Hi iK!»5«tt!lI,'»:i,:I>,J,;t,fJ*!',.li>'t;,^:,> I S u£t*FlJI SfinRE*! M $m m Bacteria in Relation to Plant Diseases BY ERW1N F. SMITH In charge of Laboratory oj Plant Pathology, Bureau of Plant Industry, U. S. Department of Agriculture VOLUME TWO HISTORY, GENERAL CONSIDERATIONS, VASCULAR DISEASES WASHINGTON, D. C. Published by the Carnegie Institution of Washington 191 1. CARNEGIE INSTITUTION OF WASHINGTON Publication No. 27, Vol. two. ^11 Cop were firs or PRESS ov GIBSON BROTHERS WASHINGTON, D. C. CONTENTS. Page. Introduction 3 History 7 The Earliest Workers 7 The General Attitude of Pathologists and Bac- teriologists 9 Literature 21 General Considerations 23 On the supposed normal occurrence of Bacteria in Plants 23 Literature 27 Bacteria on the Surface of Plants 28 Obstacles to their Entrance into Plants — Obstacles to their Multiplication in Plants 2S The Epiphytic species 30 Literature 35 The Entrance of Bacteria into Plants. The Ques- tion of Parasitism — What Constitutes a Parasite? 36 The Carriers of Infection. . . 40 Specific Diseases 40 The Experimental Production of Parasites. . 41 Literature 50 Inception and Progress of the Disease 51 Manner of Infection 51 Wound Infections 51 Infection through Natural Openings 54 Nectarial Infections 54 Water-pore Infections 56 Stomatal Infections 57 Lenticellate Infections 63 Extrafloral Nectaries — The Stigma .... 64 Period of Incubation 64 Duration of Disease 66 Final Outcome 67 Tissues Attacked 69 Mass-action of Bacteria 70 Secondary Tumors and Metastasis . . 71 Literature 75 Solvent Action of Bacteria — Destruction of Middle Lamellae — Bacterial Solution of Cell-walls — Fermentation of Cellu- lose— Destruction of Wood 76 Literature S9 Reaction of the Plant 90 Hyperplasias 90 Hypertrophies 91 Atrophies 92 Enlargement of the Nucleus 92 Changes in the Chromosomes 93 Page. Inception and Progress of the Disease — Cont'd. Reaction of the Plant — Continued. Antibodies 93 Literature 94 Individual and Varietal Resistance — What Constitutes Immunity? Immune Varieties — Intra- varietal Selection — Cross-breeding for Resistance — Is it Possible by Special Foods to obtain Resistant Plants? 95 Symbiosis 96 Root-nodules of Leguminosae 97 Synonymy of Bad. leguminosarum 99 Summary of Leading Papers 100 Literature 138 Bacterial Symbiosis in other Green Plants. . 147 Insectivorous Plants 147 Bacteria in Hop Glands 153 Bacteria with Algae 153 Literature 154 Bacterial Symbiosis in Cryptogams 155 Bacteria with Yeasts 155 Kefir 155 The Ginger-beer Plant 162 The So-called " Beer Seed " 166 Bacteria with Fungi 168 Bacteria with Myxomycetes 169 One Bacterium with Another 172 Literature 173 Are any Bacteria known to cause Disease in both Plants and Animals — Evidence from Inoculating Plant Parasites into Animals — Evidence from Inoculating Animal Parasites into Plants — Do Plants Harbor Animal Parasites? . ... 174 Animal Parasites Inoculated into Plants 174 Plant Parasites Inoculated into Animals .... 181 Inoculations of Bacterium tnmefaciens into Fish and Frogs . 182 Do Plants Harbor Animal Parasites 184 Literature 187 Hygiene of Plants 188 Recovery by Excision 191 Germicides 192 Germicidal Treatment of Seeds 196 Germicidal Treatment of Dormant Plants 201 Germicidal Treatment of Growing Plants. 201 Insecticides 202 Formulae 204 Literature 206 m IV CONTENTS. VASCULAR DISEASES. Page. Wilt of Cucurbits 209 Definition 209 Host-plants 209 Geographical Distribution 209 Signs of the Disease 211 Etiology 212 Field, Hot-house and Laboratory Notes. ... 217 Effect of Water on Wilt 2S4 Varieties Attacked 284 Morbid Anatomy 285 The Parasite — Bacillus tracheiphilus 286 Resume of Salient Characters 2t;4 Treatment 295 Pecuniary Losses 297 History 298 Literature 299 Black Rot of Cruciferous Plants 300 Definition 300 Host-plants 300 Geographical Distribution 300 Signs of the Disease 301 Etiology 304 Morbid Anatomy 314 The Parasite — Bacterium cumpestre 316 Resume of Salient Characters 327 Page. Black Rot of Cruciferous Plants — Continued. Treatment 328 Pecuniary Losses 330 History 331 Literature 333 Yellow Disease of Hyacinths 335 Definition 335 Host-plants 335 Geographical Distribution 335 Signs of the Disease 335 Etiology 336 Varietal Resistance 337 Table Showing Sensitive and Resistant Varie- ties 339 List of Hyacinth Varieties which have entirely or nearly Disappeared in Consequence of the Yellow Disease. . 34 1 Morbid Anatomy 342 The Parasite — Bacterium l:\acinthi 344 Resume of Salient Characters 350 Treatment 35 1 Pecuniary Losses 352 History 352 Literature 353 ERRATA. Page 8 line 20 read only. ' 33 " 1 " Metcalf. " 56 " 4 " 1897- ' Plate 3 last line read See plate 4. LIST OF ILLUSTRATIONS. PLATES. Faces page Plate i. Cucumber Leaves attacked by B. tracheiphilus (colored); also Tubes of Litmus Milk i 2. Water-pore Infections on Cabbage Leaf (plant No. 402) at end of two months 58 3. Infections through Stomata in Black Spot of Plum 58 4. Early Stage of Infection in Black Spot of Plum, showing Bacteria Escaping through Central Rift 60 5. Coconut Palms in middle and late stages of the Budrot 70 5a. Cross-section of a daisy stem between tumors, showing the tumor strand. 72 5b. Cross-section of a secondary tumor in a daisy petiole, showing stem structure 72 6. Olive Shoots showing Tubercle Me- tastasis 72 7. Paris Daisy with leaf showing Sec- ondary Tumor 72 8. Result from Inoculating Bad. tumc- faciens into Shoots of Oleander, Daisy, and Olive go 9. Result from Inoculating Bad. savas- lanoi into Shoots of Oleander, Daisy, and Olive 92 10. Atrophy and Death of Shoot due to Bad. tumefaciens 02 Faces page Plate i i . Four Varieties of Potatoes Inoculated with Bacillus phytophthorus, show- ing variable Resistance and differ- ence in Color of Decayed Tissue. - 96 12. Serradella grown with and without Root-nodules on Nitrogen-free Soil. 102 13. Stages in Bacterial wilt of Cucumbers, Anacostia, D. C, 1893 210 14. Six Cucumber Plants destroyed within three weeks by B. tracheiphilus as a result of Infected Needle-pricks .... 212 15. No. 1, Effect of Inoculating Squash Bacillus into Cucumber; No. 2, Transmission of Disease by Diabrotica 214 16. Two stages of Bacterial Wilt of Cucumber due to Inoculations by Needle-pricks; No. 1, Inoculated from Cucumber ; No. 2 , from Squash 222 17. Brown Stain in Vascular Bundle of Cabbage attacked by Bad.campestre. Also tube of Bad. stewarti and of Bad. phaseoli (on potato) (colored) . 300 18. The Ragged Leaf of Cabbage 304 19. Wakker's Yellow Disease of Hyacinths — Inoculated Leaves and Scape, and Infected Bulbs (colored) 336 20. Yellow Disease of Hyacinths — Inocu- lated Scape, badly Diseased Bulbs, and Cultures in Litmus Milk. 536 TEXT FIGURES. Page. Fig. 1. Stem of Potato No. 13 (1895,1 at- tacked by Bad solanacearum. 5 2. Peas growing under Sterile conditions on Gelatin Media 23 3. Ciri of Bad. niori oozing from Len- ticels on young shoots of Mulberry . 2S 4. The Gum-bud Disease of Carna- tions 34 5. Gelatin Impression of Helianthus Leaf. 35 6. Cross-section of Vascular Bundles of Cucumber attacked by Wilt show- ing location of Bacterial Infection . 52 7. Bacterially Infected (wilted) Leaf of Cucumber gnawed by striped Cu- cumber Beetles 53 8. Effect of Light Frost on Leaves of Magnolia 54 Page. Fig. 9. Water-pore Infections of Cabbage Leaves 28 days after Spraying 55 10. Cabbage Plant No. 400, attacked on margins of Older Leaves, i. <\, through Water-pores, as a result of Pure-culture Inoculations by Spraying 56 11. Section parallel to surface of Leaf through Water-pore region of Cab- bage attacked by Bacterium cam- pestre 57 12. Earliest stage of Infection in Black Spot of Plum 58 13. Natural versus Artificial Infection in Black Spot of Plum 60 14. Inner Husk of Sweet-corn attacked by Bacterium stewarti 61 v VI LIST OF ILLUSTRATIONS. Page. Fig. 15. Cross-section of Beau-leaf in an early- stage of Infection showing relation of Bacterium phaseoli to the Tissues. 62 16. Cross-section of Leaf of Broom-corn showing an early stage of Stomatal Infection 63 17. Like Fig. 16, but the Bacteria have passed beyond the Substomatic Chamber and are wedging apart neighboring cells 64 18. Angular Leaf-spot on Rivers Cotton produced in hot-house by spraying with water containing Bacterium malvacearum 65 19. Coconut Budrot: Sound enveloping Leaves removed to show Terminal Bud attacked by the Soft Rot 67 20. Coconut Budrot, showing inability of rotted bud to support its own weight, when the outer leaf-sheaths are removed 68 2 1 . Root-hairs of Bur-clover infected by Bad. leguminosarum 71 22. Bacterial Strands in root-tubercle of Bur-clover 71 23. Olive Leaves with incipient Metastatic Tubercles due to Stem Inoculations. 72 24. Secondary tumors in various places on leaf of Paris Daisy developing from the internal tumor strand. ... 73 25. Solvent Action of Bacillus phytoph- thorus on middle lamella; of cells of Potato tuber 76 26. Aerial Roots on Daisy due to pres- ence of the Tumor farther up on stem 90 27. Swelling on Potato-shoot due to Inocu- lation with nonvirulent Bad.solana- cearum 91 28. Sugar-beets Inoculated with Bad. tumejaciens (plated from tumor on Paris Daisy) 92 29. Sugar-beet Inoculated with Bad. tumejaciens (plated from Crown- gall of Peach) 93 30. Tyloses due to Bad. mori in the vessels of the Mulberry 94 31. Bacterium leguminosarum: Effect of Bacterial Occupation on Cell- nucleus in Soy-bean 97 32. Effect of Root-Nodules on a Pea grown in confined air on Nitrogen- free Soil 98 33. Worouine's figures ioj 34. Bacterium leguminosarum: Cross-sec- tion of small Root-nodule of Soy- bean showing tissue occupied by the Bacteria 1 04 35. Bacterium leguminosarum: Effect of repeated Freezings 105 Page Fig. 36. Bacterium leguminosarum: Zoogloeae threads growing through cells in Root-nodule of Common Pea 1 10 37. Bacterium leguminosarum: Stages in formation of Y-shaped Bacteroids from simple Rods . . : 113 38. Kefir grains. After Kern 156 39. Dispora caucasica. After Kern 157 40. Bacillus caucasicus. After von Freu- denreich 161 41 . Marshall Ward's Bacterium vermiforme 164 42. Same, showing splitting of sheaths. . . 165 43. The Ginger-beer Plant in a suitable Saccharine Medium 165 44. The same, in an ordinary unsuitable medium (bouillon) 165 45. Grains of "California beer-seed". .. . 166 46. Section free-hand through one of the preceding, the darker bodies being groups of Yeast imprisoned in Masses of the Bacteria 167 47. A bit of the preceding crushed out in water 167 48. Another view showing oblong and round yeasts, contents omitted.. 167 49. Spores in Yeast Cells taken from "Beer-seed" 167 50. Hubbard Squashes Wilting from an attack of Bacillus tracheiphilus 210 5 1 . Bacterial Ooze from vessels of Cucum- ber Stems 211 52. Figure showing Viscidity of B trachei- philus 212 53. Squash plant No. 215 dwarfed by B tracheiphilus (upper figure) 213 54. Cross-section of Cucumber-stem, show- ing location of Bacteria in a bundle . 214 55 Diabrotica vittata, distributor of the Bacillus of Cucumber- wilt 216 56. Progress of wilt by hours on an Inocu- lated Leaf of Cucumber (plant No. 2) 220 57. Bacteria from Viscid Slime in plant No. 8 222 58. Diagram of vine No. 15, attacked by Bacillus tracheiphilus 223 59. Diagram showing distribution of Bacillus tracheiphilus in the Vascular Bundle of a Cucumber Plant 8 days after the first appearance of the wilt (2 days after the appearance of sec- ondary wilt, and 16 days after the Inoculation) 225 60. 61. Two stages of Wilt in Inoculated Leaf on plant No. 27 228, 229 62. Cucumber from a field, showing in detail the Infection of a Single Bundle of the stem by B. trachei- philus 230 63. Progress of Wilt on Inoculated Leaf of plant No. 28 (cucumber) 231 LIST OF ILLUSTRATIONS. VII Page. Fig. 64. Wound Reaction (third dayjin Pricked Potato-leaf, plant No. 61 239 65. Leaf of Cucumber Plant No. 91 .show- ing Pricked and Wilted Area (the shaded part) 242 66. Progress of Wilt in 8 days on Inocu- lated Leaf of plant No. 113 247 67. Cross-section of Inoculated Plant No. 149 at extreme top of stem, showing Infection of all the bundles 250 68. Lamina of Squash-leaf (in section), showing the Bacteria confined to the bundle although Wilt involved the Parenchyma 251 69. Infected Bundle of Inoculated Musk- melon No. 150 252 70. Sketch of Swollen Bacteria from the Interior of Vine No. 199 256 71. Wilt of Inoculated Winter Squash at end of 9 days, plant No. 215 258 72. Progress of Wilt in Inoculated Squash leaf on plant No. 216 259 73. Inoculated Squash-leaf on plant No. 223 261 74. Progress of Wilt on Leaf of Inoculated Plant No. 245 265 75. Wilt on Leaves of Cucumis melo, var. dudaim, plants 276, 277 269 76. Wilt on Inoculated Leaf of Cucurbita californica, plant No. 280 269 77. Cross-section of Petiole of Inoculated Leaf of Cucurbita foetidissima (Plant 273), showing Bacteria con- fined to the Vascular Bundles 272 78. Detail from Vascular Bundle of Cucur- bita foetidissima occupied by B. tracheiphilus 273 79. Single Spiral Vessel of Cucumber Stem in cro^s-seetion, showing Lumen plugged by B. tracheiphilus 285 80. Cross-section of Petiole of an Inocu- lated Squash-leaf — 12 bundles oc- cupied by B. tracheiphilus 286 81. Cross-section of a Squash Bundle destroyed by B. tracheiphilus. It also shows occupation of surround- ing parenchyma 287 82. Camera drawing of unstained B. tracheiphilus from stem of a Wilted Cucumber from the field (1893) .... 288 83. Sketches of dividing rods of B. tracheiphilus after staining with nitrate of silver 288 84. Flagellate rods of B. tracheiphilus, capsule also stained 288 85. Gelatin-stab cultures of B. trachei- philus 288 86. Gelatin-streak culture of B. trach- eiphilus 290 87. Petri Dish Agar Poured-plate of B. tracheiphilus at end of 6 days 291 Page. Fig. 88. Small portion of Agar Poured-plate at end of 7 days in thermostat 291 89, 90. Enlarged Colonies of B. trachei- philus 291 91. Streak Cultures of B. tracheiphilus, showing discrete growth 292 92. Behaviorof B. tracheiphilus in Fermen- tation-tubes 293 93. Simple apparatus used for Testing Growth in Hydrogen 294 94. Restricted growth of B. tracheiphilus on a Slant Agar Streak buried under more agar 29s 95. Restricted growth of B. tracheiphilus in Acetic Acid Agar 29.=; 96 Crystals in old Litmus-milk Cul- tures of B. tracheiphilus 296 97. Germicidal Action of Sunlight on B. tracheiphilus 297 98. Cabbage-leaf showing upward move- ment of Brown Venation due to Bad. campestre 301 99. Dwarfing and Loss of Leaves in Cauliflower attacked by Bad. cam- pestre 302 100. Head of Cabbage showing Blackened Vascular Ring due to Bad. cam- pestre 303 101. Cauliflower Stem from Texas, show- ing Blackened Vascular Ring due to Bad. campestre 303 102. Cross-section of Petiole of Cabbage, all the bundles of which are black- ened by Bad. campestre 304 103. Cauliflower Stems from Florida, show- ing disease further advanced, i. e., cavities in the pith 305 104. Cross-sections of Kohlrabi showing Blackened Vascular Bundles in White Flesh 306 105. Stem of Collards attacked by Bad. campestre 307 106. Turnip-root hollowed out by Bad. campestre 308 107. Cavity in fleshy part of Kohlrabi due to Bad. campestre 309 108. Cabbage-leaf showing wedge-shaped area of Bacterial Infection (Black Venation) attributed to insect Gnawings 310 109. Cross-section of Cauliflower Petiole showing tissues occupied by Bad. campestre 311 no. A detail from fig. 109 3'2 in. Cross-sec tioii of Turnip-root showing two vessels, one fully occupied by Bad. campestre 313 112. Longitudinal section of Turnip-root showing bacteria in a vessel 314 113, 114. Bacterium campestre occupying Intercellular Spaces 315.316 VIII LIST OF ILLUSTRATIONS. Page. Fig. 115. Solvent action of Bad. campestre on middle lamella' 3 ' 7 116. Beginning of cavity in a Turnip-root. 318 117. Closed cavity in a Turnip-root due to Bad. campestre 318 118. Cross-section of Turnip-root showing many cavities and vascular occlu- sions due to Bad. campestre 318 119. Cross-section of Rape-petiole showing Bacteria confined principally to the Bundles 3 1 9 120. Longitudinal section showing non- lignified cells of Turnip-root filled with Bad. campestre. These border 1 hi a reticulated vessel 3-° 121. Flagellate Rods of Bad. campestre from an Agar Culture 321 122. Rods of Bad. campestre from Stem of Charlock (polar staining 3) 321 123. Rods of Bart, campestre from Kohlrabi from a Gelatin Culture 321 124. Rods of Bad. campestre from a smear from a Cabbage-stem, enlarged from a photomicrograph 321 125. Rods of Bad. campestre from a Potato Culture kept at 120 C 322 126. Ditto from a 48-hour Bouillon Culture at3o" C 322 127. Ditto from a very old browned Potato Culture ;-- 128. Portions of three Agar Plates poured from stem of Collard (fig. 105 at xi Size of colonies is reduced by crowding 3-i 129. Liquefaction of Gelatin by Bad. campestre 325 130. Nongrowth of Bad. campestre over Chloroform 327 1300. Left. Cabbage leaf extruding fluid from water pores. Right. Cabbage leaf showing early stage of marginal (waterpore) infection 332 Pay e Fig. 131. Cross-section of Hyacinth Bulb, show- ing an early stage of Infection (7 Yellow Bundles) due to Inocu- lation of Scape .\\<> 132. Cross-section of a Hyacinth Bulb Infected from a Flower Stalk 337 13.?. Haarlem Bulb Scale, showing Dis- organization of Parenchyma 34 1 134. Cavity in Parenchyma at extreme base of a Bulb-Scale (inoculated plant) 342 135. A detail from fig. 134 342 136. Cross-sections of badly Yellowed Bulbs (Wakker's disease) obtained by the writer in Holland 343 137. Cavity in an Inoculated Leaf 344 138. A detail from 137 345 139. 140. Rods of Bad. hyacinthi 3;s 141. Non-segmented threads (filaments) of Bad. hyacinthi 345 142. Flagellate rods of Bad. hyacinthi 343 143. Action on Milk of lab ferment pro- duced by Bad. hyacinthi 346 144. Stab-eultures in Gelatin containing Malic Acid 347 145. Behavior of Bad. hyacinthi in fermen- tation-tube in presence of Maltose and Peptone 34s 146. Behavior of Lead Acetate Paper ex- posed over cultures of Bad. hyacinthi 349 147. Parts of Agar Poured-plates of Bad. hyacinthi, showing slow growth of buried and surface colonies in pure culture 349 148. Shagreened surface sometimes seen in Streak-cuLures of Bad. hyacinthi on Sugar-agars !5o Bacteria in Relation to Plant Diseases By ERWIN F. SMITH WILT OF CUCUMBER. (I) Cucumber leaf from Anacostia, D. C wilt due to In the center of the wilted area are old gnawings due to Z> I from these the inlection is believed to have originated. (2) A cucumber stem from the same field. ing the disease much further advanced, although the stem and base of petioles are normal externally. Figures painted Ju Litmus-milk-culture of ,.ish strain) inoculated August 28, 1903. Painted September 23. (4) Uninoculated tube for comparison. Stem and leaf by D. G. I Bacteria in Relation to Plant Diseases. BY ERW1N F. SMITH. INTRODUCTION. This volume really begins the subject of bacterial diseases of plants, the first volume having had for its aim only the clearing of the ground by a discussion of methods of work and the general subject of bacteriology. Whatever in that volume relates to specific dis- eases of plants was introduced merely by way of illustration, or to provoke interest in what should follow. The first part of this second volume deals with general questions relative to bacterial diseases of plants; the history of the subject, the distribution of bacteria on the surface of plants, the questions involved in the terms parasitism and symbiosis, the action of the bacteria on various tissues, the reactions of the plant, the interrelations of animal and plant parasites, and, finally, the problems relating to prevention. The wilt of cucurbits, the black rot of crucifers, and the yellow disease of hyacinths are then dealt with in separate chapters. In researches of this kind, covering as they do a relatively new and rapidly enlarging branch of science, the point of view changes with great frequency. To-day the interest will be centered on one phase of the subject, to-morrow, perhaps, on some quite different aspect. Fortunate for the experimenter if the new aspect do not require entrance into unfamiliar fields of discouraging complexity. For anyone to cover adequately by the experimental method a whole branch of science if it be a large one, is manifestly impossible. There will always be portions slurred over. The best that one can do is that which I have tried to do, viz., to point out at frequent intervals gaps in our knowledge, to express things clearly and honestly, to distinguish between verified fact and speculation, and, finally, to leave each subject in somewhat better shape than I found it. The following pages are based in great part on data obtained as the result of a multi- tude of experiments made by the writer and his assistants, but all sources of published information have been considered. During the writing of this monograph it has often happened that the ink on some chapters would scarcely be dry before the results obtained from new experiments would require some part of it to be rewritten. In this way during the last ten or fifteen years the subject has been worked over and over, some chapters being rewritten a dozen times, in whole or great part. This, while greatly increasing the scientific value of the work, has certainly not tended to improve its style. The long period covered by these experiments must also serve to explain why the description of particular organisms does not in all respects follow the recommendations of the chart recently issued by the .Society of American Bacteriologists, many of these studies having been completed before that was begun. To make them all conform would delay 4 BACTERIA IN RELATION TO PLANT DISEASES. this publication indefinitely since the writer no longer has living cultures of some of these organisms. Troubles of other sorts have been encountered frequently and the reader would be surprised, no doubt, to know how difficult it has been to obtain exact information on some of these subjects. Some definite rule must govern the citation of literature when it is very abundant as in case of some of these diseases, e. g., potato rots and pear-blight. Purely agricultural or horticultural literature has not been cited in this monograph unless it has a direct bearing on questions under consideration. Many speculative writings have been excluded. For this reason literature on any given disease earlier than that definitely ascribing it to a bacterial origin is, as a rule, not cited. Any other rule would have led to an endless number of citations of very little worth. Perhaps not enough exclusions have been made. In this matter it has been thought best to err on the side of fullness. Probably also some papers have been overlooked since the literature is scattered through many languages. Occa- sionally a citation has been made simply to show geographical distribution. As mentioned in vol. I the writer considers it advisable to state whenever possible the exact temperatures at which experiments were made, but it happened frequently in the great mass of notes from which the following pages have been compiled, that the expression "room-temperature" is used. It may therefore be useful to certain readers to know that the Washington room-temperatures, i. c, those of our laboratory, vary roughly as follows: Summer temperature 250 to 350 C, occasionally 380 to 400 C. ; winter temperatures (heated rooms) i8°C.to 270 C, usually about 250 C; spring and autumn approximately 200 to 25°C. While not averse to synthesis, the writer has usually followed the analytical method. In general, "lumping" things not known positively to belong together is a worse proclivity in natural history than excessive subdivision. Further experiments are often necessary, and until such time it is best to keep separate subjects not demonstrated to be identical; at least the writer has striven to follow this rule. The whole trend of modern scientific research is toward analysis of phenomena, and only in the later stages of knowledge do combinations come in properly to round out a subject throughly worked over. When one has to deal with many diseases some sort of nosology becomes necessary. That which appears to be most convenient for the purposes of this treatise is, first of all, the simple subdivision into three large groups: (1) the vascular diseases; (2) the parenchyma diseases without hyperplasia; and (3) cankers, tubercles, and tumors in which there is a more or less distinct hyperplasia. The reader should remember, however, that classifica- tions are only conveniences. There is marked bacterial occlusion of the vessels in those diseases which I have classed as vascular — occlusions so extensive as to render this feature of the disease most conspicuous, but it does not follow that there is not also some destruction of the paren- chyma. Vascular bundles are not on the surface of the plant, and some preliminary bac- terial destruction of the surrounding parenchyma must always occur before the disease can take on its true vascular character, except perhaps in those comparatively rare cases where the inoculation happens to be made directly into some bundle. Moreover, in later stages of these vascular diseases bacterial pockets of greater or less extent are often formed in the parenchyma, especially in its softer parts. The extent to which these closed cavities occur varies greatly in different diseases. In the brown-rot of potato and tomato they are numerous and often fuse into large tracts of disintegrated tissues (fig. 1). In Stewart's disease of sweet-corn, on the contrary, they are neither very large nor very numerous. Exception, however, should be made of the inner husks where they are common. In the remaining groups of diseases, occurring with or without hyperplasia, it is not uncommon to find occlusion of some of the vessels, although the first and principal dis- turbance occurs in the parenchyma. As examples of this may be cited the basal stem-rot of INTRODUCTION. Fig. 1.* *Fig. I. — Cross-section of potato stem of plant 13 (1895) inoculated with Bad. solanacearum byjneedle-pricks May 27, and fixed in strong alcohol June 11. Outer tissues are very little affected except on one side (top of drawing) where there is much disorganization. At X there is a deposit stained red by safranin. This local red stain occurs here and there in other parts. Twelve crystal-sand cells occur in the outer phloem. Stem hollow. Actual diameter of section in longer axis, 4.5 mm. Slide 194(2). 