy 1931, 23:396-401, p. 400, as a cross between Pedigree S Oderbrucker male and Leiorrhynchum female made in 1916. '' Pedigree number not recorded, but probably Pedigree 6. •' The number or numbers following the dash after X numbers or Pedigree numbers represent individual plant selections. Head selections were made from the superior plants in the seg- regating populations from the crosses. The grain from the heads selected was sown in head rows in the nursery the following year. If the plants in the head row showed heterozygosity, reselections were made, and the most desirable heads propagated. When rela- tively homozygous lines were established, they were grown in tripli- cate rod rows the next season. When undesirable plant characters appeared in the rod row trials, reselections were made if superior individuals appeared ; if not, the strain was discarded. The test for yield and quality was made in l/20th acre plots. Many of the surviving selections were eliminated in the yield plot where competition was more nearly comparable to that in the field. Strains which were conspicuously inferior were eliminated early in the test. See Table V. The final test to determine adaptation to Wisconsin Research Bulletin 116 PED. 5-1 PED. 38 FIG. 2.— COMPARATIVE HEADS OF ODER- BRUCKER. PEDIGREE S-1, AND WISCONSIN BARBLESS, PEDIGREE 38. Awns of central spikelets have been removed to show relative size and arrangement of kernels. different soil and climatic areas of the state was made by the Branch Ex- periment Stations and also by farmers. Head selections were made to secure the Oder- brucker quality combined with the smooth awns. Sel- ection studies were made for size and color of head; size, shape and color of kernel ; smoothness of awn; length of internodes of the culm and rachis; strength of straw; date of heading and maturity; dis- ease resistance and, fin- ally, yield in both the rod rows and plots. The re- sults of the agronomic studies were reported in a previous paper (4). Some of the hybrid selections showing desirable qualities were again crossed with Oderbrucker, Pedigree 5 or 6, and reselected. Detailed studies on resistance to the stripe disease were made at several locations in the state. Rod row plantings of the various hybrid selections were made with stripe-infected Oderbrucker sown on each side of the hybrid selection to secure maximum uniform dis- tribution of conidial inoculum during the period of floral infection. The seed thus secured was sown at the different locations under conditions favorable for stripe development. Stripe counts were made in the hybrid selections and in Oderbrucker controls soon after the barley was headed when the disease was most conspicuous. Agronomic Characters and Yield THE PLANT CHARACTERS of the hybrid selections differed somewhat from the Oderbrucker type in that the best yielding hybrid selections matured four to five days later than the Pedi- greed Oderbrucker. As late maturing barleys are often injured by mid- summer heat and drouth, selection for earliness was continually prac- ticed in the barbless hybrids, but the long head and large kernel ap- peared to be associated with later maturity. However, reports over the state from the crop of 1931, a record season of heat and drouth, Stripe Resistance and Barley Hybrids indicated that the Pedigrees 37 and 38 withstood the drouth and heat fully as well as the Oderbrucker. The number and size of the barbs on the awns varied within the hybrid selections. The smoother variants, having fine barbs only near the tip of the awn, were selected for continued line prop- agation. These selections have