6 BACTERIA IN RELATION TO PLANT DISEASES. potato, the spot disease of beans, the blaek-spot of plums, and the olive-tuberele. Bacteria frequently occur in the vascular bundles of plants suffering from these diseases especially after considerable destruction of the parenchyma, but they are not vascular diseases in any such specific sense as the wilt of cucurbits or the black rot of cabbage. Occasionally also tumefactions, or the stimulation to growth of dormant organs, may be induced by organisms which appear to have little in common with those which regularly produce tubercles and tumors, e. g., the normal action of Bad. solanacearum on the tomato, inducing the premature development of adventitious roots and the occasional swelling of tissues when inoculated with cultures which have lost their virulence. The true tumors, *. e., the crown-galls, appear to be an exception to this rule. In these I have not seen any bacterial occupation of the vessels or intercellular spaces. Various subdivisions of these three groups, especially of the second and third, will become necessary and will be made use of in the proper place. With these explanations and qualifications, we may proceed to the subject in hand, noting, in conclusion, that in this volume as in the preceding the illustrations, so far as possible, have been drawn from the writer's own material and were made under his personal supervision, mostly by James F. Brewer. In case of drawings from sections the writer not only selected the part to be illustrated but also checked up the finished drawing line by line under the microscope, so that a fair degree of accuracy may be assumed. In most instances, however, the slide number is given under the figure, and, in case of doubt, these slides are on file for reference in the collections of the Department of Agriculture. HISTORY. THE EARLIEST WORKERS. The earliest workers in any field of science deserve special consideration. They are like pioneers in a new country, and are usually poorly equipped for their task. If, under such circumstances, a man makes substantial additions to human knowledge, he deserves corresponding credit. Such a man is what we might call a born investigator. He is able not only to see his problem as a whole, but also to see it in its parts, and to determine their interrelations. The earliest investigators in the field of plant-diseases due to bacteria were Burrill, Prillieux, Wakker, Comes. These men wrought independently — the first in the United States, the second in France, the third in the Netherlands, and the fourth in Italy. All are yet living. To these names I would add that of Woronin, whose one contribution was the discovery of bacteria in the root-nodules of the Leguminosae. Burrill's principal contribution consists in the discovery of the bacterial origin of pear- blight. The disease had been known in the United States for a hundred years and at times had been very destructive. A multitude of hypotheses had been propounded to explain the mysterious phenomenon, none of which really explained. Into this obscurity and con- fusion Burrill let a flood of light by addresses and papers published between the years 1878 and 1883. So far as this country is concerned, he may be said to have won over the public in 1 88 1 . Many things yet remained to do — things afterwards done by Arthur and Waite — but on the main proposition, namely, that pear-blight could be attributed only to bacteria, Burrill's experiments appeared to be conclusive. Burrill's work was done at the University of Illinois, located at Urbana, where he holds the chair of Botany. Burrill subsequently published short papers on several other bacterial diseases, i. e., potato-rot, disease of maize, and disease of broom-corn. But to none of these subjects does he seem to have been able to give his undivided attention, most of the conclusions depending in part at least upon the work of students. He likewise published papers on fungi, notably the Uredineae of Illinois, in connection with Seymour. Prillieux published his first paper on a bacterial disease of plants in 1879. This con- sisted of an account of a microscopic examination of wheat-kernels in which he found clouds of a micrococcus-like organism eroding the interior into distinct cavities. This is generally known in literature as the rose-red disease, or Micrococcus disease, of wheat- kernels. Prillieux did not make any pure cultures or inoculations, and the disease seems to be a rather uncommon one, so that no one has been able in recent years to control his observations, but the account of his microscopic examination is explicit and his figures are not obscure. Subsequently, Prillieux published on various other diseases of plants ascribed to bacteria, e. g., on the olive-tubercle and the Aleppo pine knot, but most of his energies have been given to the elucidation of diseases of a fungous nature. His text-book on diseases of plants is well known. He is professor at I/Institut National Agronomique and was formerly in charge of the plant pathological laboratory, No. 1 1 Rue Alesia in Paris. Wakker published his first paper on the yellow disease of hyacinths in 1883. Subse- quently he published four other papers on this disease, the last in 1889. For a long time his statements, mostly in Dutch, were overlooked or not generally accepted as conclusive, but none of the early work was better done, and numerous experiments made by the writer have shown that he was entirely right in his main contentions. The yellow disease of hyacinths is a genuine bacterial disease and can be induced in the bulb by inoculating the 8 BACTERIA IN RELATION TO PLANT DISEASES. leaves with a yellow organism, the disease progressing in just that slow regular way described by Wakker as a result of his own inoculations. Wakker worked at that time in the labor- atory of Hugo de Vries in Amsterdam, under a royal grant obtained by the hyacinth growers of Haarlem. Afterwards, for five years, he was in charge of one of the Javan sugar experiment stations. He has published various papers on fungous and other diseases of plants, those of recent years being devoted largely to the sugar-cane. Comes was one of the earliest workers in this field. He recognized bacteria in the tissues of various diseased plants in southern Italy as early as 1SS0, and published a number of papers on Bacterium gummis, which he believed to be a widely distributed para- site attacking many plants. He did not, however, grow the organism properly in pure cultures and secure infections, nor describe it so that one can now be certain of its identity. He early turned his attention to other subjects, much of his energy being given to teaching botany in the Royal Agricultural College at Portici, of which he is now director. He is the author of a general text-book on botany, of a book on diseases of plants, and of elabor- ate treatises on tobacco, as well as of numerous minor publications, and his students are scattered all over Italy. He seems never to have had any doubt as to the occurrence of bacterial diseases in plants. Sorauer also being one of the voluminous writers on plant pathology should be men- tioned here, because, as early as 1886, he saw clearly that Wakker and the others were right. Sorauer regarded the bacterial diseases of plants ony from the general standpoint of a writer and student of plant pathology. He examined various such diseases microscopi- cally but did not make pure cultures or inoculations therefrom, not having had the neces- sary grounding in bacteriological technique. He endured much obloquy in his earlier years for steadily maintaining the existence of such diseases in opposition to Hartig and his school, but he has lived to see his contentions established. Sorauer is the author, among other works, of the most elaborate and important German handbook on diseases of plants, the third edition in three volumes having been completed recently. He is also the founder and editor of the Zeitschrift fiir Pflanzenkrankheiten. Savastano and Arthur began work a little later than the others when there were not so many difficulties in the way. Savastano conceived the idea that the olive-tubercle was caused by bacteria. He demonstrated the constant occurrence of bacteria in the knots, isolated them therefrom in culture-media, and produced typical overgrowths on healthy olive shoots by punctures into which minute quantities of the culture were inserted. Cavara went over Savastano's experi- ments soon afterward with similar results, and the writer and his assistants have done the same thing in recent years (see vol. I, plate 2). This was Savastano's most important incursion into bacteriology. He described, however, in an imperfect and fragmentary way a number of other diseases as bacterial, most of which are really such. He is the author of a handbook, Patologia Arborea Applicata, and the editor of the Bollettino della Arboricol- tura italiana. Savastano held for many years the chair of forestry in the Royal Agricultural College at Portici and is now director of the experiment station at Acireale in Sicily. Arthur repeated and verified Burrill's work on pear-blight and carried the investiga- tions somewhat further, demonstrating that infectious fluids filtered through porous clay cups lost their power to produce the disease, whereas the residue containing the bacteria was as infectious as ever. Arthur began his experiments about the time that Burrill ceased, and continued them for a number of years, publishing a half dozen papers on pear-blight and the organism to which it is due, these papers forming his most important contribution to plant bacteriology. He was at that time stationed at the Geneva experiment station in central New York. Since then he has been professor of botany in Purdue University, Indiana, and has published interesting papers on fungi and other plants. He is the author, with Barnes and Coulter, of A Handbook of Plant Dissection and has published with THE EARLIEST WORKERS. 9 MacDougal a collection of essays entitled Living Plants and their Properties. His cross- inoculations of Uredineae in recent years have settled many doubtful points. For many years he was one of the editors of the Botanical Gazette. Earlier than any of these writings is the paper by Woronin (1866) announcing the discovery of bacteria in the root-tubercles of lupins. His conclusions were based simply on microscopic examinations, but they were confirmed by Beyerinck two decades later (1888) and since that date a great literature has arisen. Woronin studied under De Bary and afterwards published several papers in conjunction with him. He was born in St. Peters- burg and lived there during the latter part of his life. He is justly famous for his beau- tifully executed monographs on various pathogenic fungi: Tuburcinia, Sclerotinia, etc. Beyond this one early research, so far as known to the writer, he did nothing with bacteria. About the same time (1868) the Frenchman, Davaine, a man of marked originality and excellent powers of observation, showed that certain bacteria were able to produce a soft-rot in plants when inoculated, and this observation was subsequently confirmed by van Tieghem and many others (see Destruction of Cell-walls, etc.). In conclusion, one ought not to forget Anton De Bary (1831 to 1888). The present has its roots in the past and his profound influence still lives. De Bary published nothing on bacterial diseases of plants but he made it possible for others to do so. Directly or indirectly all the early workers in this field were influenced by him. The same is true of those who followed, and all of us are much indebted to him. THE GENERAL ATTITUDE OF PATHOLOGISTS AND BACTERIOLOGISTS. In the consideration of this topic it is hardly worth while to dip into the ill-digested mixture of fact and speculation published prior to 1874. I shall therefore begin with Sorauer. In the first edition of his comprehensive Handbook of Plant Diseases, published in 1874, Sorauer makes no mention of bacterial diseases of plants. A great many diseases are included, but none of this type. De Jubainville and Vesque writing in 1878 mention a ' ' cellular rot ' ' of potatoes, radishes carrots, and beets, occuring in the soil or in cellars, but they attribute it to soil ill-adapted to the plants or to improper cultivation. No mention is made of bacteria either as the cause of this rot or of any other disease mentioned by them. There is nothing on this subject in Winter's little book, published in 1878. Reinke and Berthold, who published in 1879, found that potato-rot could go on inde- pendently of the presence of fungi. They say: "But there are also wet-rotten potatoes which show no trace of these Pyrenoniycetes." A little farther on we are told correctly that : "A tuber can become wet-rotten without ever having been diseased by Phytophthora.' ' And again, It has already been mentioned above that in addition to the Myxomyeetes occasionally found upon wet-rotten potatoes, bacteria are also present. In fact, these bacteria are a constant accom- paniment of the wet-rot, and this is an indication as to the true cause of this decay. If a potato- tuber infected by Phytophthora lies in the wet, it passes quickly over into the wet-rotten condition. As soon, however, as the first symptoms of the wet-rot appear, bacteria are found in quantity in the wet-rotten tissue. Notwithstanding this one might think that the Phytophthora caused the wet-rotten decomposition, * * * ancj that the bacteria hastened the decay only secondarily. But if one inoculates a tuber, which is entirely sound and free from Phytophthora, with the bacterial fluid from a wet-rotten potato, there is always a local production of the wet-rot in the tuber which has been kept moist, and not rarely does it become totally wet-rotten in a very short time. The cause of the wet-rot can be considered to be, therefore, only the bacteria and the ferments produced by them, the tissue of a potato-tuber being only specially predisposed for the action of the bacteria by the Phytophthora. IO BACTERIA IN RELATION TO PLANT DISEASES. These men repeatedly produced wet-rot of the potato by direct inoculations. Their method consisted of cutting out a little tetrahedron from a tuber, to a depth of 10 to 15 mm. A drop of fluid containing the bacteria was put into this wound. The tetrahedral piece of the tuber was then put back and pressed in. When the tuber was placed with the wounded side up, the cut portion usually dried out, a cork-layer formed under the cut surface, and the wet-rot did not occur. On the contrary, when the tuber was placed with the wounded side down against wet paper under a bell-jar, in a saturated atmosphere, there was always more or less decay of the tuber, and sometimes within 2 or 3 days the whole interior became wet-rotten. In no case did Reinke and Berthold experiment with pure cultures nor could they have had any active parasite, but we are warranted in believing that sometimes at least their mixtures of bacteria were free from filamentous fungi. They found different tubers to possess very different powers of resistance, as will be mentioned in another place. The first edition of Frank's Diseases of Plants, published in 1880, contains a brief chapter on root-tubercles of Leguminosae, but nothing on bacterial diseases of plants. The following translation from page 27 of Hartig's Lehrbuch, published in 1882, shows the general attitude of botanists and pathologists at the time Burrill and Wakker were working out their interesting and beautiful results. With the pathological processes in plants they have nothing whatever to do; in fact, I have never on any occasion found the schizomycetes in the interior of a closed plant tissue, and they have nothing to do with the falsely so-called rotting processes of dead plant tissue. Of course, this does not exclude them from a share in the destruction of dead vegetable substance whenever they find an easy access to it. Evidently the interior of the plant is difficult of access to them because the open circulatory paths are wanting, which in animals make possible their rapid distribution with the blood. Also the circumstance that the wall of the plant-cell is nitrogen-free and is mostly very thick in comparison with the size of the schizomycetes, must be an obstacle to the wandering of the schizomycete from one cell to another. Finally, also, the formation of humus acids in the dead plant tissue will tend to prevent the multiplication of the schizomycetes. In Worthington G. Smith's book (1884) there is nothing on bacteria as a cause of disease in plants, although two long chapters are devoted to "The Potato Disease," as though there were but one. In 1884, in his Comparative Morphology, etc., De Bary considered the subject very briefly, but much more cautiously than Hartig: As Hartig has already pointed out, bacteria living in plants parasitically have scarcely been observed. The generally acid reaction of plant parts may be a partial explanation of this. Recently, however, Wakker has described as the yellow sickness, a disease of hyacinths in Holland, in which the characteristic symptom consists in the presence of slimy yellow bacterial masses in the vessels, etc. More exact investigations upon this phenomenon must be awaited. The following year, in his Vorlesungen, De Bary devoted two pages to this subject. Parasitic bacteria as causes of plant disease have been only infrequently observed. The most of such diseases are due to animals and plants of other groups, especially to fungi. Wakker's yellow disease of hyacinths is described briefly from that author's first papers with the remark that "successful infection experiments and the exact following out of the life history of the bacterium are still to be awaited." In the same way, Burrill's work on pear- blight and apple-blight is mentioned, without other comment than that "in Europe this phe- nomenon, so far as I know, is not known, or at least has not been carefully investigated."* Prillieux's studies of changes in wheat-grains are mentioned with the remark respecting the micrococcus that "its importance as a cause of disease can not be judged with any certainty from the short account. It may turn out to be only a saprophyte appearing in consequence of other injuries." Finally, De Bary mentions the wet-rot of potatoes, studied by Reinke *The disease of the peach tree, of the Lombardy poplar, and of the American aspen, mentioned by De Bary on Burrill's authority as bacterial diseases are now believed to be due to other causes. ATTITUDE OF PATHOLOGISTS AND BACTERIOLOGISTS. II and Berthold, with the remark that recent experiments by van Tieghem on potato-tubers, bean-seeds, eaetus-stems, etc., seem to confirm these results. Otherwise expressed, these facts may be stated as follows: As a rule, saprophytic bacteria may, under special conditions, attack, make sick, and destroy living plant tissues as facultative parasites. In the third edition of Zopf's Spaltpilze, published in 1885, bacterial diseases of plants are discussed as follows: Much rarer [than in men and animals] are the cases in which we can speak of the genuine parasitic action of schizomycetes in plant-organs. The best-known example is the familiar disease of potato- tubers known as "wet-rot" caused by the butyric acid ferment (Clostridium bulyricum), through which the tissues of the potato are entirely destroyed and converted into a vile smelling fluid-mass (Reinke and Berthold). It remains to be seen whether the phenomenon known in Holland as "the yellow disease of hyacinths," and recently described by Wakker, belongs here strictly speaking. Its characteristics are the appearance of enormous numbers of yellow schizomycetous colonies in the vessels and (at blossoming time) also in the intercellular spaces and cells of the parenchyma. Perhaps the rarity of schizomycetous diseases of plants lies in the generally acid reaction of the juices of plants, perhaps also in their lower temperature as compared with the animal body, and finally the formation of protective cork is perhaps also to be considered (p. 3). In the second edition of De Bary's Vorlesungen (1887) Arthur is said to have confirmed and extended Burrill's work on pear-blight. In the second edition of his Handbook, published in 1886, Sorauer devotes 38 pages to diseases due to bacteria, namely, to rot of the potato, white rot of hyacinth bulbs, rot of onion bulbs, Comes's gummosis of the tomato, Prillieux's rose-red wheat grains, and various stem and leaf reddenings. These pages deal with field appearances, the results of micro- scopic examinations, and to a limited extent with what I have designated in vol. I as direct infection experiments. Sorauer accepts the doctrine of bacterial disease of plants without reserve and says: "Beyond doubt, in course of time, a large number of rot diseases will be recognized." Of exact bacteriological methods there are no suggestions in this book. Dr. Sorauer's own observations appear to have been limited to microscopic examinations and a repetition of such crude infections as were made by Davaine, Hallier, Reinke and Berthold, van Tieghem, and others. The bacteriosis which he had in mind is that which occurs when tubers and bulbs are exposed to excessive moisture, with a restricted supply of oxygen. From his own observations and experiments on potatoes, he could say with reasonable assurance that "the wet-rot or rot may be produced artificially without the aid of Phytophthora by inoculating bacteria into sound tubers. The decomposition phenomena of the two diseases are essentially different." He had also observed that the bacteria could penetrate through the open lenticels into the tubers. The statements in Sorauer's book on Die Schaden der Ehiheimischen Kulturpflanzen (1888) are essentially the same as in his Handbook, only much more condensed. His gen- eral standpoint is expressed in the following sentence: "The bacteria are certainly much more dangerous to living plants than has hitherto been recognized." De Toni and Trevisan in vol. VIII of Saccardo's Sylloge Fungorum (1889) gave short descriptions of all the species of Schizomycetes known to literature. Under bacillus "Sectio 6 species endophytobiae, destruentes," the following species are included: Bacillus vuillemini Trev., B. oleae (Arcang.) Trev., B. ampclopsorae Trev., B. radicicola Beyerinck, B. hyacinthi (Wakk.) Trev., B. hyacinthi septicus Heinz, B. sorghi Kell., B. amylovorus (Burr.) Trev. Laurent, writing in 1889. has the following: Many fungi which invade the higher plants have the property (propriete) of perforating the cell membranes by the intervention without doubt of a special zymase. The germs of the ordinary bacteria could easily penetrate into the leaves by way of the stomata when they are brought there by the wind or other agents. But having reached the stomatic chamber, they would have to traverse the cell membranes or to insinuate themselves between the cells. In order that this last mode of 12 BACTERIA IN RELATION TO PLANT DISEASES. infection might succeed, it would be necessary to suppose bacteria mobile, capable of crawling into the intercellular spaces, something quite improbable (ce que est assez peu vraisemblable), or else filamentous forms with continuous development, in the manner of the mycelial filaments of fungi. As to short non-motile bacteria, they would have to traverse the cell membranes. * * * Now I have determined that the cellulose, even of the most tender varieties, resists perfectly, exposed to the air, a great number of kinds of common bacteria. The solvent action of Bacillus amylobacter takes place only in the absence of oxygen. Nevertheless, according to Vignal, the Bacillus mesen- teric us vulgatus secretes a zymase which digests the most tender celluloses. I have made the same observation in case of a Bacillus subtil is, which, when developed in mycoderma on the surface of a liquid, separated the cells of a bit of potato situated in the depths of the same liquid. In con- clusion, if the penetration of the cell-membranes of plants is not a general property of the bacteria it does occur and may perhaps be developed in a particularly favorable medium. This condition is nevertheless not sufficient to enable the bacteria which are outside to estab- lish themselves in the tissues of plants. It is necessary also to take account of the resistance peculiar to living cells, a resistance the mechanism of which is still entirely enigmatic. Among animals, the pathogenic bacteria overcome this difficulty by the production of substances more or less toxic, rapidly diffused through the entire organism by way of the blood stream. The higher plants have the advantage of being much more resistant to the movement of the microbes and of their secretions through their tissues. Consequently there exist few bacterial diseases among plants, while in the animal kingdom there are a great many of them. Kirchner (1890) mentions two bacteria, Bad. termo Ehr. which has been said to take part in the destruction of cells in the interior of sorghum-stems, and Clostridium butyricum Prazm., which "causes the wet-rot and dry-rot of potato-tubers and the rot of onions; also on the roots of apple trees, pear trees, plum trees, and cherry trees." Scribner (1890), says in his preface: "We are told that bacteria cause pear-blight and some other plant diseases," but does not mention the subject in the body of the text. In his Diseases of Plants, published in 1890, Marshall Ward does not mention bac- teria as one of the causes, although he also has a chapter on "the potato disease." Comes (1891), like Sorauer, admits without reserve the existence of such a class of diseases and treats the subject constructively, in a space of 38 pages. The organisms con- sidered are Micrococcus amylovorus , Bad. gummis, Bad. zeae, Streptococcus bombycis, Bacillus sorghi, B, amylobacter, B. hyacinthi, B. caulivorus, B. vuillemini, B. oleae, B. ampelopsorae, and B. radicicola. Ludwig (1892) also recognizes the existence of bacterial diseases of plants and devotes about 8 pages to the subject, but this does not include any original work. Loverdo (1892) recognizes the existence of bacterial diseases and devotes a number of pages to an account of the sorghum blight attributed to Bacillus sorghi. By far the best paper of its time (1892) is that written by Migula for the Middle Java Experiment Station. Without personal knowledge of bacterial diseases, but with a knowl- edge of most of the literature, and a logical mind, Dr. Migula applies the ordinary rules of pathological inquiry to the question of the existence of bacterial diseases of plants, and conies to the conclusion that five only out of about twenty which are mentioned deserve to be considered as clearly established. These are pear-blight or apple-blight, sorghum- blight, Burrill's bacterial disease of maize, Heinz's rot of hyacinth, and Kramer's wet-rot of the potato. Russell, who published the same year as Migula, also admits the existence of bacterial diseases of plants, but like Migula, his observations were based mainly on the work of others. In tables at the end of his paper, he mentions 13 diseases as of established bacterial origin and 9 as "probably of bacterial origin," i. e. 22 in all. The following were included in his first class: pear-blight, Burrill's sorghum-blight, Burrill's corn-blight, Wakker's yellows of hyacinth, Heinz's hyacinth rot, tuberculosis of olive, and of Aleppo pine, blight of oats, Arthur's carnation blight, Bolley's potato scab, Burrill's wet-rot of potato, Kramer's wet-rot of potato, sugar-beet disease of Arthur and Golden. In the second class were geranium- blight of Prillieux and Delacroix, Halsted's cucumber and tomato-blight, Halsted's root-rot ATTITUDE OF PATHOLOGISTS AND BACTERIOLOGISTS. 13 of vegetables, Garman's eabbage-rot, Sereh disease of sugar-cane, Comes's gum diseases, Halsted's celery-blight, Ludwig's white slime-flow, Tudwig's brown slime-flow. Bacteria are not recognized as the cause of any grape diseases by Viala in the second edition of his Diseases of the Vine, published in 1887. In the third edition of this book, published in 1893, two French diseases of the grape are said to be due to bacteria: Pour- riture des grappes and maladie du coup de pouce. In 1894, in the English translation of Hartig's Diseases of Trees, edited by H. Marshall Ward, it is stated that "Only in extremely isolated cases has it been placed beyond doubt that these low organisms are the primary cause of disease in plants." Wakker's disease of hyacinth, the wet-rot of potatoes, the tubercle of Aleppo pine and of the olive tree, and the blight of the pear are more or less grudgingly admitted to be diseases of this class. Concerning the first, it is said that "under normal conditions the bacteria do not attack perfectly healthy bulbs" and that "a species of Hyphomyccs almost always accom- panies the disease. ' ' Under wet-rot of the potato the editor hazards the following statement : It is extremely probable that in this and other similar cases the minute bacteria travel in the tissues down the tubes of the filaments (hyphae) of the fungus, feeding on the decomposing pro- toplasmic contents of the latter. Concerning pear-blight we have the following : Lately a disease of apple and pear trees has been described by J. Burrill, of Urbana, 111., under the name of "blight," the cause of which, according to this investigator, is to be ascribed to the invasion of a bacterium. The disease appears to bear resemblance to the tree-canker produced by Nectria ditissima; and as, in the case of this fungus, large numbers of small gonidia resembling bacteria are produced in the cortex, it remains to be seen whether this disease has not been errone- ously ascribed to a bacterium. Prillieux in his text-book on Diseases of Plants, published in 1895, devotes 37 pages to bacteria. He admits the following as diseases of bacterial origin: Rose-red disease of wheat grains; wet-rot of potatoes; white-rot of hyacinths; potato scab; gangrene of potato stems ; disease of grape bunches in France ; Mosaic disease of tobacco ; blight of mulber- ries; point-rot of tomato fruits; black spots in potato tubers; spot disease of sorghum; yellow disease of hyacinths; gummosis of the vine; pear-blight; olive tubercle; pine tumors; white slime-flow of trees; brown slime-flow of trees; black slime-flow of trees. In general, Prillieux's book gives the impression of one who works rapidly and only with the microscope. Von Tubeuf in his book on Plant Diseases, published in 1895, devotes 10 pages out of 611 to this class of diseases. In the introduction he says: While only a few diseases of men and warm-blooded animals are due to the true fungi, etc. * * * the infectious diseases of plants are caused almost exclusively by fungi. * * * Even the few bacterial diseases thus far described are almost all still incompletely investigated and abun- dantly doubtful in two directions. In case of some we have unquestionably to do with a plant disease, more or less exactly known, and it is only the cause of the same which is not yet fully investigated. In these cases, the question then arises whether the disease is due to a microorganism at all and whether this is or is not really a bacterium. But in other cases it is doubtful whether the phenomena, in connection with which the appearance of the bacteria has been observed, are truly to be considered as diseases. On this account we will speak with reserve and briefly upon the bacterial diseases, a labor essentially lightened by the bringing together of the bacterial diseases in the Lehrbuch der niederen Kryptogamen of Professor Ludwig, 1892, and the critical examination of the same from the bacteriological standpoint, by Dr. W. Migula. Then follows a brief account of the 5 diseases reckoned by Migula as of bacterial origin, and also the mention of 17 others ascribed to bacteria by various persons, and concerning which Dr. von Tubeuf says in the introduction, speaking very cautiously, as one unfamiliar with the subject : But we will also here refer briefly to those diseases in which bacteria are suspected of being the'cause. 14 BACTERIA IN RELATION TO PLANT DISEASES. Hallier (1895) asserts the existence of bacterial rots but does not attempt to make a list of them. They follow fungi or act independently, but in all cases, Hallier would have us believe that the bacteria themselves have developed out of the plastids (protoplasmic granules) of the fungi. The second edition of Frank's book on plant diseases, published in 1896, contains 1,213 pages (3 volumes), 13 only of which are devoted to bacterial diseases of plants. There is a good deal of internal evidence (careless proof-reading, etc.) going to show that this book was thrown together very hastily. The arguments in the chapter on bacteria in particular are vague and inconclusive. This is made sufficiently plain by the following paragraph : On the contrary, in the plant-world the bacteria have a very subordinate place in the production of diseases. The striking bacterial action on the plant is consequently not of a pathological char- acter, but a profitable symbiosis, to wit, that in the root-tubercles of the Leguminosae. Where one has perhaps the right to speak, in connection with plant diseases, of bacteria as causes of disease is in a number of rot phenomena of certain underground plant parts. Sorauer proposes, under the hypothetical assumption that these diseases are caused by bacteria, to designate the same by the common name rot or bacleriosis. But, in truth, we have here to do, for the most part, with very ordinary rot phenomena which represent the regular end stage of other diseases, in which, demon- strably, the genuine higher fungi, or also other external factors, are the true primary disease- producer, and decay-bacteria appear only secondarily in the tissue, dead in consequence of the disease, and, powerfully hasten the progress of the destruction of the diseased plant parts on account of the decay which they set up; not rarely, also, are associated with other decay-loving fungi, especially moulds. But because, in isolated cases, it has been possible to produce similar decay phenomena by inoculating sound plant parts with bacteria taken from rotting plants, a number of pathologists insist on viewing these bacteria also as primary causes of disease. Moreover, some cases of hypertrophy, that is of true gall formation, are known in which bacteria are said to be the cause. In the following pages we register all that is known of an authoritative character. From this it will be seen that a satisfactory proof for the acceptance of pathogenic bacteria has not been furnished, and that many times people have sought to help out with a supposition of bacteria as a cause, in diseases which may be brought about through another cause, or the cause of which is not easy to discover, or which also have not been sufficiently investigated by the observer in question. Then follows a discussion of the wet-rot of potato, in which there is no mention of the then most important paper on the subject, viz., Kramer's; the white and yellow rot of the hyacinth, in which the two diseases are confused, in which Wakker's five papers are con- densed into four lines, and in which there is no mention of Heinz's paper. The most of the two pages on this disease is devoted to Sorauer and the digest concludes with the state- ment that "there is at least yet no proof of a pathogenic bacterial action." The rot of onions is discussed briefly from data published by Sorauer. Bolley's work on potato scab and beet scab is then considered, following which are notes on various other diseases: olive-tubercle, Aleppo pine gall, rose-red disease of wheat grains, pear-blight, etc. Under bacteriosis of the sugar-beet there is no mention of Kramer's paper or of Sorauer's second note published in 1892. There is evidence throughout that many of the original papers were never seen or, if seen, were not read. The last edition of Fliigge's large general work on microorganisms (1896) contains 1,385 pages, of which three and one-half are devoted to bacterial diseases of plants, following Migula and Ludwig (vol. ir, p. 418; vol. 11, pp. 308 and 328). Most general treatises on bacteriology do not discuss this subject at all and even so extensive an annual compendium of bacteriology as Baumgarten's Jahresbericht omits all mention of bacterial diseases of plants, although the title is inclusive. In the first volume of his System (1897) Migula devotes 12 pages out of 376 to this subject, going over the literature in the same careful way as in his earlier publication. Out of 29 diseases mentioned, the 8 following are considered to be of proved bacterial origin: Sorghum-blight, pear-blight, Cobb's gum disease of sugar-cane, olive-tubercle, Kramer's wet-rot of potato, Heinz's rot of hyacinths, Arthur and Bolley's spot disease of carnation, and Smith's wilt of cucurbits. ATTITUDE OF PATHOLOGISTS AND BACTERIOLOGISTS. 1 5 The English edition of von Tubeuf (1897) does not differ essentially from the German. Frank's Kampfbuch, published in 1897, is chiefly interesting in this connection, because in it the author announces his changed views respecting the existence of bacterial diseases of plants. Concerning them we have the following very cautious recantation (p. 201) : Whether bacteria can be the cause of disease in plants is always a question to be considered with circumspection. In case of the potato-rot this doubt was formerly so much the more justified because we had learned to know a genuine thread-fungus, the Phytophlhora, as the cause of this disease, and consequently the suspicion at once arose that perhaps this fungus was really the true cause of the disease and might have paved the way for the entrance into the potato of the decay bacteria. I myself have held fast to this doubt until quite recently, but must give it up as a result of my own investigations recently instituted. In his Vorlesungen, published in 1897, Fischer takes the ground that there are no bacterial diseases of plants and can not be any for reasons cited, to wit, the bacteria can not enter the plant except through wounds, and their development in the latter is soon stopped by the formation under them of an excluding layer of cork. Stomatal infection is altogether impossible for the reasons stated : Die unverlelzle Pflanze steht mit der Aussenwelt nur durch die Spaltoffnungen in offener Ver- bindung, die selbst sich darauf beschrankt, dass das gegen die Zellen ganz abgeschlossene System der lufterfiillten Intercellularraume mit der Aussenluft kommuniziert. Wenn durch den Wind oder durch Regen Bakterienkeime in die Spaltoffnungen gefuhrt werden, so gelangen sie von hier aus nur in diese Intercellularraume, wo ihnen ausser dampfgesattigter Luft nichts weiter geboten wird, wo alle Nahrstoffe fehlen, ohne die keine Bakterienspore auskeimt, keine Bakterienzelle sich vermehrt. * * * Alle diese Fahigkeiten fehlen den Bakterien, gegen die eine unverletzte Pflanze vollkom- men geschiitzt ist. Aber auch die verwundete Pflanze wurde nur in den geoffneten, verletzten Zellen Nahrstoffe fiir Bakterien darbieten, eine Quelle, die bald dadurch abgeschnitten wird, dass unter der Wundflache eine undurchlassige Korkschicht (Wundkork) entsteht, die jeden weitern Safteaus- tritt aus der Wunde verhindert. Die Wunde bleibt nicht feucht, die verletzten Zellen schrumpfen und trocknen ein und damit ist den Bakterien der Eingang genau so versperrt, wie an der unver- letzten Pflanze. Ihr drohen demnaeh auch keine Wundinfektionskrankheiten durch Bakterien, deren Weiterverschleppung in der Pflanze gleichfalls unmoglich ist. The following is a translation of the entire paragraph : Exclusive of the tubercle bacteria whose wonderful relation to the Leguminosae has been described already (Vorl. X), no single example is yet known of bacteria which can insinuate them- selves into the closed living cells of a plant. The uninjured plant stands in open connection with the outer world only through the stomata, which connection is so limited that the system of air filled intercellular spaces connects with the outer world but is entirely closed to the cells. When bacterial germs are forced into the stomata by wind or rain, they here reach only into these inter- cellular spaces where nothing further is offered to them than vapor-saturated air, where all nutrient substances are wanting, without which no bacterial spore can germinate, no bacterial cell can multiply. Even when such bacteria as can dissolve cellulose (the methane bacteria) are brought into the intercellular spaces they can not nourish themselves here, and can not develop their pecu- liarity of dissolving the cell-wall. Consequently only those parasitic organisms can penetrate into the plant with results, whose spores have brought along with them sufficient nutrient substance so that they can germinate in pure water, so that they can overcome the lack of nutrient substances which they meet with at first, and can open their attack on the protective cell-wall at their own expense. This requirement is fulfilled by the spores of parasitic fungi, which with their reserve stuff push out a germ-tube, which now bores directly through the epidermis of the plant (potato fungus, Phytophlhora infeslans) or which first penetrates into the intercellular system through a stoma (rust fungi), and from here boring through the cell-wall, multiplies in the cells, or at least sends into them special side branches of its mycelium as sucking organs (haustoria). All these peculiarities are wanting in the bacteria, against which an uninjured plant is fully protected. But also the wounded plant offers food for bacteria only in the opened, injured cells, a source which is soon removed by the formation under the wounded surface of an impenetrable cork layer (wound cork) which entirely prevents any further flow of sap from the wound. The wound does not remain moist, the injured cells shrivel and dry out, and consequently the entrance of the bacteria is exactly so barred out as in the uninjured plant. Consequently, there is not the least danger of wound- 1 6 BACTERIA IN RELATION TO PLANT DISEASES. infections by bacteria, whose further progress in the plant is also impossible. Moreover, the result of an injection into the living plant of bacteria, even those pathogenic to animals and men, is easily predicted: There is no development in the intercellular spaces, or in extensive wound surfaces there is a wholly insignificant and soon extinguished multiplication. The experiments have exactly so concluded and need no further mention. Nevertheless new descriptions of plant diseases caused by bacteria keep springing up, and, truly, what worthless descriptions and what non-critical experi- ments. That in diseased plants bacteria are often found in great numbers is certain, but they have here always settled down only saprophytically (Metatroph) upon tissues broken down and destroyed by genuine fungi, and now, of course, help in the further work of destruction, and may also lend to the further progress of the disease a special aspect. But exclusive of other injuries, such as frost, animals, etc., the first attack upon the plant is brought about by fungi, not only in sickenings of uninjured plants, but also in case of wound-infections, which often are greatly extended by fungi and are converted into incurable injuries. From the bacillary gummosis of the grape vine to the scab of the potato, all so-called bacterial diseases of plants are of other origin, the bacteria being only saprophytic contaminations (metatrophe Verunreinigungen) , not self -conquering parasites (pp. 131,2). Wehmer's views, propounded in 1898, are not essentially different from the earlier views of Frank, or those of Alfred Fischer. They are sufficiently indicated by the following translation from his long paper on potato diseases : Thorough investigations of the bacterial rot are not so far to be found in literature. The few contributions which exist take up at random several specific cases and explain the problem with especial regard only to the bacteria, the tuber as a living organism being very little considered. The conclusions reached as to the "pathogenic" characters of the bacteria are indeed generally accepted to-day, but are not yet really sufficiently well grounded. Moreover, they are not perti- nent, as I shall endeavor to show. * * * For us, in this connection, the first sort of decay (primary rot) is of interest practically to the exclusion of the other, and will be somewhat fully considered in various directions. The question arises here especially whether we actually have to do in these cases with Schizomycetes capable of attacking living sound tissue. This supposition, all things considered, is to be definitely denied: There is manifestly no bacterial sickening of sound tubers; consequently in a literal sense also no "primary" rot — this is always secondary. In proof of which I have gathered together a pretty comprehensive mass of experiments. Wehmer also regrets that Frank should have abandoned his former safe position to accept the doctrine that bacteria can be independent causes of disease in plants. That other factors are truly primary and that the rot with its bacteria is only secondary has been already pointed out by Frank in opposition to earlier statements (Pflanzenkrankheiten, 2 Aufl. Bd. II, 1S96, p. 22). For leaving this standpoint there was really from first to last no reason; on the contrary its soundness was rather to be more exactly established by experiment. * * * Bacterial decay is only the last stage of the injury begun by environment, and even where it apparently attacks sound uninjured tubers one can demonstrate without difficulty that such is not the case. But of course it is easier from the simple discovery of the bacteria in decayed tissue to infer the pathogenic action of these organisms; in this way without trouble the numerous plant bacterial diseases, such as fill the literature of the day, are established. Smith (1899) criticised Fischer's statements and maintained the existence of bacterial diseases of plants, citing numerous experiments by various people in proof of his contention. Fischer (1899) answered Smith, maintaining that no one would doubt his having gone over the literature quite carefully; that all of the statements he had thus far examined rested, manifestly, on inexact observations or worse; that, for a book of the scope of his Vorlesungen, his statements were entirely sufficient; and, finally, that "there has not yet been published a single proof for bacterial plant diseases which meets all the requirements of exact bacteriology." Smith then made reply (1899 and 1901) to Fischer's criticisms, defending himself and other investigators, the validity of whose statements had been called in question, illus- trating three bacterial diseases by means of numerous heliotypes from photomicrographs. Since 1901 no one has ventured to question their existence. Peglion (1899) describes briefly the following as bacterial diseases: spot of hemp stems, mal nero of the vine, tubercle of the vine, tubercle of the olive, and blight of the mulberry. ATTITUDE OP PATHOLOGISTS AND BACTERIOLOGISTS. 1 7 Nadson's Russian paper, published in 1899, is a popular account drawn from the literature of the subject. He admits the existence of plant diseases due to bacteria. In his Text-book of Plant Diseases, published in 1899, Massee devotes 4 J pages out of 470 to bacterial diseases, mentioning bacteriosis of tomatoes, hyacinth bacteriosis, pink bacteriosis of wheat, black-rot of cabbage, olive tuberculosis, and the brown-rot of tomato, eggplant, and potato The author's perfectly safe standpoint is expressed in the following words: "At the present day numerous plant diseases are attributed to bacteria, some truly, others doubtfully so." Out of 1,350 odd species considered in the second volume of Migula's System (1900), 30 are at present of more or less interest to the plant pathologist. The action of the remainder, when introduced into living plants, is nil or unknown, mostly unknown. This book, like the Sylloge Schizomycetum of De Toni and Trevisan in Saccardo's Sylloge Fungorum, is devoted to a description of species rather than to a consideration of their pathogenicity. The last views of Hartig, which did not differ very materially from those held by him 20 years earlier, are sufficiently illustrated by the following quotations from his Lehrbuch, published in 1900: In fact, bacteria have been found thus far only in the tissue of those plants the cells of which are of a parenchymatic nature or are very thin-walled, as in bulbous and tuberous plants. The yellow disease of hyacinths is erroneously ascribed to onions and is further dis- cussed as follows: The bacteria do not attack sound well-ripened bulbs under normal conditions. Some sort of wounding is necessary, such as readily occurs during the lifting of the bulbs and their storage in another place, or else the bulbs are already attacked by fungi, among which a Ilyphomycctc in par- ticular is an almost constant accompaniment of the rot disease. In a damp situation the bacteria force their way into the wound and cause its decay. Bacteriosis of the potato is dismissed with 6 lines upon "wet-rot of the tubers." Pear- blight is discussed in 4 lines; sorghum-blight is mentioned in 6 lines. Of the olive tubercle it is said : In the olive forests gall-formations from the size of a pea to that of a walnut often occur in enormous numbers. These galls soon die and show in the crevices [of the dead galls!] large bacterial masses (figs. 203 and 204). But ■whether these are the cause of the gall-formations is not yet proved. [The italics are mine.] No other diseases are mentioned and this chapter closes with the following paragraph: Recently still other diseases have been ascribed to bacteria, without, however, furnishing the convincing proof by means of infection experiments that the Schizomycetes are the cause of the diseases. With these belong also the slime-flows of trees. In December 1900, Brwin F. Smith gave a lantern-slide lecture in Baltimore, Maryland, before a joint session of the Society for Plant Morphology and Physiology, and the Society of American Bacteriologists, illustrating fully three bacterial diseases from original photo- graphs and photomicrographs in his possession (Science, February 15, 1901). Weiss (1901) mentions, as of bacterial origin, wet-rot of the potato, white or yellow rot of hyacinth bulbs, beet-tip rot, and scab of potatoes. Less important bacterial diseases are probably — rot of onions, rose-red wheat kernels, * * * and the mosaic disease of tobacco. This author's standpoint is expressed in the following introductory remarks : The plant diseases due to Schizomycetes are of subordinate importance, while the bacterial diseases of animals and men are of the greatest importance. The bacterial diseases of the cultivated 1 8 BACTERIA IN RELATION TO PLANT DISEASES. plants in general presuppose previous disease due to fungi. They bear the name of rot diseases or bacterioses. Through the transfer of bacteria from diseased to sound plants the sickening of sound plants can be induced. In 1901, in his Disease in Plants, under "Exudations and Rotting," Marshall Ward, then leading English writer on plant pathology, has the following on bacterial diseases: In many of these cases bacteria abound in the putrefying miss, and some evidence exists for connecting these microbes causally with the disease in a few of the more thoroughly investigated cases, but in no case has this been sufficiently demonstrated; and considering the ease with which bacteria gain access via wounds caused by insects and fungi, as well as by other agents, the neces- sity for rigid proof must be insisted upon before we can accept such alleged examples of Bacteriosis. * * * Wet-rot of potatoes may be due to various fungi, and, in excess of water, to putrefactive bacteria. * * * The principal agent in the destruction of the tissues is Clostridium, an anaerobic bacillus which consumes the cell-walls but leaves the starch intact. * * * The rotting of bulbs, roots, etc., has been much discussed during the last few years in the pages of the Gardeners' Chronicle, Zeitschrift fur Pflanzenkh., and elsewhere. The principal references to Bacteriosis — the rot in which bacteria are stated to be the primary agent causing these and similar diseases — may be found in Massee, Diseases of Plants, pp. 338-342, and more fully in Russell, Bacteria in their Relation to Vegetable Tissue, Baltimore, 1S92; and in Migula, Kritische Uebersicht derjenigen Pflanzen- krankheiten, welche angeblich durch Bakterien verursacht werden, Semarang, 1892. The most convincing accounts, however, are since that date; see Smith, Pseudomonas campcstris, Cent. f. Bakt., B. Ill, 1897, p. 284, and Arthur and Bolley, Bacteriosis of Carnations, Purdue Uni- versity Agr. Expt. Station, 1896, vol. VII, p. 17. Woods has lately shown that this disease is due to Aphides only, the bacteria having nothing to do with the disease primarily, Siigmonose, Bull. 19, U. S. Dept. Agr., 1900: but it is necessary to bear in mind that actual penetration of the cell-walls from without must be proved, as De Bary proved it for germ-tubes of fungi, before the evidence that bacteria are truly parasitic in living plants can be called decisive. This is a difficult matter, but until it is settled we do not know whether these organisms are really parasitic in the sense that Phyla phthora is, or merely gain access by other means — I have traced them through dead fungus- hyphse — to the vessels, dead cell-walls, etc. The proof of infection via water pores and vessels is given for one species by Harding, Die Schwarze Faulnis des Kohls, etc., Cent. f. Bakt., Abt. II., B. vi., 1900, p. 305, with literature. * * * On Bacteriosis in Turnips, see Potter, Proc. R. S., 1901, vol. lxvii., p. 442. In Conn's Agricultural Bacteriology, published in 1901, 5 pages out of 419 are devoted to bacterial diseases among plants, but some of the specific statements will scarcely pass muster, e. g., those respecting the olive-tubercle. The author's standpoint is sufficiently illustrated by the following citation : It has been claimed that there is no likelihood that bacteria can live under such conditions and that bacterial diseases are, therefore, on a priori grounds, improbable or impossible. Even in very recent years this claim has been very vigorously supported, and disputes are still going on in the pages of bacteriological journals, in regard to the question of the existence of bacterial disease in plants. Almost to the very present day, it has been insisted that there is no demonstration that bacteria can produce disease in plants. Although this claim was legitimately urged a few years ago by conservative scientists, it can no longer be held in the light of recent experiments. In the last few years the evidence for such diseases has accumulated rapidly, and to-day the proof of the existence of bacterial plant diseases stands on identically the same basis as the proof of bacterial diseases among animals. In Neppi's Italian translation of Kirchner's book, published at Turin in 1901, various bacterial diseases are mentioned with the organisms said to be their cause. These are: The red disease of wheat kernels (M. tritici), internal rotting of corn-stalks and sorghum (B. termo), celery rot (B. a/>«), Bolley 's potato scab, Kellerman's sorghum disease (B. so rghi), gangrene of potato stems (B. caulivorus), brown-rot of potato (B. solanacearum) , beet disease (B. betae), hemp disease (B. cubonianus) , wilt of cucumbers, etc. (B. tracheiphilus), disease of vine stems (B. vitivorus), rot of grape bunches, soft rot of potatoes, onions, etc. (Clostridium butyricum), tumors on peach branches described by Cavara. Plates ix-xiv in Delacroix's Atlas (1901) are devoted to bacterial diseases of plants, viz., to tubercle of the olive and of the Aleppo pine, gummosis of vine, red disease of wheat, ATTITUDE OF PATHOLOGISTS AND BACTERIOLOGISTS. 