been found somewhat more difficult to thresh than the Oderbrucker due to the smooth and somewhat flexible awn and the slightly looser attachment of the hull. Setting the concaves in the separator close enough to remove all the awns resulted in some damage to the kernels by peeling the hull. Due to genetic linkages it has been impossible to eliminate the slightly gi-ay- ish color of the hull of the kernel as well as a slightly brownish discoloration from weathering. Crossing the hybrid selections back to the Oderbrucker strains and reselecting have improved the color somewhat. Strains selected so far have not given the pure white kernel characteristic of the Oderbrucker. Kernel Weight The consistent increased yield of the Wisconsin Barbless, Pedi- gree 38, over the Oderbrucker suggested a comparative study of the weight of kernel and hull in relation to this increased grain production. Grain from the yield plots at Madison, Wisconsin, was used for comparative study over a three-year period which included the hot, dry season of 1931 relatively unfavorable for barley pro- duction. The kernel weight of the two barley varieties was not significantly different under favorable conditions for production of the crop. In general, Oderbrucker, Pedigree 5-1, developed a slightly larger kernel than the Wisconsin Barbless, the latter being 1.45 per cent lighter in the two favorable seasons than the standard Oderbrucker. See Table II. Table II. — Kernel Weights of Wisconsin Barbless, Pedigree 38, AND Oderbrucker, Pedigree 5-1, Grown at Madison, Wisconsin, Seasons of 1930 to 1932. Season and place grown Average weight of 100 kernels in grams Difference in kernel weight — Ped. 38 com- pared with Ped. 5-1 Oderbrucker Ped. 5-1 Wisconsin Barbless, Ped. 38 Grams % 1930 Yield plots 3.3651 2.6539 3.5728 3.28i9 3.2520 3.5531 0S32 ' 7 4' 1931 Yield Plots 1932 Yield plots -I- .5951 -.0197 + 22.5 - 0.5 This difference in kernel weight was probably due in part to the increased number of kernels per head in Pedigree 38. In the unfav- Wisconsin Reskarch Bulletin 116 orable season of 1931, however, a very significant difference in kernel weight was found in favor of the Wisconsin Barbless, Pedigree 38, which was 22.5 per cent heavier than the Oderbrucker, Pedigree 5-1. While yield was lowered in both varieties, the kernel weight of Pedigree 38 was reduced 4.8 per cent based on the average of the two favorable seasons in contrast with 23.5 per cent in Pedigree 5-1. In other words, the size and weight of the kernel in Pedigree 38 were maintained even under unfavorable conditions. Since grade and price of malting barley are closely correlated with kernel uni- formity and size, this is an important factor in marketing barley for commercial uses. During the past four years. Pedigree 38 has shown this advantage in uniformity of kernel, size and weight, in both plots and fields. Hull Percentage The percentage of hull* based upon kernel weight was slightly higher in Pedigree 38 than in the Oderbrucker. In the two favorable years, the hull of Pedigree 38 averaged 1.52 per cent higher than that of Pedigree 5-1. See Table III. In 1931, because of the larger sized berry of Pedigree 38, the two varieties contained approximately the same percentage of hull. The hull of the Pedigree 38 was not as firmly attached (especially toward the tip of the kernel) as in Oder- brucker. This condition resulted in some peeling in the rough treat- ment of dry grain and in excessive mixing and handling of the steeped moist grain. Considerable variation in the adhesion of the hull to the kernel occurred in the hybrid selections; therefore, it