1 9 yellow disease of beets, grease spot of beans, gangrene of potato, etc. The text is very- brief and closely follows Prillieux, except in case of the spot disease of beans, which is based on Delacroix's own work. In the second revised edition of his large Text Book, issued in 1901, Dr. Sternberg mentions bacterial plant diseases for the first time, devoting 7 pages to the subject, quoting exclusively from the publications of Smith and Waite. Chester's book (1901) includes descriptions of a few plant parasites, but does not venture any statements as to pathogenesis. In their large Treatise published in 1902, Miquel and Cambier devote 10 pages out of 1059 to the micro-organisms of plants. They mention 28 species as being of more or less interest in this connection. For general remarks by these authors on the uncertainties hanging over this subject, see citation in the preface to vol. I. Van Hall's Thesis (1902) maintains the existence of bacterial diseases as proved beyond dispute. He mentions many diseases, having a very good grasp of the literature; and admits the following 15 as of clearly-established bacterial origin: The black vein disease of crucifers due to Ps. campestris; the wilt of Solanaceae due to B. solanacearum, the wilt of cucurbits due to Bacillus tracheiphilus; the yellow disease of hyacinths due to Ps. hyacinthi; the bacterial gummosis of sugar-beets due to B. betae; the maize disease due to Ps. stcwarii; pear-blight due to Bacillus amylovorus; lilac-blight due to Ps. syringae; the olive tubercle due to B. oleae; the spot disease of beans due to Ps. phaseoli; potato-rot due to various bacteria (B. solaniperda, B. solanacearum, B. atrosepticus, etc.); carrot-rot due to B. carotovorus; turnip-rot due to Ps. destructans; iris-rot due to Ps. iridis and B. omnivorus: hyacinth-rot due to B. hyacinthi septicus. The original matter in this thesis will be discussed under the various diseases. In 1903, in the second edition of his Vorlesungen, Fischer repeats many of the inad- missible statements of his earlier edition, but, nevertheless, gives several pages to a review of a few bacterial disease of plants, dealing briefly with the rot of fleshy roots, potato-rot, the black-rot of cabbage, the mosaic disease of tobacco, and tree-cancers. Concerning the latter we have the following : Bacteria as the cause of cankers are unknown, for the Bacillus oleae which is said to cause the canker-like swellings of the olive-tree is no more legitimatized by pathological experiment than many other bacteria described as pathogenic for plants. It is fitting that these citations should end as they began, with Sorauer's Pflanzen- krankheiten. The second volume of the third edition devotes many pages to the subject of bacterial diseases of plants. This purely didactic review published in 1905, contains the best summary in any general treatise on plant diseases. About 70 bacterial diseases are considered. The statements in it, carefully as the literature has been gone over by Dr. Ivindau, show, however, perhaps as clearly as anything, the great need for a re-examination of the whole subject by some one experimentally familiar with it. Most of the conclusions I have cited in this chapter are to be regarded simply as ex cathedra judgments, or to put it somewhat differently they are to be regarded only as so many evidences respecting the ability of the particular writers to reason logically and arrive at sound conclusions from a maze of contradictory statements. In other words they are literary or legal judgments rather than scientific ones. A good judge must have not only a keen, well-balanced mind, but he must also know the case and the law. Very few of the writers I have cited appear to have had any extensive acquaintance with this class of diseases, or with the rules of evidence guiding in pathology, and those who have rendered adverse judgments seem to have had none whatever, i. e., they made few observa- tions and no experiments, or only some irrelevant ones. It is no wonder, therefore, that the insight of some of these writers has been much shrewder than that of others, or that most of them should have mingled fact and fancy in nearly equal portions in what they 20 BACTERIA IN RELATION TO PLANT DISEASES. have published. Nearly all of them have grossly neglected the experimental method. Under special diseases I shall have something to say about particular misconceptions, but, I have not felt called upon to point out all the errors strewn over the pages of these handbooks. Their name is legion and some of them have done service for a generation, having been handed down from one author to another, e. g., the " hyphomycete " that is almost always present in the yellow disease of the hyacinth. After a consideration of the treatment accorded to bacterial diseases of plants in these handbooks, the reader is well prepared to accept Migula's statement that: " Dieses Gebiet der Bakteriologie gegenwartig zu den verworrensten und wissenschaftlich am wenigsten durchgearbeiteten gehort" (System Bd. I, p. 312). In reality, however, the subject is not extraordinarily difficult, if it is approached by the experimental method, not more difficult than new researches in any other branch of science — to cut underbrush and break ground in any field of science is laborious work. This stage is now largely passed and there is considerable definite information on the subject as will appear from what follows. ATTITUDE OF PATHOLOGISTS AND BACTERIOLOGISTS. 21 LITERATURE. 1874. SorauER, Paul. Handbuch der Pflanzenkrank- heiten. Berlin, Wiegandt, Hemple und Parey. 1874, pp. iv, 406. Mit 20 Holzschnitten und 16 Tafcln in Farbendruek. 1878. De Jubainville, A. DArbois, et Vesque, J. Les maladies des plantes cultivees des arbres fruitiers et forestiers produites par le sol — L'atmosphere — Les parasites-vegetaux, etc. pp. viii, 328. Avec. 48 vignettes et 7 planches en couleur. Paris, J. Rothschild, Editeur, 13 rue des Saints-Peres, 1878. 1878. Winter, Georg. Die durch Pilze verursachten Krankheiten der Kulturgewachse. Leipzig, 1878, Karl Scholtze, pp. 151. 1879. Reinke J. und Berthold, G. Zersetzung der KartofTel. Berlin, 1879. 1880. Frank, A. B. Die Krankheiten der Pflanzen. Breslau, Eduard Trewendt. 1880, pp. vn, 844, mit 149 in den Text gedruckten Holz- schnitten. 1882. HarTig, Robert. Lehrbuch der Baumkrank- heiten. Berlin, Julius Springer, 1882, pp. VIII, 198, mit 186 figuren auf 11 Lithographirten Tafeln und 86 Holzschnitten. 1884. Smith, Worthington G. Diseases of field and garden crops, chiefly such as are caused by fungi. London, Macmillan & Co., 1884, pp. xxiv, 353 with 143 illustrations, drawn and engraved by the author. 1884. De Bary, A. Vergleichende Morphologie und Biologie der Pilze. Mycetozoen und Bacterien. Leipzig, Whilhelm Englemann, 18S4, pp. xvi. 558, mit 19S Holzschnitten. 1885. De Bary, A. Vorlesungen iiber Bacterien. Leipzig, Whilhelm Englemann, 1885, pp. VI, 146, mit 18 figuren in Holtzschnitt. 1886. Sorauer, Paul. Handbuch der Pflanzen- krankheiten. Zweite neubearbeitete Auflage. Zweite Theil, Die parasitaren Krankheiten. Berlin, Paul Parey, 1886, pp. xi, 456, mit 18 lithographirtin Tafeln und 21 Textabbildungen. 18S7. Viala, Pierre Les maladies de la vigne. Paris Deuxieme edition, ornee de 5 planches en chromo et 200 fig. dans le Text, pp. 462; A Delahaye et E. Lecrosnier, Libraires-editeurs; Montpellier, Camille Coulet. Libraire-editeur, 1887. 1887. De Bary, A. Comparative morphology and biology of the fungi, mycetozoa, and bacteria. English translation by Henry E. F. Garnsey, with 198 wood cuts. Oxford, at the Clarendon Press, 1887, pp. xviii, 525. Third part, Bac- teria or Sehizomycetes, pp. 454-490. 1888. Sorauer, Paul. Die Schaden der einheimischen Kulturpflanzen durch tierische und pflanzliche Schmarotzer sowie durch andere Einfliisse. Berlin, Paul Parey, 1888, pp. vii, 250. Die Spaltpilze (Sehizomycetes), pp. 152-154. 1889. DE Toni, J. B. and TrEvisan, V. Schizomy- cetaceae Naeg. in vol . vm of Saccardo's Sylloge F'ungorum, Padua, Dec. 20, 1889, p. 923. 1889. Laurent, EmilE. Sur l'existence de microbes dans les tissus des plantes superieures. Brus- sels, 1889, Bull, de la Soc. Royale de Bot. de Belgique, Tome xxvni, pp. 233-244 Also a separate. 1890. Kirchner, Oskar. Die Krankheiten und Beschadigungen unserer landwirtschaftlichen Kulturpflanzen, Stuttgart, 1890, pp. x, 637. Eugene Ulmer. 1890. Scribner, F. Lamson. Fungus Diseases of the grape and other plants and their treatment. Little Silver N. J.. 1890, pp. 136. J. T. Lovett Company. 1890. Ward, H. Marshall. Diseases of plants. Lon- don, Society for Promoting Christian Know- ledge, Northumberland Ave., W. C; New York: E. and J. B. Young & Co., 1890, 196 pp. 1 89 1. Comes, O. Crittogamia agraria. Vol. Unico, Napoli, Riccardo Marghieri di Gius, 77, Galleria Umberto I. 1891, pp. 600. Cap. xxx, Schizomiceti, pp. 493-530. 1892. Ludwig, Friederich. Lehrbuch der neideren Kryptogamen, pp. xv, 672, mit 13 fign. Stutt- gart, 1892, Ferdinand Enke, Die Bakterien als Urheber von Pflanzenkrankheiten, pp. 89-97. 1892. Loverdo, Jean. Les maladies cryptogamique des eereales. pp. 312, avec 35 fig. intercalees dans la Texte. Paris, Librairie J. B. Bailliere et Fils. 1892. 1. Sehizomycetes, pp. 15-26. 1892. MlGULA, W. Kritische Uebersicht derjenigen Pflanzenkrankheiten, wclche angeblich durch Bakterien verursacht werden. Mededeelingen van het Proefstation "Midden Java" te Klaten, 8 vo., pp. 18. Semarang, G. C. T. van Dorp & Co., 1892. There is also a Dutch edition. 1S92. RUSSELL, H. L. Bacteria in their relation to vegetable tissue. A dissertation presented to the Board of Univ. Studies of Johns Hopkins Univ., for the degree of Doctor of Phylosophy, pp. 41. Friedenwald Co., Baltimore. 1893. Viala, Pierre. Les maladies de la vigne. 3d edition, pp. vi, 595, 20 colored plates and 290 figures in the text. Montpellier, Camille Coulet, Libraire-editeur 1893, Chap, vm, Bacteries, pp. 414-417. 1894. Hartig, Robert. Text-book of the diseases of trees. English Translation by William Somer- ville, Macmillan & Co., London 1894, pp. xvi, 331. Revised and edited, with a preface, by H. Marshall Ward, Bacteria or Sehizomy- cetes, pp. 37-38. 1895. Prilliuex, Ed. Maladies des plantes agricoles et des arbes fruitier et forestiers causee par des parasites vegetaux. Paris, Librairie de Firmin-Didot et Cie, 56 rue Jacob, 1895, pp. xvi, 421 , Tome 1. 1895. Tubeuf, Karl, Freiherr von. Pflanzciikrank- heitendurchkryptogame Parasiten verursacht. pp. xii, 599, mit 306 in den Text gadruckten Abbildungen, Berlin, Julius Springer, 1895. 1895. HalliER, Ernst. Die Pestkrankheiten (Infec- tionskrankheiten) der Kulturgewachse, Stutt- gart, Erwin Nagele, 1895, pp. xiv, 144, mit 7 Tafeln. 1896. Frank, A. B. Die Krankheiten der Pflanzen. 2te Auflage. Zweiter Band. Die durchpfianz- liche Feinde hervorgerufenen Krankheiten. Breslau, Eduard Trewendt, 1896, pp. xi, 574, mit 96 in der text gedruckten Abbildungen. 2 Kap: Spaltpilze oder Bakterien, pp. 19-33. 22 BACTERIA IN RELATION TO PLANT DISEASES. 1896. 1897- 1897- 1897- 1897- 1898. 1899. 1899 1899- 1899. 1899. 1899. 1900. FluggE, C. Die Mikroorganismen. Mit besonderer Beriicksiehtigung der Aetiologie der Infektionskrankheiten. Dritte, vollig umgearbeitete Auflage bearbeitet von Dr. Frosch in Berlin, Dr. E. Gotschlich in Breslau, Dr. W. Kolle in Berlin, Dr. W. Kruse in Bonn, Prof. R. Pfeiffer in Berlin, Herausgegeben von Dr. C. Fliigge. Leipzig, Verlag von F. C. W. Vogel. Erster Theil, mit 57 Abbildungen im text, pp. xvi, 596, 1896. Zweiter Theil, mit 153 Abbildungen im text, 1896, pp. xxn, 751. Migula, W. System der Bakterien. Erster Band, Jena, Gustav Fischer, 1897. pp. vin, 368, mit 6 Tafeln. Tubeuf, Karl, FreihErr von. Diseases of plants induced by cryptogamic parasites. English translation by William G. Smith. 330 illustrations. Longmans Green & Co., London, New York, and Bombay. 1897, pp. xvi, 598. The Pathogenic Bacteria. Schi- zomycetes, pp. 530-539- Fischer, Alfred. Vorlesungen iiber Bakterien. Jena, Gustav Fischer, 1S97, pp. vin, 186, 29 figu., 2d Edition in 1903. Frank, A. B. Kampfbuch gegen die Schad- linge unserer Feldfriichte. Mit 46 Textab- bildungen und 20 Farbendrucktafeln. Berlin, Paul Parey, S. W. Hedemannstrasse 10, 1897, pp. vin, 308. Pages 144, 147, 175 and 200 refer to bacterial diseases. WehmER, C. Untersuchungen iiber Kartoffel- krankheiten in. Centralblatt f. Bakt., etc., 2 Abt., iv Bd., July 8, 1898, p. 540 et seq. Smith, Erwin F. Are there bacterial diseases of plants ? A consideration of some statements in Dr. Alfred Fischer's Vorlesungen iiber Bakterien. Centralbl. f. Bakt., etc., 2te Abt., Bd. v, 1899, pp. 271-278. Fischer, Alfred, in Leipzig. Die Bakterien- krankheiten der Pflanzen. Antwort an Herrn Dr. Erwin F. Smith. Centralbl. f. Bakt., 2te Abt., v Bd. 1899, pp. 279-287. Smith Erwin F. Dr. Alfred Fischer in the role of pathologist. Centralbl. f. Bakt., 2te Abt., V Bd., 1899, pp. 810-817. Peglion, ViTTOrio. Le malattie crittogamich delle piante coltivate. Biblioteca agraria Ottavi, vol. xxi, Casale (Carlo Cassone), 1899, pp. vii, 311. 2d Edition, 1905. Nadson.G. A. Bacteria as the cause of diseases of plants. 8°, 12 pp., St. Petersburg, 1899. [Russian but containing a brief French resume] A discourse delivered at a solemn sitting of the Imperial Society of Horticulture in May, 1899. Massee, George. A text-book of plant diseases caused by cryptogamic parasites. London, Duckworth & Co.; New York. The Macmillan Co., 1899, pp. xii, 458. Bacteria. PP- 338-342. Migula, W. System der Bakterien. Handbuch der Morphologic, Entwickelungsgeschichte und Systematik der Bakterien. Zweiter Band, PP x. 1068, mit 18 Tallin und 35 Abbildungen im Text. Jena. Gustav Fischer, 1900. 1900. 1900. 1901. 1 90 1. 1 90 1. 1901. 1 90 1. 1 90 1. 1 90 1. 1901. 1902. 1902 1905- Hartig, Robert. Lehrbueh der Pfianzen- krankheiten. Dritte Auflage. Berlin, Julius Springer, 1900, pp. ix, 324, mit 280 Text- abbildungen und einer Tafel in Farbendruck, iv. Schizomycetes (Spaltpilze), pp. 209-211. Wehmer, C. Zur Frage nach der Existenze Pflanzenpathogener Bakterien. Centralbl. f. Bakt., etc., 2te Abt., Bd. vi. No. 3, Feb. 3, 1900, pp. 88-89. Weiss, J. E. Kurzgefasstes Lehrbueh der Krankheiten und Beschadigungen unserer Kulturgewachse. Stuttgart, Eugeu Ulmer, 1901, pp. vin, 179. Smith, Erwin F. Entgegnung auf Alfred Fischer's "Antwort" in betreff der Existenz von durch Bakterien verursachten Pflanzen krankheiten. Zweiter Teil, mit XI Tafeln. Centralbl. f. Bakt., etc., 2te Abt., vii, Bd., 1901, pp. 88-100,128-139, 190-199. Ward, H. Marshall. Disease in plants. London, Macmillan & Co., Ltd., New York, The Macmillan Co., pp. xiv, 309. Kirchner, Oskar. Le malattie ed i guasti delle piante agrarie coltivate: manuale per l'avviamento alia indentificazione ed alia difesa ad uso degli agricoltori, degli ortolani, ecc. Versione italiana del Carlo Neppi rinnovata ed arricchita di copiosissime aggiunte ed annotazioni, 8 vo. vin. 873, pp. fig. 268. Torino (Unione tipogr. editrice), 1901. Conn, H. W. Agricultural bacteriology. A study of the relation of bacteria to agriculture with special reference to the bacteria in the soil, in water, in the dairy, in miscellaneous farm products and in plants and domestic aminals. Illustrated. Philadelphia, P. Black- iston's Son & Co., 1901, pp. 412. Delacroix, Georges. Atlas des conferences de pathologie vegetale professees a l'institut national agronomique, Paris. Libraire me- dicale et scientifique Jacques Lechevalier, Sept. (?), 1901. Sternberg, George M. A text-book of bac- teriology. Second revised edition. Illus- trated by heliotype and chromo-lithographic plates and two hundred engravings. New York, William Wood and Company, 1901, pp. xi, 708. Chester, Frederick D. A manual of deter- minative bacteriology. New York, The Macmillan Co.; London, Macmillan & Co., Ltd., 1901, pp. vi, 401. Miquel P. ET Cambier, R. Traite de bac- teriologie pure et applique a la Medicine et a l'hygiene. Paris, 1902, C. Naud, pp. XV, 1059- van Hall, C. J. J. Bijdragen tot de kennis der bakterieele plantenziekten. Academisch Proefschrift ter verkrijging van den graad van Docter in de Planten en Dierkunde aan de Universiteit van Amsterdam. Amster- dam, 1902, pp. x, 198. Lindau, G. Schizomycetes (Spaltpilze) in third edition of Sorauer's Handbuch der Pflanzenkrankheiten, 2 Bd., pp. 18-93 (includ- ing the nitrogen-gathering bacteria). GENERAL CONSIDERATIONS. ON THE SUPPOSED NORMAL OCCURRENCE OF BACTERIA IN PLANTS. We now believe that bacteria do not occur normally in the interior of sound plants. The case is quite different, however, with wounded plants or wilted ones. Frequently sapro- phytic bacteria have been found in such plants and occasionally mistaken for parasites. When bacteria are found in the tissues of plants in any great number we may assume that they are disturbing elements, and that if they continue to multiply the result to the host or some portion of it must be some considerable diminu- tion of vitality, even if no specific disease supervenes. Compen- sations due to symbiosis are not here under consideration. The former great uncertainty as to the life-history and habitat of bacteria led to many speculations respecting their normal occur- rence in the interior of both plants and animals. The belief that they might occur normally in the interior of plants arose from the inexact observations and experiments of various early workers, notably Bechamp and Hallier. The dispute continued for a number of years but was finally settled in the negative. Bechamp went so far as to maintain that his microzymes were always present in plants and animals, were in fact the simplest components of the tissues and led an independent life after their death and disintegration. Hallier believed that the protoplasmic granules of fungi were converted into bacteria capable of an inde- pendent existence.* Fremy maintained the existence of hemi- organized bodies in the juice of fermentable substances which bodies were converted into yeasts. Trecul believed in similar transformations: granules of organic matter became motile bac- teria. In a later time it was still believed by some that bacteria could be cultivated out of the sound interior of plants and animals and were normally present therein, and by others that they arose spontaneously in all sorts of organic substances. That the organic must have developed from the inorganic during some period in the history of the earth seems probable, but we must look elsewhere than to Bechamp and Hallier for evidence. The amount of ignorance and credulity respecting micro- organisms prevalent in the middle of the last century seems aston- ishing in the light of our present knowledge. It is, however, the history of all subjects hedged about by difficulties. The beginnings are always foggy. Pasteur appears to have been the first to show that the sound interior of plants is free from micro-organisms. He experimented on grape-berries, taking some of the juice from the interior under such conditions as to preclude the entrance of surface bacteria and placing it in sterile must which remained sterile, while flasks treated to the washings from the surface of the grapes invariably developed growths of some sort. In 1 879-1 880, Chamberland working in Pasteur's laboratory showed that beans taken directly from the interior of their pods were free from bacteria, i. e., did not contaminate culture-media when put into them (see fig. 2). *Even in very recent times we have similar views occasionally coming into print, c, g., Dunbar's Zur Frage der Stellung der Bakterien, Hefen und Schimmelpilze im System (1907), in which it is maintained that bacteria, yeasts and fungi, are the product of algal cells. fFio. 2. — Peas taken from pods less than 18 hours after picking and placed on sterile nutrient gelatin where they have sprouted and grown entirely free from the presence of bacteria. Photographed Oct. 3, 1908, from sample tubes sent the writer by Mrs. A. W. Bitting. Two-thirds natural size. 23 Fig. 2.f 24 BACTERIA IN RELATION TO PLANT DISEASES. In 1882, Prof. T. J. Burrill described a Micrococcus toxicatus as the essential virus of the Poison Ivy (Rhus toxicodendron). It is now believed that the poison of Rhus is not due to a micro-organism but to a chemical substance: A non-volatile oil fPfaff) ; a gluco- side (Syme). In 1884, Jorisson ascribed the formation of diastase in the higher plants to the pres- ence of bacteria in the tissues. In the following year, however, Laurent pointed out errors in Jorisson's work. Laurent himself reached the conclusion that there are no bacteria normally present in living plants. Galippe (1887) examined the inner tissues of many kinds of vegetables from the vicinity of Paris and found bacteria so constantly present, except in garlic, that the reader is at once led to suspect some serious error in his methods of work. No quantitative tests were made of the number of bacteria per gram of tissue or per cubic centimeter of juice, and it is possible that those which appeared so regularly in the author's tubes are to be ascribed either to the use of wounded or wilted vegetables, to air contaminations occurring at the time the cultures were made, or to imperfectly sterilized media, especially as his results are not in accord with those of Fernbach. Galippe first experimented with vegetables grown on a soil supersaturated with sewage- bacteria, i. e., with those grown in the municipal experiment gardens on the plain of Gennevilliers, near Paris. The vegetables tested were exposed to the Bunsen flame until the surface was car- bonized. They were cut with a hot, sterile knife. Each surface of the section was then flamed. Finally, by means of a hot sterile knife (heated above ioo° C.) "I detached fragments of the vegetable which were put directly into the culture-fluid, choosing the most central parts. I strove as far as possible to remove sources of error." His culture-media were: (1) ordinary bouillon; (2) peptonized bouillon with sugar; (3) same, neutralized; (4) saliva peptonized and sugared; (5) same, neutralized; (6) broths from the vegetables experimented upon; (7) same, with peptone and sugar. A long series of experiments was instituted, the culture-media being inoculated from carrot, onion, celery, parsnip, turnip, potato, beet, lettuce, salsify, leek, cabbage, Brussels sprouts, Jerusalem artichoke, garlic. All gave positive results except garlic. The juice of the latter is said to sterilize culture-fluids. A second series of experiments was made, using vegetables taken from the market- gardens about Paris. Concerning their origin only one thing was ascertained carefully, namely, they did not come from Gennevilliers. These are styled normal vegetables. The author obtained substantially the same results with these vegetables, viz., the clouding of most of his culture-media. His general conclusions, therefore, are: (1) the micro-organisms of the soil can pene- trate into the tissues of the vegetables with which they are in contact, the mechanism of this penetration remaining to be elucidated; (2) the number of the micro-organisms con- tained in the vegetables seems to vary with the number in the dung used. There is no statement in the paper as to how the culture-media was sterilized ; where the inoculations were made, c. g., whether in a clean room, free from air-currents; nor as to whether check-tubes of the various culture-media were generally held for comparison and remained sterile. All we are told is that some of the many inoculated tubes remained sterile. Concerning these results, as already hinted, conclusions quite different from those of the author might be drawn. Fernbach carefully repeated the experiments of Galippe and published a paper on the subject in 1888. He found all of Galippe's conclusions erroneous. In all cases, except that of the tomato, pieces of the interior tissue were removed and thrown into the culture- media. In all, 98 different specimens were examined. Of the 555 inoculations only 35, i. c., 6.3 per cent, developed any growths. This number is considered about the minimum of necessary contaminations, arising from air-currents and other imperfect conditions under ON THE SUPPOSED NORMAL OCCURRENCE OF BACTERIA IN PLANTS. 