should be possible to select a strain with the hull as firmly attached as in the standard Oderbrucker. Table III. — Average Percentage of Hull by Weight on One Hundred Ki:rnels Samples of Oderbrucker (Ped. 5-1) and THE Wisconsin Barbless (Ped. 38) Barley. Grown at Madison, Wisconsin, Seasons of 1930 to 1932. Season and place grown Average percentage of hulls on kernels of Difference. Ped. 38 pared with Ped. 5- in per cent com- Ped. S-1 Ped. 38 1930 Yield plot 1931 Yield plot 1932 Yield plot Average difference in 3 seasons 9.66 11.87 S.99 10.88 11.8? 10.^ _ 6 45.5 46.7 44.3 45.2 38.7 59.1 36.3 39.3 2 > 0- 0^ CO 47.7 44.1 37.3 37.0 33.5 47.4 30.1 29.4 00 q r~ -, •* •■a- 1 CO d 46.4 67.2 53.1 52.4 49.2 42.9 0 * t-- 00 rt 00 r^ r^ ro CO •0 10 rooi^;ON:;::::::::::::: — dr^ioit^;:::::::::::;::: «3- vomio • rr 1 c> 58.5 69.0 65.4 60.4 58.0 52.5 49.9 48.1 .0 2 V 0 c X cs c C X C X C X X or 1 c 6 X C X ■o -c XX X X ly- c X fs C X 00 *^ CO CO *^ 0- o>_ CO co-C XXI (x X I 1 0- X Stripe Resistance and Barley Hybrids Table VI. — Yields of Pedigree 37, Pedigree 38, and Oderbrucker, Pedigree 5-1, at the Wisconsin Branch Experiment Stations Bushels per acre Station Oderbrucker Ped. 5-1 Pedigree 37 Pedigree 38 1930 Ashland Marshfield Sturgeon Bay 1931 21.3 Sl.O 14.2 29.2 38.5 25.8 34.8 28'.'3 30.8 38.0 80.8 46.0 18.7 Marshfield 1932 Ashland 34.2 40.0 33.3 • The yields of the two selections, Pedigree 37 and 38, in farm trials have been considerably higher than the Oderbrucker. In 1930 sev- eral farmers reported yields above 60 bushels per acre from the Pedigree 38. The heat and drouth in 1931 reduced barley yields throughout the state; the yields of Pedigree 38 reported, however, were without exception higher than those obtained with Oderbrucker. In three widely separated counties where yields of Pedigree 38 and Oderbrucker were compared on farms in the same locality, 28 farmers reported an average yield of 31.7 bushels per acre for Pedigree 38; 22 growers, an average yield of 26.7 bushels per acre for the Oder- brucker. Again in 1932, Pedigree 38 out-yielded the standard varieties. County agents and farmers in the better barley sections of the state reported around 60 bushels per acre with five reporting 70 bushels or higher. Since the smooth-awned, hybrid selections have given high yields of good quality grain in both test plots and field trials, Wisconsin Barbless, Pedigree 38, has been accepted as the standard variety for the state. Stripe Resistance Studies DATA ON STRIPE infection, taken in 1927, Table VII, suggested differences in resistance to stripe. In order to study the disease reaction of these crosses under different environmental condi- tions, plots were sown in 1928 in South Central Wisconsin at Janes- ville and Madison and in the lake shore region near Milwaukee, Cleveland, and Sturgeon Bay. These locations were selected be- cause stripe is a very important factor in barley production along the shore of Lake Michigan as well as in the vicinity of Janesville, and it was thought that the maximum amount of infection could be obtained at these places. (Fig. 3). Seed grain for the 1928 sow- 10 Wisconsin Research Bulletin 116 ings came from the plot supplying the data in Table VII. The opportunity for infection, therefore, should have been rel- a t i V e 1 y uniform among the various selections. In order further to expose the hybrids to a uniform and maximum amount of infection, a row of stripe-in- fected Pedigree 6 barley was sown al- ternately with each row of hybrids, so that there was a row of stripe-infected Pedigree 6 on either side of the hybrid selection. In all locations