25 which the work was carried on. An examination of the contents of the fertile tubes also confirmed this view. Fernbach took vegetables just as they came into the market without inquiring where grown, assuming that any soil adapted to vegetables was rich enough in micro-organisms so that they would penetrate into vegetables if this was physiologically possible. He tested potatoes, carrots, turnips, beets, and tomatoes, judging it not worth while to try additional sorts, since experiments on all of the above showed the conclusions of M. Galippe to be erroneous. For culture-media he used neutral veal-bouillon and sugared turnip-water which was very slightly acid. These media were in test-tubes and Pasteur flasks. The surface of the vegetables was first heated to light carbonization by means of a thermocautery. The tomato juice was aspirated out by means of a pipette drawn out at one end and plugged with cotton at the other. The other vegetables were punched with brass cork-borers, cotton-plugged above and sterilized inside of cotton-plugged test-tubes at 1650 C. I obtained with these tubes cylinders of vegetable tissues which I pushed out little by little, and which were sowed immediately by sectioning them with a flamed scalpel. I thus introduced into each tube a volume of tissue varying from 0.5 to 1 cc. A summary of Fernbach's results is given in the following table: Number of sow- Vegetable Tomatoes. Turnips . . Carrots. . . Beets. . . . Potatoes. . Total . Number of vege- Number of tables sampled. sowings f ertil 26 52 2 36 199 19 13 IOI 4 12 I03 10 1 1 IOO 0 98 555 35 We see that there are a certain number of fertile sowings, 63 per cent. It could scarcely be otherwise. As a matter of fact, in experiments so delicate there are several sources of error which it is impossible to remove absolutely. The most important is that which arises from germs of the air. These vary greatly in number but are always abundant in a laboratory where there are goings and comings and where one is constantly exposed to currents of air. The practice of filling Pasteur flasks almost daily shows me that, in the laboratory where I have made my experiments (Sorbonne), out of 100 flasks rilled there are always 4 or 5 which show growth. The organisms which developed in 6.3 per cent of the cultures were of various sorts — bacilli, micrococci, molds, etc., in general each tube being occupied by a single sort. A source of error also to be considered is the possible penetration of germs into vegetable tissues as the result of insect depredations, or of injuries due to digging or transportation. We conclude, therefore, that normal vegetable tissues constitute a perfect filter for microbes and that they can be invaded by them only as a result of causes wholly accidental (p. 570). In 1888 A. di Vestea repeated Galippe's experiments. This author experimented under much the same conditions as Galippe, i. c, with plants obtained from the market-gardens around Naples and grown on the lowlands where the filth of the city is dumped. He tested especially a variety of lettuce called Roman lettuce. In sampling he made use of a special glass-punch, consisting of a thin tube through which slid a thick-walled tube longer than the first, closed with cotton at the free end, and sliding in a plug of cotton. The whole apparatus was sterilized at 1500 C, after the lower half had been introduced for protection into a test-tube plugged with cotton. The plant to be tested was then cut with a flamed knife. The punch was removed from the test-tube and the cutting end of the outer thin-walled tube was then plunged into the cut surface of the vegetable and a piece of tissue removed. The apparatus was then replaced in the sterile test-tube, the tissue was pushed out of the cutting tube by means of the inner tube, which was then used to crush it in the bottom of the test-tube. Finally, some of the liquid which resulted from the crushing was sucked 26 BACTERIA IN RELATION TO PLANT DISEASES. up into the inner tube and from this transferred to the culture-medium. At the same time sterile bouillon was put on the crushed tissues, thus giving two series of cultures. The culture-medium was sterilized veal bouillon. Numerous experiments led di Vestea to the following conclusions: (i) Cultures made from plants which he gathered himself or which were brought to the laboratory (Lab. de la Clinique Cantani, Naples) fresh from the country remained regularly sterile whether in vacuo or exposed to the air. (2) If the same vegetables were left exposed to the air for a day or more, a new sample very often gave positive results. (3) Finally, whenever the author worked with vegetables brought in from the market he always obtained fertile cultures. "This last result is to be explained, I believe," says di Vestea, "especially by the fact that the gardeners and produce dealers water their vegetables, to keep them fresh, with water which is usually swarming with bacteria." In 1888, Dr. Bernheim claimed to find bacteria in the interior of seeds of cereals, but Lehmann, in whose laboratory he had worked, discredited his views, stating that Bernheim's studies were very incomplete researches made under direction and published during a vaca- tion period without his teacher's knowledge or consent. The same year in discussing Bernheim's paper Dr. Buchner says that he (Buchner) obtained negative results from histestsof normal vegetable tissue for the presence of bacteria. In 1889, Dr. Lehmann, discussing Bernheim's work said: "Normal plant seeds are germ-free." In 43 gelatin-roll-cultures made out of at least 800 fragments from the interior of beans, chestnuts, maize, peas, and almonds, he obtained only 6 bacterial colonies, which undoubtedly came from the air. At the conclusion of his paper in the Archiv fur Hygiene he says: "I summarize the second part of my work thus: That Buchner certainly is right in considering Dr. Bernheim's pellicles for non-living formations — but that in opposition to Buchner's view they did not consist of fat, but of salts." Laurent experimenting on several occasions with various plants, viz., seeds of barley, maize, and lupin, tubers of potato, bulbs of onions, roots of carrot and chicory, and the fleshy tissues of Cereus, Agave, and Carica, obtained the same results as Fernbach. The interior of these plant parts was found to be free from bacteria. In 1890, Laurent showed that the sap flowing from the cut surface of healthy vine- stems contains no bacteria, i. c, none of these organisms were taken up by the roots. Eleven tests were made on as many young potted grape-vines, placed for a few weeks in winter in a hothouse. In each case the shoot was flamed, cut, the cut end renamed, and then plunged into a tube plugged with cotton and containing 10 cc. of sterile veal broth. In 24 hours, 5 to 10 cc. of sap had oozed out and mingled with the culture-fluid. The branches were then removed and the tubes incubated at 300 C. In a week's time only 1 tube developed bacteria. Some of the fluids were neutral, others were slightly acid, and the rest were slightly alkaline. In 1 89 1, Kramer found no bacteria in the interior of sound potato-tubers (for a long review in English see American Naturalist, 1897, pp. 123-138). In 1892, Russell reached the same conclusion respecting several plants. In more recent years Hiltner has reached the same conclusion respecting the interior of sound seeds. He found the tissues of those he tested always free from bacteria. The writer has sometimes found bacteria in fleshy roots supposed to be normal, and the surface of which had been properly sterilized, but these had been dug for some time. The parenchyma of healthy plants is always or almost always free from bacteria. Prob- ably the vascular system, especially of some parts of the roots, frequently contains bacteria, and certainly they must be present to some extent in those plants in which tyloses are abundant, if the latter are due to the stimulus of bacterial products as believed by the writer (see fig. 30 and page 91). ON THE SUPPOSED NORMAL OCCURRENCE OF BACTERIA IN PLANTS. 27 LITERATURE. 1865. Trecul, A. Matiere amylacee et cryptogames amyliferes dans les vaisseaux du latex de plusieurs Apocynees. Compt. Rend, des Se. de LAcad. des Sciences, Paris, 1865, T.lxi, pp. 156-160. See also T. lxi, pp. 432-436, and T. Lxv 1867, pp. 513 to 521. Trecul reported finding Amylobacter in the pitch- celts of fig and in the cortex of Sambucus, Solanaceae, and Crasulaceae. 1868. NylandER. Animadv. circa historiam Arnylo- bacteriaceam. Flora, 1868. 1876. Pasteur, Louis. Etudes sur la biere, etc. Paris, Gauthier Villars, 1876, pp. 54-57. 1879. Chamberland.Charles-Edouard. Reeherches sur l'origine et sur le developpement, des organismes microscopiques. Ann. de l'Ecole Normal super., 1879, Paris. 1880. Chamberland, Charles-Edouard. These, Paris, 1880, p. 35. 1882. Burrill. T. J. The bacteria, 1882, p. 42. See also some vegetable poisons, Tr. Am. Asso. Adv. Sci., 1882, p. 515, and Am. Mo. Micro. Jour., Oct., 1882, p. 192. 1884. JorissEn, A. Les proprietes reductrices des graines et la formation de la diastase. Bull. d TAcad. Royale de Belgique, T. vm, 3rd Ser., PP- 550-555- 1884. Ralph, S. On the occurrence of bacteria in living plants. Trans. Roy. Soc, Victoria, vol. xx, 1884. 1885. Laurent, E. Sur la pretendue origine bac- terienne de la diastase. Bull, de lAcad. Royal de Belgique a Bruxelles, 3 serie, Tome x, No. 7, 1885, pp. 38-57. Also a separate, 22 pp. 1885. Duclaux, E. Sur la germination dans un sol riche en matieres organiques, mais exempt de microbes. Compt. rend. hebd. des seances de l'Acad. des Sciences, Paris,5 Jan., Tome C, 1885, pp. 66-68. 1885. Brasse, Leon. Un moyen de debarrasser les graines des germes de microbes adherents a leur surface. Compt. Rend, et Mem. Soc. de Biologie, Paris, 1885. T. 37, 8 Ser., pp.196, 197. 1886. Frankhauser . Der Bund, No. 26, Berne. Not seen. 1887. Galippe, V. Note sur la presence de micro- organismes dans les tissus vegetaux. Compt. Rend. hebd. de la Soc. de Biologie, Paris, 1887, pp. 410-416. See also Jour, des con- naissances medicales, Paris, 30 Juin, 1887. 1888. Fernbach, A. De l'absence des microbes dans les tissus vegetaux. Ann. de l'lnstitut Pasteur, 2 annee, Paris, 1888, pp. 567-570. A review of Galippe's work. 1888. di Vestea, A. De l'absence des microbes dans les tissus vegetaux. Ann. de l'lnst Pas- teur, 2 annee, Paris, 1888, pp. 670-671. 1888. Bernheim, Hugo. Die parasitaren Bacterien der Cerealien. Miinchner med. Wochen- schrift, 1888, pp. 743-745. 767-770. 1888. Buchner, - — . Notiz betreffend die Frage des Vorkommens von Bacterien in normalen Pflanzengewebe. Muenchner med. Wochen- schrift, 1888, No. 52, pp. 906-907. 1889. Lehmann, K. B. Erklarung in Betreff der Arbeit von Herrn Dr. Hugo Bernheim: "Die parasitaren Bakterien der Cerealien. ' ' Muench . med. Wochenschrift, 1889, p. no. 1889. Lehmann, K. B. Erklarung in Betreff der Arbeit von Herrn Dr. Hugo Bernheim: "Die parasitaren Bakterien der Cerealien," Nebst Weitern eigenen Versuehen. Archiv f. Hygiene, Bd. ix, 1889, pp. 350-361. 1890. Fazio, Eugenic I microorganismineivegetali usati freschi neH'alimentazione. Rivista inter- nazionale d'igiene, Anno 1, 1890, No. 1-3, pp. 15—23, 99-107, 162—167. Reviewed in Cen- tralbl. f. Bakt, etc., 4 Jahrg., 1890, Bd. vn, P- 798. 1890. CorneillE, A. V. and Babes, V. Les bac teries 3 ed., Paris, 1890, Felix Alcan, p. 20. They assert that Duclaux has shown that the germination of plants is impossible without the presence of bacteria, but this is incorrect. Duclaux showed only that when peas and beans were germinated in milk they behaved exactly as when germinated in pure water. In a comment on Duclaux's paper, Pasteur hazarded the supposition that animal life required the presence of bacteria, and hence perhaps the origin of the confusion. 1890. Brown, Horace T., and Morris, G. Harris. Researches on the germination of some of the Gramineae. Jour. Chem. Soc, vol. 57, 1890, pp. 458-528 (see especially p. 512). Healthy seeds contain no bacteria. 1890. Laurent, EmilE. Sur la reduction des nitrates par Levurede Biere et parquelqueMoisissures. Bull, de l'acad. Royale de Belgique, Tome xx, 1890, pp. 309 to 317. Also Bull. Soc. Roy. de Botan. de Belgique, T. 28, p. 233. Healthy seeds contain no bacteria. 1890. Laurent, EmilE. Experiences sur l'absence de bacteries, dans les vaisseaux des plantes. Bull, de l'Academie Royal des Sciences, des Lettres et Beaux-Arts de Belgique, 1890, 3me series, T. 19, pp. 468-471. 1900. Weil, R. Die Entstehung des Solanins in den Kartoffeln als Product bakterieller Einwirk- ung. Pharmaceut, Ztz., 1900, No. 93, p. 901. 1 89 1. Kramer, Ernst. BakteriologischeUntersuchun- gen ueber die Nassfaule der Kartoffelknollen. Oesterreichisches landw. Centralb. I, Heft 1, 1891. 1892. Russell, H. L. Bacteria in their relation to vegetable tissue. Thesis, Johns Hopkins University, 1892, 8vo., pp. 41. 1892. Wohl, A. Ueber die Bildung des Lupulins und den Micrococcus Hiimuli Launensis. Oesterr. landw. Centralb. 1892, from Allgem. Brauer und Hopfenzeitung, 1892, No. 47. 1902, Ellrodt, Ueber das Eindringen von Bakterien in Pflanzen. Centralb. f. Bakt., 2 Abt. Bd. ix, 1902, p. 639. 1893. Dixon, Henry H. On the germination of seeds in the absence of bacteria. [Read Dec. 21, 1892.] Scientific Trans, of the Royal Dublin Society, vol. v (series 11), part 1, May, 1893, Dublin, pp. 1 to 4, 1 fig. Also a separate 4 pages. Experiments undertaken as a result of the erroneous state- ment of Corneille and Babes. 1894. Kochs, — — . Giebt es ein Zellenleben ohne Mikroorganismen? Biologisches Centralbl. Bd. xvi, 1894, No. 14. Rev. in Centralbl. f. Bakt., etc., Bd. xvi, 1894, pp. 633-634. The fact that plants can be grown to maturity from sterile seed in sterile vessels shows that plants can develop independently of bacteria. BACTERIA ON THE SURFACE OF PLANTS. OBSTACLES TO THEIR ENTRANCE INTO PLANTS— OBSTACLES TO THEIR MULTIPLICATION IN PLANTS. The young vegetative parts of plants are covered by the epidermis, a skin of close-set cells interrupted here and there by stomata, but not easily permeable to water. This epidermis when unbroken offers great resistance to the entrance of harmful micro-organisms. Its surface is often reinforced by cutin a still more resistant layer which is sometimes developed to a very marked degree. Some plants also turn aside water and whatever that may contain, by a waxy bloom, e. g., the cabbage. A dense layer of soft hairs may have the same function, as on the surface of a peach fruit or the stem of a com- posite. These devices render it difficult to wet the actual epidermis lying under the cutin, wax, or lanugo. In older parts the epidermis is displaced by cork a many-layered, close-celled, very imper- vious, very indestructible covering which keeps out fluids and also keeps them in so perfectly that the special kind found on the cork-oak is used by civilized man everywhere for this very purpose. Its use to the plant is obvious. Only through wounds or through certain natural open- ings, known as lenticels, can bacteria pass this very perfect barrier (see fig. 3). The plant then is naturally very well pro- tected against bacteria, except as I will point out in a following chapter. The surface of plants, as we shall see a little later, is often covered by a variety of bacteria and some of these are likely to find their way into the tissues whenever they are wounded, but if they do gain an entrance either through wounds or through some natural opening, they can in the vast majority of cases take no advantage of it because they are saprophytes, i. e., they are not adapted to the conditions present in the plant. And even if they happen to be parasitically in- clined they are often debarred from further pro- gress by the fact that the wounded plant does not contain enough water for their needs. In such cases they make either no growth or such a very slow growth that the plant has time to erect a physical barrier to further progress in the form of a cork-layer cutting out the affected tissues from the body of the plant. This happens very frequently in potato-tubers attacked by various soft-rots. It *I'"io. 3. — Young shoots of mulberry inoculated with Bant, mori, showing cirri of bacterial slime oozing to surface through lenticels. Inoculated by needle-pricks Jan. 4, 1909. Photographed (enlarged) Jan. 9. The dark stripe is a sunken diseased area. 28 Fig. V BACTERIA ON THE SURFACE OF PLANTS. 29 occurs often also in leaf-spot diseases. It is not always a perfect protection, however, since in some weak portion the parasite may break through the barrier and form a new center of infection. This I have observed many times. The chemical obstacles are equally interesting, although our knowledge of many of them is far from exact. There can be no doubt, however, as to their existence. There are probably few substances in plants which bacteria can not be educated to tolerate in test-tube cultures, especially when the inoculations are very copious and the doses of the antiseptic are small at first. The conditions in nature, however, are somewhat different. Especially are the number of bacteria accidentally introduced into the plant undoubtedly, in most cases, vastly fewer than we introduce on our needles or with the hypodermic syringe. It is a common laboratory experience that culture-media which will not cloud when inoculated with minute doses of bacteria will do so when inoculated with larger quantities of the same bacteria. I found the acid parenchyma juice of cucumbers exerted a decidedly retarding influence on Bacillus tracheiphilus and the same was true of hyacinth juice on Bad. hyacinthi; with Bad. stewarti I could not get any growth in a very acid tomato-juice. In one instance a slight reduction in the acidity of a potato juice by the use of sodium hydrate enabled an organism to grow readily — a very slight reduction in proportion to the total acid present. Many bacteria are quite sensitive to the organic acids occurring in plants, e. g., malic acid, citric acid, tartaric acid, and there can be little doubt that these exert a protective influ- ence. I have never found any bacterium that would grow in pieplant juice* or in orange juice. Galippe states that he could not obtain any growth in garlic juice. I made one trial with the same result. Whole families of plants contain bitter or aromatic substances, and some of these must undoubtedly be regarded as protective substances, indeed, some of them as salicin, methyl salicylate, thymol, menthol, camphor, cinnamon oil, mustard oil, are distinctly antiseptic. Other plants contain alkaloids, glucosides, etc., which may be assumed to be more or less protective. Tannin is a substance very widely distributed in plants ; we do not know its functions very well, but, as Tschirch has suggested with reference to germinating seeds, one of them may be that of antisepsis. It might inhibit bacterial development either directly or by oxidation into more active colorless or brown compounds. In this connection see inter- esting observations by Hiltner on a germicidal or inhibiting substance extruded by sprout- ing seeds of legumes (p. 124). Anthocyan is thought by some to be antiseptic. Appel has pointed out that potato-tubers, the flesh of which turns a reddish brown on cutting (oxidation of tannin compounds) , are much more resistant to Bacillus phytoph- thorus than those which remain white. The latter were rotted easily on inoculation, the former either not at all, or very slowly. In an experiment by the writer made in test- tubes using this same organism, growth was certainly more rapid in potato-juice steamed at once and consequently remaining pale, than in a portion of the same juice that was allowed to oxidize 24 hours before steaming. Dr. O. Iyoew states that he found bacteria absent or rare in brown, curing tobacco, and the writer confirmed microscopically some of his results. In the samples shown to me bacteria were certainly not abundant. The list of weak to moderately strong antiseptics derived from plants is a long one, as every bacteriologist knows. We are accustomed to look on them, for the most part, with little favor because they do not accomplish all we desire, i. 2 months meanwhile having elapsed. Natural size. Fig. 14.* 62 BACTERIA IN RELATION TO PLANT DISEASES. stomata. The disease has been obtained a number of times during the last seven years by placing the bacteria in water and spraying this upon the plants. The leaf-serratures also blacken in this disease, and here infection probably occurs through the groups of water pores situated on their apex. The genuine bacterial spot of carnations is a fourth disease of this type. It was pro- duced a number of times by Lloyd Tenny, one of my assistants, who kept the plants moist under bell-jars for a day or two so as to get a deposit of water drops on the foliage, and then sprayed upon the plants sterile water inoculated with pure cultures of the organism. Infec- tions are visible within a few days. They always begin in the substomatic chamber, and Petri-dish poured plates made from the interior of the spots on several different occasions have yielded pure cultures of the parasite, which when reinoculated by spraying has again produced the disease.* The spot disease of beans, caused by Bacterium phaseoli, is another example of stomatal infection. Serial sections through very young spots demonstrated this to me beyond reasonable doubt (fig. 15). Moreover, the disease was subsequently produced experimentally under my direction by Deane B. Swingle. The spots appeared in large numbers in about 6 days as the result of spraying experiments and the earliest bacterial nidus was in the substomatic chamber. This manner of entrance explains, I be- lieve, the fact once observed by Halsted that nine-tenths of the spots in this dis- ease were on the western side of the pods — that is, as I interpret the pheno- menon, on those parts where rain drops or dew drops would persist longest and thus give most opportunity for infection. The writer has observed the same thing in connection with the black spot of the plum. The spot disease of broom-corn also arises by stomatal infection and has been so produced by the writer, using pure cultures but not of Bacillus sorghi. For early stages see figs. 16 and 17. The angular leaf-spot of cotton (vol. I, fig. 80) and the brown spot of Pelargoniums are two other bacterial diseases in which infection commonly begins in the substomatic chamber. The writer has studied both of these diseases in serial sections and has repro- duced the first by simply spraying the bacteria upon the plants (fig. 18). The Pelargonium leaf-spot was also so reproduced in 1906 by John R. Johnston, one of the assistants in my laboratory, using pure cultures. j^ ]u/&. /^-i-. #^4gJ I - 1 / A '■J Jk ■ K-^*c&x0!0<% / r»— ^ -'=5' ^^^j^ ■^^rv \-m^^*® ^ 1 v»-V^ W VJi^jA .£L-- /\ sections to either side in the series. Xylem and phloem free from infection. In upper part of section bacteria lie over ion I three cells, sharply delimited on one side by cell /, and on other side by cells ee. At a deeper focus these three cells are free from bacteria, except for a few lying between cell-walls. Similar collections of bacteria along cell-walls may be seen in extreme upper part of picture. Contents of epidermis cells omitted. Slide 4 [_• Ao, upper row. sixth section from left. Paraffin embedded section, stained with carbol fuchsin. Drawn with a Zeiss 3 mm. 1 .40 n. a. oil immersion objective, No. 1 2 compensation ocular and Abbe camera. PERIOD OF INCUBATION. 65 The time between inoculation and visible disease may be as short as 24 to 48 hours, or as long as 3 or 4 weeks. It varies not only with different organisms, but with the same organism under different conditions. In some species long cultivation on artificial media destroys or greatly weakens the ability of the organism to attack tissues. In other cases a similar reduction of virulence occurs within the host. Experimenting with juicy suscepti- ble plants and such organisms as Bacillus carotovorus, B. olcraceae, B. aroideae, B. melonis, or B. hyacinthi (Heinz) , the result of a single needle-prick is often visible in 24 hours, and by the end of the third day the necrosis of tissue is often quite extensive. With the same organisms and in the same host-plants, but in rather woody or somewhat dry spongy tissues, the progress of the disease is slow, and after a slight development it may stop altogether. Potter states that his Ps. dcstructaus inoculated into turnips caused very distinct signs of the disease in 24 hours. For their rapid development most of the soft-rot organisms require tissues full of water. With Bacillus phytophthorus, using virulent cultures, susceptible varieties of potatoes, and optimum temperatures, and inoculat- ing by needle-pricks, rot is always visible in 24 to 48 hours, and the entire tuber may be rotted in a week's time, even in dry air. In pear-blight the blackening of the shoots usually occurs in from 3 to 10 days after inoculation by needle-punctures from fresh agar cultures (vol. I, plate 28), but may sometimes be delayed 23 days (Arthur). Much depends on the weather and on the immaturity of the shoots. The pear-blight develops soonest in moist, warm weather and in rapidly growing shoots. In blossom- infections there is a distinct browning in the nectaries in 48 hours, and on the third or fourth day the whole flower collapses and is blackened, together with its pedicel, which has become infected. In the writer's experiments with Bacterium solanacearum in 1895 and 1896, blight appeared in young shoots of the potato and tomato in about 4 to 6 days when they were inoculated by needle-pricks from young cultures. On the contrary, in an old and woody tomato plant wilt did not become general until 8 or 9 weeks after the punctures, but then the organism was found to have multiplied enormously in the vascular system, extending to a distance of several feet from the pricked part of the stem. In large tomato plants in a field in South Carolina, during wet weather in July 1895, direct infections by needle-stab induced plain signs of the disease only after many days. Similar tardy results were obtained in Washington in a hothouse in 1909. When Colorado potato beetles were used as carriers of the organism the first signs of the disease in potato appeared in 7 to 9 days from the time the plants were bitten. In more recent experiments with this organism, especially some tests made in 1904 with an extremely virulent strain obtained from a blighting potato, wilt appeared in young tomatoes in 48 hours after inoculation by needle-pricks, and the entire plant was destroyed in 6 days (see vol. I, plate 26). These plants were in pots in the hot- house, were 1 to 1.5 feet high, and were growing rapidly when the stems were pricked. Fig. 18.' *Fig. 18. — Angular leaf-spot on Rivers cotton. Inoculated by spraying Jan. 21-23, '9°5- Photographed March 15. Natural size. Spots in water-soaked stage. 