the same quantity of seed was used in so^^^ng each rod row, which made counts of striped plants comparable. By using the same uniformly infected seed at the different locations, it seemed possible that any large differences from place to place in any given hybrid selection or in the check, Pedigree 6, could be attributed to en- vironmental conditions. FIG. 3. -MAP OF WISCONSIN SHOWING LOCATION OF STRIPE PLOTS IN 1928 AND 1929. Table VII. — Number of Striped Barley Plants in Three Rod Rows OF Barley Selections Grown at Madison, Wis., 1927 Number of striped plants in — Selection Series 1 Series 2 Series 3 Total X39-9-3-4 0 0 3 0 0 0 0 3 1 0 2 b 3 4.2" 0 0 0 1 0 2 1 I 0 1 3 0 0 2 0 0 0 0 0 1 0 2 3 2 2 0 XS7-12-4-13 0 XS7-24-6-9-3 3 XS7-24-6-10-1 1 XS7-24-6-10-S X57-24-10-2-3 0 3 XS7-24-6-6-4 1 X57-24-6-6-2 6 X60-2-1 4 X60-2-3 3 7 0 0 5 Pedigree S, (controls) .... " Average per row for 41 rod rows distributed every fifth row throughout the plot. Stripe Resistance and Barley Hybrids 11 Table VIII.'— Number of Stripe-Infected Barley Plants Per Rod Row in Hybrid Selections and Pedigree 6 Grown at Various Locations in 1928 from Seed Grown at Madison, Wis. in 1927 Hybrid sel- Mad- Mad- Mad- ection and Mil- ison ison ison Stur- pedif;ree Janes- wau- Series Series Series Cleve- geon nos. ville kee 1 2 3 land Bay Total Hybrid sel- ections 1 X39-5-8-4-1 0 0 0 0 0 0 0 0 X39-5-8-4-2 0 0 0 0 0 0 u 0 X39-5-8-4-3 0 0 0 0 0 1 0 0 0 X39-9-3 0 0 — — — 1 0 0 0 X39-9-3-4 0 0 _ — — 0 0 0 X39-9-3-6-1 0 0 0 0 0 0 0 0 X39-9-3-6-2 0 0 0 0 0 I 0 1 X39-9-3-6-4 0 0 0 0 0 0 0 0 X39-9-3-6-5 0 0 0 0 0 0 ! 0 0 X39-9-3-6-7 0 0 0 0 0 0 0 0 X39-9-3-6-8 0 0 0 0 0 0 1 ° 0 X39-9-10V-1 0 0 0 0 0 0 ! 0 0 X39-9-10V-5-3 0 0 0 0 0 0 0 0 X39-9-10V-S-6 0 0 0 0 0 0 0 0 X57-5-3 0 0 — — — 0 2 2 XS7-12-4-13 0 0 — — . — 0 1 1 XS7-12-S-2-1 0 0 0 0 0 0 0 0 X5 7-24-6-6-4 0 0 0 2 0 0 2 4 XS7-24-6-9-3 1 0 2 6 0 2 0 5 X57-24-6-10-5 0 0 1 0 0 1 2 4 X57-27-5-5 0 0 — — — 0 3 3 X60-2-1 0 0 — — — 0 1 1 X60-2-3 0 0 — — — 0 0 0 X66-1-3 0 0 0 0 0 0 0 0 X66-1-4 0 0 0 0 0 0 0 0 X66-1-7 0 0 0 0 0 0 0 0 X67-1-1 0 0 0 0 0 0 0 0 X69-1-2 2 0 0 0 0 0 0 2 X69-1-4 1 0 0 0 0 0 1 3 X69-2-2 0 0 0 0 1 0 0 1 X69-2-3 0 2 0 0 0 0 0 2 X69-2-6 2 0 0 0 0 0 0 2 X69-2-7 0 0 0 0 0 0 0 0 X69-2-9 1 0 0 0 1 0 0 2 X69-2-10 0 0 0 0 0 0 0 0 X102-1-5-4 1 ' 0 0 0 0 0 0 1 X102-1-6-1 3 0 0 0 0 0 1 4 X102-1-6-2 S 1 1 0 2 2 2 13 X102-1-6-3 0 0 0 0 0 0 0 0 X102-1-S-1 1 1 0 0 0 2 1 S X102-2-1 0 0 0 0 1 0 0 1 X102-2-2 0 0 0 0 0 0 0 0 X102-2-3 1 0 0 0 1 1 0 3 X102-2-3-4 0 0 0 0 0 0 0 0 X102-2-4-1 0 0 0 0 0 0 0 0 X102-2-4-2 1 0 0 0 0 0 1 2 X102-2-4-3 0 1 0 0 0 0 0 0 a X104-1-1-1 0 1 0 0 1 0 0 2 X104-1-1-2 0 4 0 0 2 0 1 7 X104-1-1-3 1 0 0 0 1 0 0 2 X104-1-1-4 0 0 1 0 0 1 1 3 X104-1-2-1 0 0 0 0 0 0 0 0 X104-1-2-2 1 0 0 0 0 0 1 2 X104-1-2-3 0 0 0 0 0 1 1 2 X104-1-2-4 1 0 0 0 0 0 i 0 1 X104-1-2-S 0 0 0 1 0 0 0 1 X104-1-2-6 0 1 0 0 0 0 0 1 1 X104-1-3 2 [ 0 0 0 0 1 0 3 XlOS-1-2-3 0 1 0 i 0 0 1 3 XlOS-1-4-1 0 0 0 0 0 0 0 0 X105-1-4-2 1 0 0 0 0 0 0 1 12 Wisconsin Research Bulletin 116 Table VIII. — Continued Hybrid sel- ' Mad- Mad- , Mad- L-lectiun and Mil- ir.on ison ison Stur- pedigree Janes- 1 wau- Series Series Series Cleve- geon nos. viile Icee 1 2 3 land Bay Total Hybrid sel- ections XlOS-l-S-1 0 1 0 0 0 0 0 1 XlOS-l-S-2 0 0 0 0 0 0 0 0 X105-l-5-,5 1 0 0 0 0 0 0 1 X105-2-1-1 1 0 0 0 1 0 1 3 X105-2-1-2 0 1 1 0 1 2 1 6 X105-2-1-J 1 1 4 1 1 1 0 9 X105-2-2 2 0 2 0 4 4 2 14 X105-2-2 1 1 0 1 3 0 0 0 5 X105-2-2-2 3 ' 1 0 0 0 2 1 7 X105-2-4-1 5 1 2 — ■ — — 2 1 10 X105-2-4-2 3 2 — — . — 6 4 IS S:i05-2 6-1 2 2 1 0 0 2 1 8 X105-2-6,5 4 3 0 1 1 0 1 10 XlOS-2-6-4 0 3 0 0 0 1 1 5 X105-2-7 2 1 J 2 0 