66 BACTERIA IN RELATION TO PLANT DISEASES. With Bacillus tracheiphilus, inoculating from young agar-cultures, potato-cultures, or bouillon-cultures into the leaf-blades of cucumbers and other very susceptible plants by means of needle-pricks, the writer has seldom been able to obtain signs of the disease in less than 3 or 4 days. Usually the first signs (wilt and change of color in the vicinity of the punctures) were visible in 5 to 7 days. Occasionally the wilt did not appear until after the tenth day, once it was not manifest until after 21 days, and once not till after 30 days. In leaves of hyacinths inoculated by needle-pricks with Bacterium hyacinthi signs of the disease appeared in from 1 to 3 weeks. In one experiment in cabbage leaves inoculated by way of the water-pores with Bac- terium campestre, the serratures of the leaves showed a distinct blackening within 4 to 6 days, but a period of 3 weeks elapsed before there was any visible spread of the disease down the leaf, i. e., away from the leaf- serratures. In another experiment there was a distinct blackening in the region of the water-pores in 6 days. In Russell's water-pore infections on cabbage, signs appeared in 2 to 3 weeks. Stem-inoculations, i. e., needle-punctures without injection, cause signs of the disease in the nearest leaves, viz., yellowing, flabbiness, and brown veining after 7 to 28 days (Smith, Harding, Hecke). In sweet corn infected in the seedling stage by Bacterium stewarti, a period of 1, 2, or 3 months may intervene between the first signs of disturbance in the seedling leaves and the general sickening of the plant, during which, of course, the plant has grown to many hundred times its original weight. There may be an equally long period between local infection and constitutional disturbance in case of sugar-cane attacked by Bacterium vascularmn. In Savastano's experiments with the olive-tubercle, knots began to appear upon the young shoots in a little over a month after puncture and were well developed in 2 months. In my own and Mr. Rorer's experiments, incipient knots were frequently visible as early as the end of the second week, i. e., sufficiently developed to be distinguished from the control punctures, and were very distinct in a month, but larger, of course, after several months (see vol. I, plate 2). Once in a later experiment, starting with cultures very recently plated from a knot and introducing the organism by needle-pricks from agar, I observed the beginnings of tumefaction on 5 shoots as early as the ninth day. In case of the soft-galls, due to Bacterium tumefaciens, the writer has sometimes obtain- ed the distinct beginnings of them as early as the third or fourth day, using pure cultures, needle-punctures, and very susceptible tissues such as young shoots of the Paris daisy. Earlier than this it is not possible to decide whether the slight swellings are to result in tumors or are simply a reaction of the plant to the needle-thrust. Ordinarily if the organism is virulent, the tumors are distinctly visible in 8 or 10 days if the tissues are young and tender, but they continue to grow for several months, or even many months. The bacterial leaf-spot diseases are usually visible in 1 to 2 weeks from the date of inoculation. DURATION OF DISEASE. Plants show very different degrees of resistance to a bacterial organism once ensconced in the tissues. The soft-rot organisms, as already noted, are usually prompt in their action, and a week or two is often sufficient to destroy the susceptible parts. The writer has seen a good-sized potato-tuber half rotted in 5 days at ordinary autumn temperatures when inoculated with Bacillus phytophthorus by means of a few needle-pricks, this too, in a rather dry air; others wholly rotted in 8 or 10 days. Under favorable circumstances, inoculating from a young agar-culture, the flesh of a melon one decimeter in diameter may be rotted wholly by Bacillus melonis in 4 or 5 days (Giddings, Smith). In case of the leaf-spots, progress is much slower, and the disease is generally restricted to small areas, e. g., bacterial leaf-spot of the peach, which dry out and fall away from the sound tissue, especially if excised by a cork layer. After plain signs appear, a week or two is sufficient in most cases to accomplish the destruction of the affected part. In cucumbers and muskmelons attacked DURATION OF DISEASE. 67 by Bacillus tracheiphilus, the progress of thedisease is rapid after the incubation period has passed. In squashes, on the contrary, the resistance is much greater, and it may be several weeks after a vine shows the wilt before it entirely succumbs. At night, or in moist weather, it becomes turgid, to again collapse with the reappearance of sunlight and dry air. In the growing season, pear-blight is usually a rapid disease, but in the cool weather of autumn and winter there is frequently an almost balanced activity between host and parasite, result- ing in what is known as "hold-over" blight. In this way the disease is carried from one growing season to the next. Vascular diseases, such as those of sweet corn and sugar- cane, already mentioned, kill the plant very gradually, if it is of good size when infected or when the constitutional signs first appear, but after the vascular occlusions have reached a certain volume the destruction of the plant is speedy. In maize, which has reached this stage, the leaves dry out within a few days, and the green stem then shrivels. In case of the olive-tubercle, the tree as a whole does not, so far as we know, become infected but only particular parts of it, yet there may be wide metastasis especially in young trees. Individual knots live for several months, and fre- quently portions of them for several years, the knot enlarg- ing from some particular part which has not been injured beyond the power of cell-division. Terminal twigs girdled by tubercles are frequently starved and die, but not very promptly. Knots and cankers due to bacteria are generally of slow progress and correspondingly long duration. The life of a shoot of chrysanthemum, sapped by a big tumor due to Bact. tumefaciens, varies from 6 months to a year or more. Often the plants live many months. Peach trees attacked by crown-gall generally live for several years. Galled apple trees may live indefinitely. In the recently discovered tuberculosis of the sugar beet due to Bacterium beticolum {Vide Crown Gall, etc., Bull. 213) decay is rather prompt. FINAL OUTCOME. Plants, like animals, are affected to very different degrees by the various bacterial parasites. This must be apparent from what has been said under duration of the disease. In the animal world there are protracted bacterial diseases and rapid ones, diseases terminating fatally or end- ing in recovery. The same is true of plants. The simplest cases, perhaps, are the stomatal infections resulting in leaf- spots and fruit-spots. The leaves are more or less disfigured, and the fruit may be destroyed or so spotted as to be unsal- able, but generally it is beyond the power of the organism to destroy the plant, or even to render it wholly unfruitful Fig. 19.* In bacterial blights, such as that of the mulberry or pear, much larger portions of the plant may be destroyed, twigs or even large branches, and yet it may recover. In a majority of cases, after running a *Fig. 19. — Coconut budrot of Eastern Cuba. Outer enveloping leaf sheaths removed to show condition of inner undeveloped leaves — sound below, rotted above. Tree No. 1 1. Bud itself not dead, but enveloping sheaths rotted. Color of decayed part was a mixed gray and brown. Photographed by the writer at Baracoa, Cuba, April 20, 1904. One-third natural size. 68 BACTERIA IN RELATION TO PLANT DISEASES. certain course, which is usually not shorter than several weeks, the disease stops and the organisms which caused it are then found to be dead in the blighted tissues (pear-blight). This, however, does not seem to protect the tree from new infections the following year, i. e., the disease is not self-limited and protective from new infections like the eruptive fevers. Not infrequently rapidly growing, juicy trees of pear, apple, quince, loquat, and mulberry are killed outright in the course of one season if left untreated. Even whole orchards have been destroyed, as in Georgia and California. Olive-tubercle also sometimes kills young trees, but more often it kills only some of the smaller branches and renders the tree unfruit- ful. Certain infections seem to kill almost infallibly. This is true of Bacillus trachciphilus in musk-melons and cucumbers, and of virulent strains of Bad. solanacearum in young Fig. 20.* tomatoes, potatoes, and egg-plants. Whole fields of potatoes, tomatoes, and tobacco when young may succumb quickly to this disease, particularly in moist soils containing root nematodes. In carefully made inoculations on young plants, using either of these organisms, at least 95 per cent of the infections are promptly fatal, i. c, within 2 or 3 weeks from the first visible signs of the disease, and sometimes much sooner (see vol. I, plates 24 to 27). Old plants are more resistant, especially to Bacterium solanacearum. In the same way old and slow-growing cabbages are rather resistant to Bacterium campestre and may not be wholly destroyed, but young and rapidly growing plants are very apt to die either from the direct effects of the parasite or from the action of the soft-rots which follow it. "Fig. 21). — Coconut budrot of Eastern Cuba. Outer leaf sheaths removed to show inability of diseased terminal bud to support its own weight. Tree No. 10. Photographed at Baracoa, Cuba, April 18,1904. Natural size. Photo- graphed in a room at 2 p. m., raining, with Cramer's isoinstantaneous plate, Zeiss double protar lens, series Vila, stop 250, time 30 minutes. BACTERIAL ACTION ON THE PLANT. 69 TISSUES ATTACKED. In what we may consider as the lowest type of these diseases the parenehyma-cells of storage tissues are the parts principally attacked. Often these are aggregated in fleshy organs which have reached maturity and ceased to grow but abound in water, amid, proteid and carbohydrate substances, designed for the green plant which is to be developed the coming season. Buds, bulbs, tubers, rhizomes, and various swollen underground parts of mixed structure are good examples. Examples of such diseases are certain soft-rots of potato-tubers, Jones's carrot-rot, Metcalf's rot of sugar beet, and the coconut bud-rot (plate 5, and figs. 19 and 20). Usually they do not appear in green, growing parts. They destroy the tissues by softening the middle lamella?. In a little higher grade of essentially the same type of disease, the green parts of plants are alsoattacked, e. g., iris-rhizome-rot, Appel's potato-rot, lettuce-rot, calla-lily-rot. All these organisms require tissues rich in water, otherwise they refuse to grow or make very little headway. A grade higher in the scale, perhaps, are those bacteria which attack the parenchyma of stems, roots, bark, green leaves, etc., but can do so only when the tissues are in a rapidly growing, actively dividing condition. They may retain a foothold for some time thereafter, in exceptional cases, but their power for evil is limited to a short period of the growing season. Black-spot of the plum and pear-blight (vol. I, plates 28, 29) are good examples. All of the leaf-spots are primarily diseases of the parenchyma, and some of them are limited to quite restricted areas of the parenchyma, e. g., leaf-spot of the carnation, larkspur, soy- bean. Other diseases like pear-blight and Aderhold's cherry-blight often extend a long distance through parenchymatic tissues. In case of the pear the upward or downward movement of the bacteria in the bark may be several meters, and the sidewise movement is often sufficient to girdle and kill large limbs or even the whole tree. In general, however, destruction is more extensive when the organism is able to attack the vascular system as well as the parenchyma. Then we have phenomena of occlusion, and marked interference with transportation of water. There are intermediate forms and transitions of various sorts as might be expected. The pear-blight organism, so far as I know, seldom follows the vessels, expending its energy rather on the cortical parenchyma. Some leaf-spot bacteria seem to use the vessels more than others. Appel's potato-rot organism (B. phytophthorns) makes some use of the vessels of the stem, but seems more at home in the parenchyma. On the contrary the organism causing Stewart's sweet-corn disease, to take a very striking example, develops principally in the vascular system and destroys the plant from this vantage ground. This is true also of Bacillus tracheiphilus and Bacterium vascularum. The same is true of Bad. solanacearum, in potato and tomato, only here the organism finally floods out into the tissues of pith and bark much more than in the other cases cited (fig. 1). Bad. solanacearum and Bacillus phytophthorus may be compared and contrasted in this particular since both attack the potato. Both make use of the vessels, but the former does so much more extensively and destructively than the latter; the one is primarily a vascular disease, the other a paren- chyma disease ; one destroys the stem by occluding the vessels, the other by rotting it off at the surface of the earth. All vascular diseases make pockets in the parenchyma, but in most cases these are only in close proximity to the vessels (fig. 6) and after the latter have been occluded and destroyed. In very soft tissues such as those of watery, fast-growing tomato shoots, Bad. solanacearum finally makes very extensive closed cavities, often honey- combing both pith and bark for many centimeters. Bacterium vascularum, though restricted pretty closely to the bundles for long distances in the maturer parts of the stem of sugar- cane, often excavates extensive closed cavities in the very soft undeveloped parenchyma under the terminal bud. The gum diseases rupture the bark and ooze extensively on the surface. Pear-blight does this also to a lesser degree. Some, perhaps all, of this class of bacteria reach the surface through fissures due to surface tensions set up in dead tissues by 70 BACTERIA IN RELATION TO PLANT DISEASES. the parts still alive and growing. Others reach the surface of the living plant, if at all, principally through natural openings, i. c, stomata or lenticels. I do not know of any bacteria confined to the sieve-tubes, or particularly at home therein, but probably this only amounts to saying that the whole field has not been surveyed. Parasitic bacteria, limited to the woody tissues of trees and shrubs or supposed to be peculiarly at home therein, have been described but are unknown to the writer. Examples cited in literature are Mai nero of the vine, and Janse's disease of dadap trees, both abund- antly doubtful etiologically. The writer has seen bacteria-like bodies in fossil woods, but these woods may have been occupied by them after submersion in some swamp. He has also seen a yellow Schizomycete very abundent in the wood of pear trees attacked by pear- blight, but this had no pathogenic properties. Viala and Ravaz found a very abundant multiplication of a Schizomycete in the vessels of vine cuttings buried in sand to be used later as grafts, but the organism was unable to propagate itself in the living, growing plant when these cuttings were used either as grafts or scions, and cultures made from these bacteria had no pathogenic power when inoculated into the vine. There is no apparent reason, however, why the wood of living trees should be wholly exempt from the attacks of bacteria, and cases will probably be discovered in which bacteria are confined pretty closely to the woody tissues. They must of course be sorts able to live on a minimum quantity of water. The highest type of bacterial disease, and the most interesting from many standpoints, is that in which all the growing tissues, pith, wood, cambium, and bark are involved, and are stimulated into abnormal, excessive growth, death occurring only after extensive hyperplasia. These overgrowths may attack roots or shoots. Good examples are olive- tubercle, pine-tubercle, beet-tubercle, the daisy knot, and crown-gall of the peach. In the olive, oleander and daisy they often occur on the leaves. Metastatic tubercles and secondary tumors occur. This type will be discussed more at length under Reactions of the Plant. There appear to be two forms of these overgrowths. In the olive tubercle, bacterial cavities occur and the organism is abundant in them, and is easily observed wedging its way between cells. In the crown-gall no such cavities have been observed, the causal organism is difficult to detect with the microscope, and its location in the tumor tissue appears to be unlike that of the olive-tubercle organism. Moreover, plate cultures show that it is not very abundant in the tissues, at least in a viable form. The olive tubercle organism occupies intercellular spaces. The crown-gall organism occurs within the rapidly dividing cells, as in case of the root-nodule organism of Legumes, but less abundantly and does not form a bacterial strand. MASS-ACTION OF BACTERIA. A few words are necessary on the mass-action of bacteria. It is a common observation, one made by the writer at least a hundred times, that in culture-media not exactly adapted to the needs of an organism, a scanty inoculation may not give any growth — not even after a long time — whereas a copious one will lead to a growth which gradually clouds the fluid or covers the solid. The penetration of bacterial strands from cell to cell in the root- nodules of Leguminosae is another example (figs. 21, 22). The only explanation I can think of is that a multitude of the bacteria are stronger than a few, and thus by union are able to overcome obstacles too great for the few. The same fact comes repeatedly to the attention of the animal pathologist as a result of his inoculations. The animal body, we must assume, is often able to overcome and destroy a few hostile organisms, where it would not be able to defend itself against many; otherwise whole races would be exterminated by natural infections. The same is undoubtedly true in plants. The modus operandi in plants is not altogether clear. We may advance several hypotheses: (1) The formation of a resistant cork-layer before the bacteria have multiplied to such an extent as to prevent PLANT BACTERIA. VOL. 2. PLATE 5. U i QJ hr a a o> * Phi £.£ — >. — o a a> ft-v o-S 'L J= 91 "J! ._ EH en - "N rt c ca — ■Sg ■^ £ 4_l o ^ >* o o H 0 «**— U 3 X EE O 0 •it. MASS-ACTION OF BACTERIA. 71 cell-division; (2) the destructive action of antiseptic plant-substances, e. g., acids, before these can be neutralized or otherwise destroyed by the substances produced by the mul- tiplying bacteria. In some instances, the introduction of a very considerable mass of bacteria seems to be necessary to induce disease ; in other cases a very few are sufficient. It would be extremely interesting to know the minimum number capable in any given case of inducing disease. This could be determined easily by the dilution method, and still more readily and with Fig. 21 Fig. 22 f absolute accuracy by the use of Barber's apparatus, but no one seems to have made any exact experiments. Good organisms for experimental purposes would be Bacillus phytoph- thorus, Bacillus trachciphilus, Bacillus amylovorus, and Bacterium campestrc, care being taken, of course, to select sensitive varieties and susceptible tissues, and to have all other factors comparable. SECONDARY TUMORS AND METASTASIS. Secondary foci of overgrowth occur in the olive and in the daisy as the result of internal infection. The writer has obtained them frequently in both plants by pure culture inocula- tions (plates 6 and 7, and fig. 23). The organisms pass through the tissues of the stems or leaves and set up irritations which lead to hyperplasias in particular spots in the deeper tis- sues. These tissue enlargements, later on, break through to the surface. Sometimes these secondary growths arise at a considerable distance from the primary tubercle. In case of olives inoculated in 1910 the writer observed numerous deep tubercles develop at a distance of 1, 2, and 3 feet from the point of inoculation within a period of 7 months in actively growing plants, both down and up the shoot. The movement is more apt to be up the stem or leaf, i. c, with the transpiration current, than down the stem. In the olive a distinct channel of infection is traceable from the primary to the sec- ondary tubercle. This is usually (so far as observed) a narrow pathway in some part of the inner wood, the tissues being more or less stained and disorganized, and the bacteria present in abundance and easily demonstrable without staining. Whether similar downward *Fig. 21. — Three figures from Peirce's paper: (I) Two root-hairs of Bur clover infected by nodule bacteria, showing characteristic bending at point of infection, x 50. - (2) The lower of two root-hairs in 1, showing mass of bacteria in concavity of coil and infection thread running from this point through the hair, x 300. (3) Another infected and coiled root-hair, infection thread growing close to nucleus of hair, x 300. fFic. 22. — Two figures from Peirce's paper on Root Tubercles of Bur Clover: (10) Section of a tubercle near meristem. Direction in which meristem lies is indicated by arrow. Section stained by Fleming's triple stain and differentiated, after anilin gentian violet, by Gram's iodine. Course of infection threads is definitely toward tubercle-meristem and generally toward nucleus of cell entered, x 200. (II) One cell from 10, showing solid infection strand (zoogloeae) in which separate bacteria can be dis- tinguished, x 1000. 72 BACTERIA IN RELATION TO PLANT DISEASES. channels occur in the bark has not been determined. The channels in the wood, which probably begin with an occlusion of some of the spiral vessels, generally occupy but a very small portion of the stem, although they are easily visible on cross-section even to the naked eye as small brown specks and on longitudinal section as a dark line bordering the pith and connecting the two tumors. In the Paris daisy attacked by Bacterium tumefaciens I have been able to obtain leaf- tumors in a considerable portion of my inoculations by making a single needle-prick into the soft young stem below the leaf, the infected needle being thrust into one of the three leaf-traces. A primary tumor results at the point of inoculation on the stem, and some weeks later a secondary one develops in the inner tissues and subsequently bursts through the upper surface of the leaf, i. e., through the petiole or midrib. Sometimes there are a series of such internal secondary tumors (fig. 24). In young daisy plants inoculated January 13, 191 1, upon the stems, secondary (inter- nal) tumors appeared upon the leaves very promptly and after 16 days were visible 7 to 10 cm. away from the primary tumor. The movement in this case was upward and the little tumors had not yet burst through the midrib. Sections were cut and the internal tumor tissues studied in these midribs, and motile bacteria-like rods in small numbers were seen inside the cells in the gall tissue. The channel of infection in this case is not so easily traced. There is no plain disor- ganization of tissues, and no brown stain such as occurs in the olive, and although I have studied a good many sections I do not yet know through what tissues the channel of infection passes or whether there is any definite channel. It is more likely that the tumor tissue itself carries along the organism within the rapidly dividing cells, there being a chain of tumor cells all the way from the primary to the secondary tumor. Since the above paragraph was written additional observations have been made con- firming this hypothesis. A strand of tumor tissue has been discovered connecting the primary tumor with the secondary tumors. This is usually in the inner wood or at the junction of pith and wood. Moreover, when the primary tumor is in the stem and the secondary tumor is in the leaf', the latter tumor does not have the dorsi-ventral structure of the leaf in which it is growing but rather the structure of an imperfect stem. In cross- section it is circular or nearly so, and is composed of a pscudopith of tumor tissue beyond which is xylem surrounded by cambium and what I take to be phloem. Often remnants of the petiole or midrib adhere to the surface of the tumor as a shell or as wing margins, in which one finds the unchanged dorsi-ventral structure of the leaf. The strand or wedge of tumor tissue proceeding from the primary tumor grows through the inner tissues of the stem, petiole, and leaf -ribs much like a foreign body, developing the secondary tumors from its substance in places of least resistance, or where the food supply is most abundant. These Fig. 23 : *Fig. 23. — Metastatic small tubercles (nearly erumpent) on midrib of olive leaves at xx. Inoculations were made Nov. 26, 1907, on stems about 0.5 inch below nodes, and growths developed at these points: Subsequently, these leaf- tubercles developed from wiUiin. Inoculated plants Nos. 424a and 412. Photographed Feb. 29, 1908. PLANT BACTERIA, VOL. 2. PLATE 5o. Crown Gall of Dais)-. CrOSfh';eC'i°? °f sf'T °b a" '"OCD'.ai!eu P'aIlt '«'ween a Primary tumor and secondary tumors, showing a tumor strand in the inner wood in the «"ter of the figure. Pith below. The secondary tumors are outgrowths from such strands See next plate and page ;_• Slide PLANT BACTERIA. VOL. 2. PLATE 5b. Crown Gall of Daisy. Cross-section of petiole from an inoculated plant showing the central leaf-trace enlarged and converted into a pseudo stem by the parasitism of Bad. tumefaciens. The reason for the stem structure of the leaf tumor ties in the fact that the tumor-strand infecting this leaf-trace is an outgrowth from a primary tumor on the stem. The remainder of the petiole is normal Slide 634 B 10. PLANT BACTERIA, VOL. 2. PLATE 6. Tuberculosis of Olive. Shoots of olive inoculated Nov. 26. 1907, at yv by needle-pricks. A pure culture of Bact. savastanoi, 3 days old, was used. Photographed Mar. 13, 1908. Metastatic tumors in process of development at xx. Channel of infection in vessels next to pith. PLANT BACTERIA, VOL. 2. PLATE 7. Crown Gall on Daisy. Two tumors on the stem of a Paris daisy as the result of an inoculation of Bad. turne/adens by needle-pricks, and on a branch above the upper one a secondary tumor on the petiole of a leaf. Age of primary tumors about three months ; that on the leaf is much younger, perhaps four weeks old. SECONDARY TUMORS AND METASTASIS. 73 internally developed secondary tumors contain the parasitic bacterium the same as the primary tumors, but it is not abundant in any of the tissues. The bacterium causing the disease has been found also sparingly in the strand connecting primary and ] secondary tumors. Much difficulty has been experienced in stain- ing Bad. tumejaciens in the tissues of the daisy, and ideal preparations are yet to be obtained. It would seem from an examination of numerous slides, and also from the results of many poured plates, that the organism occurs in the tissues of the tumor in rather small numbers, especially as compared with the organism causing the olive-tubercle. From what I have seen I believe it occurs only inside the paren- chyma cells, stimulating them to divide and passing on from mother cells to daughter cells in this manner. The development of these tumors depends on a very delicate series of adjustments between the attacking organism and the host cells. My present hypothesis as to the mechanism of the tumor development in crown-gall is as follows: Through wounds which have not injured the cells beyond the power of recovery (needle-pricks in case of my experiments) the bacterium gains an entrance into the cell; here it multiplies rapidly for a short time; its further growth is checked by the appear- ance of acid to which it is very sensitive — this acid being developed in the cell as a by-product of the bacterial action on sugar; the first effect of the acid is to inhibit the further growth of the bacteria and consequently there never can be very many bacteria in any individual cell; the continued action of this acid on the bacteria leads to the pro- duction of involution forms (clubs and Y's) and finally a portion of these bacteria are killed out- right, but the concentration of the acid is not suffi- cient to destroy the host cell; the nucleus of the latter now divides, either under the direct stimulus of the acid, or under the influence of bacterial endotoxins which now for the first time have been liberated, i. e., have come into contact with the nucleus by diffusion through the permeable mem- branes of the dead bacteria ; during the cell-division the bacteria are carried over into the daughter cells, where under the new conditions those not destroyed by the acid multiply rapidly for a short time; then in turn their growth is checked by more acid, whereupon ensue the other changes ending in another cell division. In this way is developed first the primary tumor, then the strands penetrating the sound tissues in various directions, and finally the secondary tumors, which I have elsewhere called metastatic tumors, but which are probably not so in the sense that loose tumor cells migrate from the primary tumor to Fig. 24.* *Fig. 24. — Leaf of Paris daisy {Chrysanthemum frutescens) developing internal tumors at points marked by letter x and beyond. The lower one has split open surface of rib, others are still sub-epidermal. Source of infection was the stem-tumor here shown, which was induced by needle-punctures introducing Bacterium tumefaciens. Photo- graphed Feb. 12, 1908. 74 BACTERIA IN RELATION TO PLANT DISEASES. form them, since the circulation in plants is not as well adapted as in animals to this sort of migration. The stages outlined above take place with great rapidity since in very sus- ceptible tissues, e. g., young, rapidly growing sugar-beets, it is possible by means of a few needle-pricks to obtain a tumor as large as a man's fist in 5 or 6 weeks. These phenomena represent to me an entirely new type of bacterial disease. They seem to me also to throw a flood of light on the mechanism of the development of malignant animal tumors, making it likely that they also are due to parasites having similar relations to the cells of man and the lower animals. The facts underlying this hypothesis may be summarized as follows: (1) The crown-gall disease is of bacterial origin beyond reasonable dispute, as shown by hundreds of poured plates and pure culture inoculations. It is also a neoplasm rather than a granulomata (vide evidence advanced in Bulletin 213). (2) The bacteria can not be found readily in the tissues by means of microscopic examinations although the poured-plate method shows that they occur there, and the vessels and intercellular spaces being free from any granules whatsoever, the bacteria must occur inside of the cells, forming some portion of the cell-inclusions. (3) The poured plates confirm the microscopic examinations. They show that the bacteria are not abundant in the tissues. They also show that these bacteria often occur in the tissues of the tumor in a moribund state, requiring 4 to 6 days or more to recover and develop colonies on the agar, although when once recovered they grow in second and sub- sequent transfers as promptly as other organisms. (4) In flasks containing water, peptone, grape-sugar, and calcium carbonate the organism (from the daisy) produces an abundance of acetic acid. (5) Chemical analysis shows an excess of acid in the tumor tissue as compared with sound parts of the same plants (daisy, sugar-beet), but up to this date a sufficient quantity of the tumor for a definitive quantitative test (10 kilos or more) has not been available. If acetic acid is formed in the tumor cells, it must be in minute quantities, and it might be oxidized by some subsequent action of the host protoplasm so as not to be recoverable on chemical analysis. (6) In artificial cultures club-shaped, Y-shaped, and variously branched bodies can be produced at will by adding small quantities of acetic acid. (7) Similar forms occur in the tissues of the tumor, and while I have not seen them in the cells they can be obtained on sterile slides in small numbers by making sections of tumors and allowing them to diffuse in sterile water for a few minutes. (8) When too strong a dose of acetic acid has been added to the agar, or bouillon cultures, the Y's and other involution forms can not be resuscitated by means of agar poured plates, but when the dose has been properly adjusted a portion of the bacteria may be recovered in poured plates, the colonies coming up slowly the same as when material is taken from the interior of the tumors. (9) Finally, the statements respecting the tumor strand, the anatomy of the secondary tumors, and the occurrence of the bacteria in these latter are supported by many observations and experiments. Schiff-Giorgini first clearly recognized metastasis in the olive, although earlier Savas- tano pointed out that some tubercles develop superficially and others from the deep tissues. WAYS OF ENTRANCE — CARRIERS OF INFECTION. 75 LITERATURE. 1856. 1887. 1891. 1895. 1896. 1903- 1905- 1905. 1905 Schacht, Hermann. Bericht an das Konig- liche Landes-Oekonomie-Collegium iiber die Kartoffelpflanze und deren Krankheiten. Berlin, Verlag von Karl Wiegandt, 1856, Quarto, 30 pp., 10 plates. Savastano, Luigi. Tubercolosi iperplasie et tumori dell' olivo. Ann. R. Sc. Sup. Agr. Portici, 1887, vol. v, fasc. 4. Waite, Merton B. Result from Recent In- vestigations in Pear-blight. Bot. Gazette, 1891, p. 259. Smith, Erwin F. Bacillus tracheiphilus sp. nov. die Ursache des Verwelkens verschiedener Cucurbitaceen. Centralb. fur Bakt. Abt. 2, Bd. I, No. 9-10, 1895, p. 365. Smith, Erwin F. A Bacterial Disease of the Tomato, Egg-plant, and Irish Potato (Bacillus solanacearum n. sp.). Bull. No. 12, Div. of Veg. Phys. and Path., U. S. Dept. of Agric, Dec. 19, 1896, 26 pp., 2 plates. Smith, Erwin F. Observations on a Hitherto Unreported Bacterial Disease the Cause of which Enters the Plant through Ordinary Stomata. Science, N. S., vol. xvn, No. 429, March 20, 1903, pp. 456-457. Smith, Erwin F. Bacterial Infection by Way of the Stomata in Black Spot of the Plum. Science, N. S., vol. xxi, No. 535, March 31, 1905, P- 502. Smith, Erwin F., and Hedges, Florence. Burrill's Bacterial Disease of Broom Corn. Science, N. S., vol. xxi, No. 535, March 31, 1905. PP- 502-503. Schiff-Giorgini, Ruggero. Ricerche sulla tubercolosi dell'ulivo. Reale Accademia dei Lincei (Anno ccci, 1904). Roma, 1905, pp. 185-210, 2 plates. Also a separate. 1906. 1907. 1908. 1910. 1911. 1911. 19". 1911. 191 1. Smith, Erwin F. Channels of Entrance and Types of Movement in Bacterial Diseases of Plants. Science, N. S., vol. xxm, No. 585, March 16, 1906, pp. 424-425. Smith and Townsend. A plant tumor of bacterial origin. Science, N. S. Vol. xxv, April 26, 1907, pp. 671-673. Smith, Erwin F. Recent studies of the olive tubercle organism. Bull. 131, pt. IV, Bureau Plant Industry, U. S. Dept of Agric. Jensen, C. O. Von echten Geschwiilsten bei Pflanzen. Deuxieme conference internat . pour l'etude du cancer. Rapport. Paris, Oct., 1910, pp. 243-254. Smith, Erwin F. Crown-gall. Phytopathology, vol. 1, No. 1, pp. 7-11, 2 plates. Feb., 1911, Ithaca, N. Y. Andrus and Church, Printers. Smith, Brown and Townsend. Crown-gall: Its cause and remedy. Bull. 213, Bureau Plant Industry, U. S. Dept. of Agric, Feb. 28, 191 1, pp. 215, 3 text-figures, and 36 plates. Smith, Erwin F. Crown-gall and Sarcoma. An. Meeting Am. Asso. for Cancer Research. Buffalo, N. Y., April 12, 1911. Circular 85, Bureau of Plant Industry, U. S. Dept. of Agiic. Barber, M. A. A Technic for the Inoculation of Bacteria and other Substances into Living Cells. Jour, of Infectious Diseases, vol. 8, No. 3, April 12, 191 1, pp. 348-360. MlCulloch, Lucia. A Spot Disease of Cauli- flower. Bull. 225, Bureau of Plant Industry, U. S. Dept. of Agric, Wash., D. C, 1911, 15 pp., 3 plates. SOLVENT ACTION OF BACTERIA-DESTRUCTION OF MIDDLE LAMELLA-BACTERIAL SOLUTION OF CELL-WALLS-FERMENTATION OF CELLULOSE- DESTRUCTION OF WOOD. Next to crushing and splitting, due to the rapid multiplication of the bacteria in closed spaces, solution of the middle lamellae uniting cell-walls is probably the most widespread and simple action of bacteria on plant tissues. This is common in a great number of diseases, but it is not always possible clearly to separate lysis from tension-splitting, when the bac- teria are multiplying rapidly in a given tissue and must have room. An excellent example of the separation of cells by a solvent action on the pectic matters composing the middle layers of the common wall may be seen in various rots of potato-tubers. A few days after an inoculation the tissue softens, and if it is then washed in water the cells float free, their starch content remaining unacted on (fig. 25). Potter asserts this solution to be due, in case of a turnip rot which he studied, to the presence of oxalic acid. The writer found oxalic acid had no solvent action on slices of turnip, but that ammonium oxa- late softened the middle lamellae decidedly. Inasmuch as a part at least of any oxalic acid liberated by any Schizomycete of this type would be converted into ammo- ; a S'a ■c £0 2 «.- « 3 »8 3 o ^ o 1 s ~- ^2 . ■oSfc S3* ■- +j o ta .ft o b ao j ■£"".= ■p a, > w o uo a o SYMBIOSIS. 97 within the scope of this treatise. Is there such a thing? We will run over some of the alleged sorts and let the reader decide for himself. The most discussed case in recent years is the relationship existing between root-nodule bacteria and legumes. ROOT-NODULES OF LEGUMINOSAE. Should the root-nodules of Leguminosae be cited as examples of symbiosis? The plant submits to distortions and enlargements and final destruction of portions of its roots, giving water, mineral foods, and carbon compounds in exchange for which it receives nitrogen compounds, at least this is the current view. The agricultural chemists appear to be satisfied that the host-plant actually receives the nitrogen and that it is from this source, and is not simply combined nitrogen drawn again to the surface of the earth by the deep feeding roots of legumes, as would be our first thought, considering how readily nitrogen compounds leach out of agricultural soils into the deeper substrata where they can not be reached by surface feeding roots. The pathologist sees a nodular growth stimulated by the presence of a foreign organism and various phenomena not unlike those of genuine parasitism as Peirce and others have pointed out. We might have, however, a local injury and yet general advantage to the plant if the bacteria really store nitrogen available to the legume. The organism appears to be able to infect only through very young roots or root-hairs (fig. 21). As soon as the cells of the roots have passed out of a rapidly dividing condition the nodule takes on a definite form and ceases to grow. vSubsequently it passes through the same stages of disorganization as other overgrowths in which there is no suggestion of symbiosis. That the micro-organisms infest the interior of the cells, rather than the intercellular spaces, does not alter the case materially. In the end they destroy plasma and nucleus (fig. 31), and the nodule decays. So far then as the morbid anatomy goes we must look upon Bact. leguminosarum as a restricted parasite. Does the host receive something in return ? The agriculturist has observed a more luxuriant growth when the nodules are present on the roots of the plants than when they are absent. This, however, by itself might mean only that infections are most abundant on rapidly growing plants, i. c, on plants in a good soil capable of inducing a rapid growth. In recent years, however, in experiments on poor soils, marked increase of growth over that in untreated check plants has been obtained sometimes, by infecting the soil or seeds with this organism at planting time (plate 12). This experiment has yielded the same result, it is said, on a large scale in field practice (see statements by Fig. 31.* Hiltner and by Moore), and that not once, but many times in widely different localities. On many fields, however, no marked difference has been observed between the check plots and those inoculated with this organism, and sometimes the check plots have given the best returns, even when nodules have been abundant on the roots of the inoculated plants. vSometimes these failures have been on fields already well stocked with nitrogen compounds, but apparently such has not been the case always. Moreover, granting the increased growth associated with an increased number of nodules on the roots it does not necessarily mean free nitrogen assimilated by the bacteria and turned over to the host- plant, unless it can be shown that combined nitrogen is absent from the soil and air, since plants often make an increased growth under the stimulus of weak poisons. This extra *Fig. 31. — Normal and shriveled nuclei from cells in root-nodule of soy-bean. The three lower nuclei are from cells fully occupied by Bact. leguminosarum — they are distorted, flattened, have taken the stain deeply, and apparently little is left except skin of nucleus. The two at top of figure are globose and faintly stained except the peripheral chromatin — they are from uninfected cells distributed sparingly among cells destroyed by the bacteria. 98 BACTERIA IN RELATION TO PLANT DISEASES. growth has, however, been obtained in some instances, it would seem, on soils practically destitute of combined nitrogen, and also, I believe, in an atmosphere destitute of all traces of combined nitrogen, i. c, of ammonia and nitric acid. Figure 32 shows a pea grown to maturity in a closed space on nitrogen-free sand (on which oats and buckwheat starved) by adding soil extract containing the nitrogen root- nodule organism. This water extract contained only 0.15 mg. of nitrogen, and the hermeti- cally sealed carboy was opened to the date of the photograph (fiftieth day) but three times and then for a few moments only to introduce a measured quantity of pure washed carbon dioxide (6 liters) necessary for the growth of the plants. The capacity of the carboy was 44 liters and while the contents of the air in combined nitrogen was not deter- mined it could not have been over a small fraction of a milligram. The pea grew for a period of 4 months and fruited, yielding a total dry weight of 10.359 grams, of which 233.5 mgs- were nitrogen. Some nitrogen was also re- covered from the sand, making (after proper deductions for nitrogen in the seeds, etc.) 248 mgs. of fixed nitrogen. The oats and buckwheat made only a very starved growth and finally perished without fruiting. It will be remembered that Boussin- gault got no growth and no nitrogen assimilation in a closed space under sterile conditions. This is exactly Boussingault's experiment plus the addi- tion of soil extract containing root- nodule bacteria, the result being entirely different. The bacteriologist finds that in pure cultures the root-nodule organism appears to be able to grow on substrata in which there is absence of nitrogen compounds or at least great paucity of such compounds. Maze states that the organism requires a minimum of com- bined nitrogen to make a good growth and store nitrogen. He says it can not do so if all initial combined nitrogen is withheld. Others (c. g., Moore) state that it can make a decided growth on media in which there is no combined nitrogen. According to these observers it makes a good growth on media believed to be free from all nitrogen compounds, and hence the assumption is that it must be able to obtain its nitrogen from the uncombined nitrogen in the air. It is, however, a difficult matter for the ordinary bacteriologist to be assured that all traces of nitrogen compounds are absent from a given culture medium and from the air *FlG. .12.— Hellricgcl and Wilfarth's pea No. 384 grown from germination time in a closed space along with an oat plant and a buckwheat plant on nitrogen-free quartz sand. The sand which had been previously heated to redness was mixed with nitrogen-free nutrient salts and enough twice-distilled water added to approximate 70 per cent of soil saturation. The sand was then inoculated with a little soil extract from earth adapted to peas. Erbse No. 384. Fig. 32.* ROOT-NODULES OF LEGUMINOSAE. 99 supplied to his flasks. Quite a good many bacteria not known to be assimilators of free nitrogen will make a little growth in some of the so-called nitrogen-free media. The chemists, therefore, have undertaken to determine whether flask cultures of Bad. leguminosarnm show any increase of nitrogen as a result of their growth. Most have found no gain, or so slight a gain of nitrogen as to be within the limits of experimental error (see Beyerinck's statement). Maze is almost the only one who has reported large gains of nitrogen in flask cultures. I do not know what opportunities there are for error in the ordinary nitrogen determinations, but on the bacteriological side I detect a good many suspicious statements in Maze's papers. Miss Dawson's comment, that Maze's statements respecting ability of this organism to store nitrogen are to be accepted only with the greatest reserve, appears to be entirely proper. See also Hiltner's comments. The whole subject of the storage of free nitrogen by this organism in flask cultures and in the plant itself ought to be worked over again carefully by the bacteriologist and chemist. Possibly the root-nodules are only indicators of a fixation of nitrogen which actually takes place in the soil. Certainly it should be determined whether Bad. leguminosarum is able to fix nitrogen outside of the plant in agricultural soils both sterilized and unsterilized. The question why the addition of pure cultures of the organism to certain soils does not increase the yield of alfalfa and similar crops, should also be determined. Hiltner has made commendable attempts in this direction. Also, it should be determined why the organism so readily loses its virulence. There are, therefore, several fundamental problems connected with this question of nitrogen fixing in legumes which require further study. Doubts also exist in some quarters as to whether what is commonly called Bacillus radicicola has anything whatever to do with the production of the root-nodules. These doubts have been sharply focussed by Gino de Rossi who maintains that a Schizomycete of quite different character is the real cause of the nodules (see abstract), and that we know nothing about its ability to store nitrogen. Hellriegel and Wilfarth postulated symbiosis. Hiltner seems to waver between sym- biosis and parasitism. Maze maintains that it is not necessary to explain the fixation of atmospheric nitrogen by the hypothesis of symbiosis, the micro-organism being able to gather its own nitrogen without aid from the plant. SYNONYMY OF BACTERIUM LEGUMINOSARUM. Frank's Schinzia leguminosarum appears to be the earliest name and therefore I write Bacterium leguminosarnm (Frank) as the proper name for the organism causing the root-nodules on Pisum, Vicia, Laihyrus, etc., since it is a Schizomycete, motile by means of a polar flagellum (see vol. I, pp. 165-171). The type form to which this name applies may be taken as that causing the nodules on Laihyrus {Orobus) tuberosus. Should Hiltner's view prevail respecting the existence of two distinct species, Beyerinck's specific name radicicola may be retained for the organism causing the nodules on Lupinus, Ornithopus, and soy-bean. The name Rhizobium beijerinckii Hiltner and Stormeris inadmissible because there is an earlier Bacillus beycrinckii Trevisan, and also because Kirchner's specific name japonicum, applied to the organism causing the root-nodules of soy-bean, is earlier. Bacillus radicicola Beyr. is still earlier and the name Bacterium radicicolum may be used in place of Hiltner's name. Moreover, there is some doubt whether the name Rhizobium should apply at all to the root-nodule organism, since Frank stated his Rhizobium to be a micro- coccus. There is no doubt, however, that Frank applied the name Schinzia leguminosarnm to the zoogloea? strands of this bacterium. That he interpreted them to be fungous filaments does not invalidate the name. Pscudorkizobium ramosum Hartleb (1900) is a name given to a non-infectious organism obtained from root-nodules. The name Bacillus beycrinckii was given by Trevisan (1889) to the white, liquefying, non-pathogenic organism isolated by Beyerinck from root-nodules. IOO BACTERIA IN RELATION TO PLANT DISEASES. The synonymy of Bad. Icguminosarum (Frank) is as follows: Syn: Schinzia Ieguminosarum Frank (1879); Bacillus radicicola Beyerinck (1888); Cladochitrium tuberculorum Vuillemin (1888); Bacterium radicicola Prazmowski (1889), Moeller (1892); Phytomyxa Ieguminosarum Schroter (1889); Rhizobium Ieguminosarum Frank (1890)?; Bacillus ornithopi Beyerinck (1890); Bacillus fabae Beyerinck (1890); Rhizobium mutabile Rh. curvum Rh. Frankii var. maiuslr- . •. / 0 \ n, _ . .. . J >Sehneider ( 1892); Rh. Frankn var. minus [ i*"y/> Rh. nodosum Rh. dubium Rhizobium sphaeroides Schneider (1894); Micrococcus tuberigenus Gonnermann (1894); Rhizobium Pasteurianum Laurent (earlier than 1899 according to Maze, but the writer has been unable to find the name in any of Laurent's papers). Bacterium (Rhizobacterium) japonicum Kirchner (1895) (applied to the soy-bean organism) ; Rhizobium Beijerinckii Hiltner and Stormer (1903) (applied to organism causing the root-nodules of Ltipinus, Ornithopus. and Soya) ; Rhizobium radicicola Hiltner and Stormer (1903) (applied to organism causing the root-nodules of Pisum, Vicia, Phaseolus, etc.); Pseudomonas radicicola Moore (1905). In 1906 Stefan suggested that the root-nodule organism is related to the Myxo- baeteriaceae. In 1910, Peklo maintained it to belong with Actinomyces. SUMMARY OF LEADING PAPERS. From the hundreds of pages relating to root-nodules the writer has culled the following statements: Apparently the first person to discover bacteria in the root-nodules was Woronine (1867). He states that he made his examinations principally upon the common lupin of the gardens (Lupinus mutabilis Lindley). He describes the nodule as composed of an interior parenchyma, an exterior parenchyma, and a vascular system between the two, the cells of the interior parenchyma being occupied by enormous numbers of small rods which he figures and describes as bacteria. His exact statement respecting their classification is as follows: " In all these respects, they have the greatest resemblance to those organisms of doubtful nature which have been designated under the names of Bacterium Duj., Vibrio Ehrbg., Zoogloea Cohn, etc., and we may, without doing violence to the analogies, range them in the same class." His figures correspond very well to the facts as we now understand them, even to enlarge- ment and shriveling of the nucleus (fig. 33). The same paper deals with the enlargements on the roots of Alnus. In these he says he found a fungus, described as Schinzia alni. Frank (1879) saw the bacterial filaments, sometimes ending in a sharp point in the middle of the cell. To him they were hyphae. The "hyphae" and the small rods and branched bodies were believed to be parts of the same fungus, although Frank was far from certain respecting this. This paper is the one in which he first used the name Schinzia Ieguminosarum (column 397). Ilt-llriegel and Wilfarth (September 20, 1886) demonstrated the great difference between the nitrogen nutrition of grains and legumes and connected the latter with the presence of the root- nodules. They made pot experiments, using quartz sand washed many times, nutrient solutions contain- ing all the mineral elements necessary for growth except nitrogen, and watered with pure distilled water (the first part of the distillate being rejected). Under these circumstances they found that the crop of oats or barley was in direct proportion to the amount of nitrate of soda added. When no combined nitrogen was added these grains soon showed nitrogen hunger and uniformly perished. Under the .same conditions, i.e., nitrogen compounds withheld but all other foods added, peas grew remarkably well and produced seeds. Many comparative experiments were made and the re- sults were uniform. It was plain, therefore, that the peas did not obtain their nitrogen from the soil. ROOT-NODULES OP LEGUMINOSAE. IOI Did they obtain it from combined nitrogen present in the air? To settle this question, peas were grown under bell-jars in washed air, i.e., in air from which all the nitric acid and ammonia had been removed, and the growth was just as good as in the unwashed air. Growth was also good on nitrogen-free soil in a closed space in a limited volume of air which could have offered to the plants only a trace of combined nitrogen (fig. 32). The conclusion, therefore, appeared to be irresistible that the peas were in some way able to assimilate free nitrogen. Boussingault having already shown that legumes can not directly assimilate free nitrogen, the only hypothesis open was some indirect assimilation through the assistance of other organisms. They were led to the conclusion that the root-nodule organism was the factor sought, by having observed that after peas had used up the stored food in the seed there often followed a period of nitrogen-hunger during which growth stopped and the leaves became pale or yellow, but that after a time the green color returned and growth was resumed. In certain plants, however, this resump- tion of vigorous growth never took place and the roots of such plants were observed to be nearly or quite destitute of root-nodules, whereas the roots of the other plants bore nodules, and the more abundant and better developed these were, the better the growth of the plants appeared to be. ' ■■'■■■-,)■ —^ mm) J7 Fig. 33.