2 1 11 X105-2-7-2 I 0 0 0 0 0 0 1 X105-2-7-3 0 0 0 0 0 0 0 0 X106-1 0 0 0 0 0 0 1 1 X106-2-1 1 1 0 1 0 0 0 3 X106-2-2 2 0 2 1 0 0 0 5 X106-2-.i 2 0 1 0 0 1 1 5 XlOo-2-t 1 0 0 0 1 0 1 3 X106-2-5 2 0 1 0 0 1 0 4 XlOu 2-6 4 0 0 0 0 0 0 4 X106-2-7 1 0 0 0 0 0 0 1 Pedigree 6 1 Av. per rod row 16.S ; 8,6 15.4 I6.,S 16.0 26.9 28.9 — Seed from the selections, grown at the five stations in 1928, was sown in a uniform plot at Madison in 1929 in order to test the possible effect of the several environments on infection. In this ex- periment, also, stripe-infected Pedigi'ee 6 was included for com- parison and for supplying inoculum. Experimental Data and Results At each location in the 1928 tests, there were approximately 250 plants per rod row. The stripe-infected plants were counted and the data are given in Table VIII and summarized in Table IX. Because of insufficient seed, two lots of stripe-infected Pedigree 6 were used, one grown at Janesville in 1925, and the other grown at Janesville in 1926. The first lot was used at Janesville, Milwaukee, and Madison, and the second at Cleveland and Sturgeon Bay. Ap- parently, the 1926 Pedigree 6 seed was the more heavily infected, judging by the relatively greater number of striped plants develop- ing in the controls at Cleveland and Sturgeon Bay where this seed v/as used. The hybrid selections, from the presumably uniformly inoculated 1927 Madison crop, showed some differences in stripe infection at the different locations. This suggested the importance of environ- ment on the development of the disease. Stripe Resistance and Bakley Hybrids 13 Table IX.— Summary of the Number of Striped Plants of Ped- igree 6 AND Hybrid Selections Grown in Rod Rows in 1928 AT Different Locations in Wisconsin. Location of plots Row Rod s of— Striped plants in — Average striped per number plants row Ped. 6 Hybrid selec- tions Ped. 6 Hybrid selec- tions Ped. 6 Hybrid selec- tions Number Nunibin Number Number Janesville 88» 88" 86 1481 760 64 28 17.61 36 40 16.83 8.64 16.35 26.92 28.92 .74 .3i Madifon (Av of 3 series') Cleveland 78" 77 1 1276 88'' 86 1 2369 .23 .42 Sturgeon Bay 88'' 86 2545 .47 " Seed grown at Janesville in 1925. *• Seed grown at Janesville in 1926. Only one striped plant occurred in the 14 selections from the X39 cross. Pedigree 5 was the pistillate parent and Leiorrhynchum the staminate in this cross. Leiorrhynchum was the pistillate parent and Pedi^ee 6 the staminate of hybrid X57. Of the seven X57 sel- ections all except one showed some stripe infection. Selections of X66 with Pedigree 6 as the staminate parent failed to develop stripe in 1928, but showed susceptibility in 1929. This indicates the greater genetic resistance of Pedigi'ee 5 over Pedigree 6. One selection from X60 showed stripe infection in 1928 while the other did not. Cross X69 was a hybrid between a selection of X57 (Leiorrhynchum X Pedigree 6) and Pedigree 5. Only two of the eight selections from X69 were free from stripe in 1928. Hybrid X102, a cross between X60-1 and Pedigree 6, showed eight selections with stripe and four with no stripe in 1928. Hybrids XI 04 and X105 both had Pedigree 5 and Pedigree 6 in their parentage and were similar to hybrids X66 and X67. Only one selection of X104 was free from stripe, while three in cross X105 were free. Most of the selections from X105 were very susceptible. In X106, a cross between X57-5-2 and Pedi- gree 6, Pedigree 6 entered into the recombination twice and all eight selections tested were infected with stripe. These 1928 data indicate rather clearly that stripe resistance and susceptibility are inherited, although the nature of the inheritance is not shown. Genau^ observed that, after head inoculation, varietal differences in stripe infection ranged from 0 to 60 per cent. Environment influenced stripe development in the 1928 experi- ments and perhaps modified the expression of resistance to stripe. The effect of environment during the seedling stage was evident by the difference in the number of striped plants occurring at differ- ent locations where the same seed lot had been used (Table IX). About twice as much stripe occurred in both hybrids and controls at Janesville as at Milwaukee; whereas, in comparing Janesville and 14 Wisconsin Research Bulletin 116 t_*_. t^OOOOrcr^'^''C(^ot^Ooor**coOrr;OfO'^i^po GC «^ r': sC <^ -^ rO ( _ fO t^ r^ O ro <^ I — O O t^ O ( lOrsiOO'^-^-^^O'^OOO- lO-^O — i'-. OO-H^^i 1^ . — ^00— (^OC a xj r^ I ^ o O >o O <■ ■ O O (^ O O t iTj-O-nrsioc— — r-Oi ) O 00 c^ — r^j o^ ' :00\0-^0''':fr:Or->-^.— ■Or^iioOO'^i'-^iOOa-OrvjTrj-— .r^O. IsO-^r-j,— lOOO^ror^*— 'fvjCvjT^CvjOr-^HvOrt'O'^O^OvO^ I O O vC ^ ^ ' Or'^-^OCrsj— .r:J-0'~00«-«r-l lOfjr^O^'^O — "r^ooro<^rrrOfN>— 'OOOt-^Or^-^"— '<;<^'— <00Oi CNOOfvioOr^OOocOONOTf< i(nOOf^r^rsiTf^.-MW^— ii Tj-OO^-r-i'-^vC^-OO— 'OoCCCCCr-isO-— 1^ — ' I O- O " rvi O vO 1.-, ^ i:> * OO — 0-*^>00^ — OrvI^^r^T^^o--00< OO'—^CoO'O— . — r^ — <^i — Ti- — O'o^i/^OOO' — f^oC"' r^ 00 *0 0n Oo Ovo\ o^o ; r^ f^ ^C Tj- . VO a-XXci.XXP.>^.XciHXX&.XX£^>^>^f^XXa.XXa.XX&.X Stripe Resistance and Barley Hybrids 15 >< Oi~- — H 3 ^ « c 2 lOO.— lOvr^Ovr^O^ro-HrOPO* • ■^^J-tvjQcr^-^vOoo^'^O «J M 2 ^ n3 H c o £ £ lOCoooOCNi^O^r^O" I O -^ lo f^ I o >o >o « I _ lO O !> — ^ ' ) sc sc — ■ — in -H I 2-^ oo c:x Ifli X O. X X Oi (in &, fc Oh rj y^ o CT3 aj t? ^ O "■ M nj o _ ■r' o ^' ^- 60 'S^ -: k- ^"S 16 Wisconsin Research Bulletin 116 Madison the stripe infection in Pedigree 6 was practically equal. In the hybrid selections, however, there was approximately three times as much stripe at Janesville as at Madison. There was no apparent reason for the difference in the ratio of striped plants in the hybrid selections and in Pedigree 6 at the two locations. Less stripe also occurred in the hybrid selections at Cleveland and Sturgeon Bay than at Janesville. This might suggest differential environ- mental reaction in the various strains of barley. The effect of high soil moisture on stripe development was ob- served at Janesville. Twenty rows of Pedigree 6 at one end of the plot were on lower ground, considerably higher in moisture content than the rest of the plot. In these twenty rows there was an average of 11.6 striped plants per row; whereas, the average for all rows of Pedigree 6 in the plot was 16.83. The other plots did not show differ- ences in the percentages of stripe that could be attributed to differ- ences in soil conditions. These results agreed with those reported by Leukel et aV, summarized as follows: "Relatively dry soil (less 2= X39-9-3-6-I 3= X39-9-3-6-4 ■ X39-9H0V-5-6 ■ PEO. 6 _ CONTROL ■ A 23456789 56789 123456789 I" Ot30 "(b) X69 SEIXCTIONS - I = X69-I-2 _ 2= X 69-2-2 3= X 69-2-3 4 = X69-2-6 5= X69-2-7 6= X 69-2-9 — 7=X69-2-iO 8= PED. 6 CONTROL K.O ■- I ■- - r^V "(c)xi04 SELECTIONS - I = XI04-I-2-3 — 2= XI04-I-2-4 3= XI04-1-2-5 4 = PED. 6 CONTROL 110 i 5 U. 11 ^?T- :®« >= f-cD. 6 CONTROL ml. hllll Mill Lllll Llll JlIlL MADIS.ON JANESVILLE MILWAUKEE CLEVELAND STURGEON BAY FIG. 4.— GRAPHS SHOWING THE RANGE OF STRIPE INFECTION IN THE VARI- OUS SELECTIONS WITHIN THE HYBRIDS X39, X69, X104, AND XIOS, GROWN AT MADISON. 'WTSCONSIN, IN 1929. THE SEED WAS PRODUCED UNDER UNIFORM CONDITIONS AT EACH OF THE FIVE STATIONS INDICATED IN 1928. Stripe Resistance and Barley Hybrids 17 than 20 per cent saturation) during the period of emergence seemed to favor stripe development as compared with very wet soil." Seed of the Pedigree 6 grown at each of the five locations in 1928 was bulked and sown in 1929 for comparison with 28 hybrid selections. Seed of stripe-infected Pedigree 6 grown at Madison in 1928 was sown every third row as a control. The seed grown at each of these different locations was sown in two series (S-1 and S-2) at Madison to study the effects of the different environments in 1928 upon the floral infection with the stripe fungus. In addition to this, a similar series of Pedigree 6 and hybrid selections was sown at Cleveland and Marshfield in 1929 with seed grown at Madison in 1928. The arrangements of rows and the results are given in Table X. The X39 selections again showed a low number of striped plants per row, X39-5-8-4-1 showing no striped plants in any of the 1929 plantings. (Fig. 4A). The other selections of X39 had a total of 6 to 13 plants for the 12 rod rows or approximately 0.0 to 0.43 per cent. The X57 selection was more susceptible having 1.88 per cent stripe. The selection of hybrid X66 from Maidson and Janesville seed was heavily infected but was somewhat less so from other locations. And again, the selections of hybrid X69 were uniformly low in stripe infection from all locations. (Fig. 4B). One of the selections from hybrid X102 was higher than the other selection in stripe infection at all of the locations. One of the selections from the X104 cross was uniformly low in infection at all stations and one was intermediate. On the other hand, selection X104-1-2-3 was higher at all stations with a marked increase in infection in the seed produced at Cleveland. (Fig, 4C). Infection in hybrid X105 was higher than in any other hybrid and the selections showed some variation in infection from the different seed sources. The reactions of the X105 selections grown at the five locations are shown graphically in figure 4D. Infection in several of the selections parallels that in Pedigree 6 while others fail to show any such cor- relation. This is illustrated by comparing X105-2-2-2, practically as susceptible as the control, with X105-2-1-3, fairly resistant. The two selections were infected about equally when grown at Madison under the conditions where they had been selected; but, when grown at Janesville and Cleveland, differences occurred that were large enough to appear significant. The infection in the XI 05 selections at Stur- geon Bay was practically the same as that at Madison. A compar- ison of infection in the various strains grown in the different sec- tions of the state suggests, insofar as resistance to stripe is concerned, a less stable condition of some strains under different environmental complexes. The effect of environment during germination and seedling de- velopment was emphasized again in comparing the amount of stripe obtained from uniformly infected seed grown at Madison and sown at Madison, Cleveland, and Marshfield in 1929. The Pedigree 6 grown at Cleveland developed less than one-third the stripe that oc- 18 Wisconsin Research Bulletin 116 curred at Madison. The infection developing at Marshfield like that at Cleveland was also very low. In contrast with this, a number of the hybrids increased in the amount of stripe under Cleveland condi- tions and two of them actually had more than the checks. (Fig. 5). Under Marshfield conditions, an unusually low infection developed during the spring of 1929 which may be partly explained by the warm weather prevailing at the late date of planting (May 14). 25