* The next step, therefore, was to add and exclude nodules or nodule products, and so determine the results experimentally. This was done by taking 40 experimental pots containing nitrogen-free soil, holding 30 for checks, and to the other 10 adding 25 cc. of an extract of fertile soil, containing only 1 mg. of nitrogen per pot, all being planted to peas. All passed through a period of nitrogen- hunger, but the plants in the pots inoculated with the soil-extract all regained their green color and grew freely and uniformly. In only 2 of the 30 check pots did the plants grow freely. All the rest continued to show nitrogen-hunger, and some became quite yellow. This difference in color and amount of growth was found to be correlated with the presence or absence of root-nodules. *Fig. 33. — Nos. 8 to 11. Stages in development of root-nodules on common garden lupin {Lupinus mutabilis). No. 12. Same in tran verse section as seen under a hand-lens. Same as 12 but in longitudinal section; ap, external parenchyma; ip, internal parenchyma; gb, vascular bundles. Cross-section, x 120. Mature cells of internal parenchyma containing rod-shaped corpuscles, x 320. Cell showing escape of rod-shaped bodies by resorption of membrane, x 320. Mass of bacteria surrounding nucleus which has survived cell membrane, x 620. Isolated bacteria which have become modified and have ceased to be motile, x 620. 14- No. No. No. No. 16. No. 17. No. 18. After Woronine. Reduced one-eighth. 102 BACTERIA IN RELATION TO PLANT DISEASES. The effect of the soil-inoeulation on legumes differed from the effect of nitrate of soda in that, in the former case after the period of germination, a peculiar and very characteristic hunger-stage supervened which was followed by very energetic and rapid development of the plants. In two experiments under sterile conditions, the peas grew well in the nitrogen-free sand until the food stored in the seed was exhausted and then dwindled, dying after about 6 leaves had been formed. On these plants not a trace of root-nodules could be found. The same negative result was obtained when the soil-extract was boiled or heated to 70° C. before adding it to the pots. They concluded, therefore, that the nitrogen assimilation of legumes was in some way connected with root-nodules and the bacteria present therein. In other words, as expressed in their final report: There are, therefore, for the Leguminosae two sources of nitrogen, viz., the combined nitrogen of the soil and the elementary nitrogen of the air, the latter being made available to them through the agency of micro-organisms which, to be effective, must enter into a symbiosis with the plant. Numerous experiments with lupins failed : No successful second growth could be obtained with pea soil and the conclusion was reached that the nodule organism of lupins must be different from that of peas. Only when inoculations were made with soil from a field where lupins grew well did the experimental plants overcome their nitrogen-hunger and do well. On this experiment two check rows of pots were held, one untreated and one inoculated with extract from pea soil. In all three the plants germinated and grew well at first. Then followed a period of starvation, each of the three rows showing equal nitrogen-hunger at the end of a month. Then the first row became green and grew well, while the other two rows continued feeble and red-brown in color. The roots of the first row (inoculated with lupin soil) bore numerous large nodules. The roots of the second row bore none whatever. The roots of the third row (inoculated with pea soil) bore none whatever, except one plant on which asingle small nodule wasfound. Serradella behaved like the lupin. Peas, vetches, and beans grew best in the third row. In Ilellriegel's own words: " Leguminosenknollchen und Wachsthum der Papillionaceen in stickstofffreiem Boden lassen sich willkurlieh hervorrufen durch Zusatz von geringen Mengen Kulturboden und Verhindern durch Ausschluss von Mikroorganismen. Bei verschiedenen Papillionaceenarten wirkt nur der Zusatz von gewissen Bodenarten Knollchen bildend und Wachsthum fordernd. " Lawes and Gilbert sum up these experiments very well in the following paragraph: "The negative result with the Gramineae, the negative result with the peas when everything was sterilized, or when the sand was not seeded by the soil-extract, the positive result with the peas when the sand was seeded by the humus soil extract, the negative result with the lupins when their soils were net seeded, or when they were seeded with the same extract as the peas, and the positive result when seeded with the extract from the sandy soil where lupins were growing, seem to exclude any other conclusion than that the micro-organisms supplied by the soil-extracts were essential agents in the process of fixation. Further, the development of nodules on the roots was, to say the least, a coincident of the fixation. The following year (1887) Dr. Wilfarth stated at the Naturforscher Versammlung in Wiesbaden that they had repeated and extended their experiments with wholly confirmatory results (plate 12). From this time on the scientific world generally accepted their views as may be seen from the fol- lowing comments of Lawes and Gilbert: "Thus it may be considered established that the Papillionaceae can take the whole of their nitrogen from the air. * * * " It will be seen that the results are not only confirmatory of those given by Hellriegel the vear before, but that they are even much more definite and striking. Thus, taking no account of the fraction of a milligram of combined nitrogen supplied in the soil-extract, the amount of dry matter produced is nearly 50 times, and the amount of nitrogen assimilated is nearly 100 times, as much with, as without, the soil-extract." The full account of their experiments was first published in 1888. No figures since published are any more striking or convincing than the six plates which appeared in this epoch-making pub- lication, two of which are here reproduced (plate 12 and fig. 32). In their experiments with serradella, Series C, 1897 (plate 12), each jar contained 4,000 grams of sterilized nitrogen-free quartz sand to which was added the necessary nitrogen-free nutrient salts (monopotassium phosphate, potassium chloride, calcium chloride, and magnesium sulphate). Fight seeds were germinated in each jar, the number of plants being reduced soon after to four. The plants were watered with distilled nitrogen-free water. To some jars additional fertilizers were added as follows: Nos. 264 and 265 each a small amount nf calcium nitrate (41 mgs.); the two end pots of each row (right side) received each 40 grams of calcium carbonate which was sterilized by heat and mixed with the sand previous to planting; No. 250 received some potassium carbonate. PLANT BACTERIA, VOL 2. PLATE 12. f? yp *4 I'M 00 5 < 2 ■= e *as 2 1= r-t - 0 '■2 ~~ ? is [fl E U .. 3 = ill — 0J2 a; *— *0 2 s« a P 2z aJ-13 0> s ^ o I WfH' ROOT-NODULES OF LEGUMINOSAE. IO3 All those plants which had not received nitrates showed nitrogen-hunger early (about time of appearance of third leaf). By June 28, the plants inoculated with the unboiled soil-extract had recovered from their nitrogen-hunger and grew rapidly from this time on. This increased growth of the plants in the upper row was correlated with the presence of nodules on the roots of each one of the plants. There were no root-nodules on the plants in the lower row. The total grams of dry substance from the upper row (28 plants) was 106.542; the total grams from the lower row (32 plants) was 1.888. According to Tschirch (1887) the filaments possess no membrane but only a hyaline border layer and, therefore, have nothing in common with fungous hypha; they are not of a fungous nature. It is very unlikely also that they are plasmodial strands. Tschirch saw and figured the trumpet-like expansions where the filaments penetrate the cell-walls. He believed the membrane was not pene- trated. The bacteroids probably are not given off from the filaments. Their variable form is opposed to the view that they are of bacterial origin. It appeared to him rather that the filaments decomposed and then later the bacteroids were developed out of the cell plasma. The filaments and bacteriods are held to be two stages in the differentiation of the cell -contents of the plant itself. The nodules are considered as transitory reserve tissue, especially for albumen ; possibly also starch. Marshall Ward's communication on this subject was read before the Royal Society of London in June 1887, and appears as the last paper in the Philosophical Transactions which was published in 1888, that is at about the same time as Beyerinck's paper in the Botanische Zeitung. Ward obtained numerous infections in water-cultures by binding on slices of the root-nodules. He saw and figured the entrance of the organisms through the root-hairs in the form of filaments or tubes, as he calls them, and the penetrations of these filaments through the cortex into the tissues where he observed, figured and described the characteristic branching. He also saw the bacteria and bacteroids within the cells. He interpreted the whole phenomenon as one of fungous infection. He regarded the parasite as related to the Ustilagincae and considered the bacteroids to be yeasts budded from special portions of the hyphae. His description is so full and his figures so distinct that there can be no doubt of his having meant them to apply to this organism. Most of his studies were made on the broad bean, Yicia faba. His cultural work did not lead to any satisfactory results. In his second paper, published in 1890, he says: "I may here say that these cultures (/. e., as micro-cultures) have given me much trouble, and little results. To obtain pure cultures is a matter of greater difficulty than Beyerinck's paper would lead one to expect, and it is not proposed at present tolay much stress on the evidence got from them." This second paper figures the entrance of the organism into the root-hairs. First a bright spot appears near the apex of the hair, and from this a little later a filament projects into the interior, and grows toward the cortex. The root-tip curves. Following Woronine, Beyerinck in Holland was the first man to recognize clearly the nature of the organisms occurring in root-nodules of Leguminosae. These he designated Bacillus radicicola. His long paper in Bot. Zeit., 1888, called general attention to this subject. The following are some of the statements made in this paper: The splitting of the primary bark for the emission of the side roots is the special means of entrance of B. radicicola. The bacteroids stain like B. radicicola but not intensely. They are of various shapes, branched, round, pear-shaped or bacteria-shaped. They are incapable of growth. Melampyrum pratense has root-nodules containing bacteria (DeVries, Beyerinck). Beyerinck says such nodules also occur on Rhinanthus major and on the roots of Alnus, Eleaginus, and Myrica. While the bacteroids can be found in nearly every cell of the nodule, and occur also in the bark and epidermis of normal roots, the interior of the central cylinder is the special tissue of the bacte- roids (see fig. 34). "The bacteroids are organized albuminoid bodies which the plant has formed out of Bacillus radicicola, for the purpose of local storage of albumen — therefore an organ of the plant protoplasm, developed from bacteria which have wandered in." (Column 732.) Two types of nodule are recognized — one in which the bacteria gain the ascendency and destroy the interior, they themselves remaining alive; the other, in which the nodule, i. e., the host plant, gets the advantage; the bacteria being converted mostly into bacteroids, incapable of growth and furnishing food for the plant. He calls those nodules normal in which no bacteria remain capable of germination, except perhaps in the meristem. The small threads which pass from cell to cell he considered to be remnants of the nuclear spindle. Sometimes when the bacteria get the upper hand, nucleus and cytoplasm are destroyed. Beyerinck distinguishes three sorts of bacteroids: (1) normal, (2) a smaller sort called "hem- 104 BACTERIA IN RELATION TO PLANT DISEASES. mung's bactcrioiden," and (3) bladder bacteroids. The third form occurs where the bacteria have multiplied enormously. The hcmmung s bactcrioiden occur outside of the bacteroid tissue in nearly all the outer cells of the nodule, and not rarely in the normal bark of the root. Beyerinck found that the nodules did not develop on the roots of plants grown in sterilized soil. Frank reached the same conclusion in 1879, Hellriegel and Wilfarth in 1S86, and Ward in 1887. Plants in soil rich in humus are sometimes free or nearly free from nodules. According to Beyerinck there is only one bacterial species, but not all the forms are identical. There are varieties. There is, for instance, a distinct difference between the bacteria occurring in Yicia, Ervum, Trijolium, and Pisum, on one hand, and in Lotus, Lupinus, Ornithopus, and Phaseolus on the other hand. In the large rapidly growing colonies one is most apt to find B. radicicola like ordinary bacteria; in the small /[ is&i slow growing ones there are more -J\\ branched bacteroids. He obtained the strongest growing colonies out of the very young nodules, or out of the outer meristematic zone of the older ones in Yicia faba, this being the plant he studied most carefully. The inner zone of the meristem Wjl*-:^*'.'^ , '-/' yielded more bacteroid elements and slower growing colonies. The same result was obtained with t. '-|8 J Lupinus polyphyllus. This he V* "*.J&1 «£&' saYs ig tne lupin in which Woronine first saw the bacteria. The nodules of this plant are very large and the swarmers in them are very minute. There are no slime threads and there is no meristem. The large watery colonies consist of a mixture of rods and swarmers, many motile. The rods exclusive of some long forms are about 4X I//. The bacteroids of Viciafaba are some- what larger and average 5X i". The swarmers are very small : Taken from Yicia faba they are 0.9X0.18/'. They are so small that granting them some plasticity they might easily penetrate the perieambium cells without leav- ing any visible wound. They possess one polar flagellum. This was inferred from behavior during slow motility rather than actually seen. Motility ceases almost immediately in hydrogen or carbon dioxide. The little slow-growing colonies also contain swarmers. B. radicicola has no special powers of fermentation, oxidation, or reduction. It does not produce spores. It is not harmed by freezing or drying (see fig. 35). Neither diastase nor invertase are produced. Cellulose and starch are not converted. Nitrates are not reduced. Oxygen is liberated from hydrogen peroxide. It is aerobic. It does not liquefy gelatin. Meat-water peptone gelatin is too concentrated for the first cultures (isolations). The addition of 0.25 per cent asparagin is useful in agar cultures. Cohn's solution is too acid for B. radicicola. It will not grow in it even after neutralization. Aklaline and neutral solutions are also injurious. For B. radicicola from Trijolium repens 0.6 per cent ' malic acid is useful. Fig. 34.' *Fio. 34. — Planar enlargement of stained section from a small root nodule on soy-bean. Great bulk of section consists of cells much enlarged and occupied by enormous numbers of liml. leguminosarum. Colorless spaces between are occupied by smaller (non-distended) cells free from infection and bearing normal nuclei. Surrounding this central mass is vascular tissue and beyond that cortex (both free from bacteria). ROOT-NODULES OF LEGUMINOSAE. I05 Beyerinck found swarmers in minute nodules which were still inclosed in the mother root. He divides the root-nodule organisms into groups and varieties as follows : Group I. — This contains the larger more hyaline colonies. Growth absent or difficult on meat peptone gelatin. Growth is favored by cane-sugar or grape-sugar. vSwarmers are very minute. The bacteroids are two-armed, globose, or pear-shaped. Meristem is always present in the nodules. The primary bark of the nodule is closed. Slime threads are distinct. The following forms belong here : B. radicicola, vars.fabae, vicia-kirsutae, trijoliorum, pisi, lathyri. Group II. — Colonies more cloudy white. Growth better on meat peptone gelatin. Swarmers more rod-shaped, somewhat longer. Bacteroids like the bacteria, that is, seldom branched. Slime threads absent or little developed. Mostly no meristem in the nodules (Robinia an exception). Three types occur: (1) Phaseolus type; (2) B. radicicola, var. lupini; (3) Robinia type. In Yicia faba, as the bacteroids are exhausted the color of the cytoplasm changes from reddish to intense green. The bacilli from this plant when grown in Faba stem gelatin in a cool place (cellar) were alive and motile at the end of a year. Active cultures can be obtained from all parts of the nodules which have been exhausted by the bacteria. They are present in a living condition therein in great numbers. The result is quite different when the host empties out the contents of the bacteroids. Then it is more and more difficult to get any bacterial growth from the meristem. The longer the bacteria remain in the nodules the more bacteroids occur. Beyerinck found saprophytes in the nodule tissues mixed in with B. radicicola and named at least two — B. luleo albus and B. agglomerans. Another green fluorescent form thought certainly to come from the nodule was identified as B. fluoresceins putidus. A form resembling B. radicicola and found in certain nodules was first named B. radicicola lique- faciens, but subsequently Beyer- inck came to regard this as an intruder having nothing to do with their formation. This lique- fying organism was afterwards called Bacterium beyerinckii by Trevisan. The bacteroids are found in other parts of the roots than the nodule, but less well developed, e. g., in the root-hairs and epi- dermis cells. Beyerinck never found them in parts above ground, except once in a stem of Yicia faba where inoculated by hypo- dermic injection. The bacteroids are always derived from the bacteria. They occur in old cultures as well as in the nodules. The swarmers easily pass through the walls of the Chamberland filter. When fresh nodules are put into water at room temperature this water clouds first with a mixture of bacteria, of which B. radicicola is the chief. Later, when the nodules decay, other bacteria appear. The tissues of legumes have a strong attraction for this organism, as is shown by the fact that in such roots placed in the water any little cracks or wounds are immediately occupied by this organism and the intercellular spaces flooded with it. These roots may be considered as a bacterial trap apparatus. The infection of the living pericambium of the root must take place through pores, possibly Fig. 35 * *Fig. 35. — Poured-agar plates of Bart, leguminosarum from bean, introduced to show effect of repeated freezings: a, Contents of a loop before freezing — several hundred colonies per square centimeter; b, Contents of a similar loop of culture fluid after 7 freezings — less than one colony per each 2 square centimeters. Each freezing lasted half an hour; time between freezing short, i. e., only long enough to thaw out tube in cool water and make necessary plates. Round colonies are on surface ; spindle-shaped ones are buried. 106 BACTERIA IN RELATION TO PLANT DISEASES. those by which the protoplasm of one cell is connected with that of another. The infection is an active and not a passive one. The nodules are not to be considered as normal organs of the plant. There is no doubt, however, that the plant takes advantage of the presence of the bacteria. "The papillionaceous tubercles are bacterial galls, useful to the host plant in so far as the normal bacteroids function as providers of albumen — useful to the bacteria in so far as the numerous tuber- cles filled with bacteria capable of growth function by their decay as centers for the distribution of their occupants." Beyerinck found no formation of nitrates or nitrites and could not establish the assimilation of free nitrogen in flask cultures. He says that in 14 days the nitrogen increased according to the find- ings of the chemists but only within the limits of error. Through its ability to use asparagin and grape-sugar, B. radicicola has the same nutrition demands as the protoplasm of the host. Beyerinck evidently believed that the B. radicicola gets its nitrogen from asparagin stored in the plant. In his Dutch paper published in 1891, Beyerinck states that by additional experiments he succeeded in showing that his B. radicicola from Vicia faba is able to obtain nitrogen from the air. A very great number of active bacteria were used. The increase of nitrogen was, however, so very slight (only about 12 milligrams per 100 cc. on an average in 6 cultures 2 to 3 months old) that he suggests the possibility of its having come from some nitrogen compound present in the air rather than from the free nitrogen. The cultures were kept at 2° to 12° C, higher temperatures being thought to induce loss of function; the weakened cultures take their food more readily from am- monia salts and nitrates than the unweakened ones. By the diffusion method in gelatin he could not be certain of his results, and with the bacillus from the root-nodules of Robinia he could not obtain any increase of nitrogen in 8 weeks. The cultures were made in Kjeldahl flasks in extract of bean stems (100 gr. germinating sprouts to 1 liter of tap water), with 2 per cent cane-sugar and with one-tenth to one-thirtieth gram of monopotassium phosphate. Sometimes without the latter. Nitrites in all dilutions are said to be injurious to the growth of the organism. Cane-sugar is a much better source of carbon than asparagin. The earlier statements of Beyerinck respecting the value of asparagin should be rejected as they were obtained from an associated organism confused at that time with B. radicicola. Peptone is a better source of nitrogen than asparagin, ammonium sulphate, or nitrate of soda or potassa. The growth of B. radicicola is greatly favored by extract of papillionaceous plants or dilute must extract. Another point insisted upon is that the concentration of the food stuff should be low, especially the nitrogen compounds and the phosphates. The negative result of his earlier attempts to prove storage of nitrogen is attributed to neglect of this point and to growth at too high temperatures. In 1897, in hisfirst paper on the subject, Maze states that heused a bouillon made by heating white beans in water for half an hour at a temperature of ioo° C, being careful not to boii it. This bouillon contained about 0.0005 0I nitrogen. To it was added 2 per cent saccharose, 1 per cent sodium chloride and traces of bicarbonate of soda. This was solidified by the addition of agar and spread in a thin layer (o to 4 mm.) on the bottom of large flasks having a side opening by means of which air freed from nitrogen compounds could be introduced into the flasks. Into these tubes he introduced air which had been passed through an asbestos plug, then over copper turnings warmed to just below redness, then through pumice stone saturated with sulphuric acid to remove the free ammonia, and afterwards through pure water. The flasks were set up in series, the last one in connection with the aspirator. This removed 20 liters in 24 hours, not counting the more rapid movement of the atmos- phere every morning to remove the gaseous products of respiration accumulated during the latter part of the night. At the end of 15 days the experiment was broken off. Microscopic examination indicated the cultures to be pure and transfers to sterile media also indicated the same thing. An analysis showed that there was a gain in the 15 days of 40.8 mgr. of nitrogen, the initial amount being 62.1 mgr. In a second experiment which lasted also 15 days the gain in nitrogen was 47.5 mgr., as shown by analysis of a mixture of the contents of the two flasks, the initial nitrogen being 70.7 mgr. In a third experiment, using bean bouillon without the agar the experiment was broken off on the sixteenth day, and the results of the analysis of two flasks united showed a gain of 32.4 mgr., the initial amount of nitrogen being 22.4. He concludes that symbiosis is not necessary to explain the fixation of atmospheric nitrogen by the nodule-forming bacteria. This is a property belonging to the organism independent of any influence exercised upon it by the plant. The lack of success experienced by former experimenters he thinks due principally to a defect in the method of experi- mentation. They placed too little value on the energy necessary to enable the nodule-forming bacilli to convert the nitrogen of the air into an endothermic combination. To place this organism ROOT-NODULES OF LEGUMINOSAE. 107 in a medium deprived of combined nitrogen, obliging it to depend for nourishment from the beginning upon the nitrogen of the atmosphere is to demand of it more than it is able to do. "We see that the dose of sugar can not fall much below 2 per cent, for the experimenters who have worked with media containing only one per cent of sugar have not found any sensible increase of the nitrogen." Easy access of air also exercises a very favorable influence on the fixation of nitrogen, and this is easily comprehended, for the rapidity of the combustion of sugar stands in relation to the quantity of oxygen furnished to the cultures. It is because he did not fulfil this condition of aeration that Mr. Beyerinck has observed only a very limited fixation of nitrogen. Maze states that the plant must furnish the bacillus 100 grams of starch in order to receive in exchange 1 gram of nitrogen. "The cultures of the bacillus of the Leguminosae in bean broth, exhaled a strong odor, not without analogy to that which is given off by soft cheeses (brie and camembert)." According to Maze's second paper (1898) Bad. radicicola does not grow in an atmosphere of nitrogen, although it remains alive for some time. Laurent's contradictory results are to be ascribed to a defective experiment, /'. e., to traces of oxygen left in his air. The organism is an aerobe. It is greedy of oxygen. It is able to fix free nitrogen without the assistance of the plant. In fixing nitrogen in flask cultures Maze states the best result to be when the combined nitrogen was 1 to 200 of the saccharose, the lower limit of the latter being 2 per cent and the upper limit 4 per cent. The minimum limit of combined nitrogen in bouillon cultures is 14 mgr. per 100 cc, and the maximum about 30 mgr. per 100 cc. Maze's evidence in favor of the storage of nitrogen is increased by another experiment. In three 50 cc. flask cultures there was more than twice as much nitrogen at the end of the experiment as at the beginning, the gain being respectively 12. 1 mgr., 12.8 mgr. and 15 mgr. Two other flasks in the same series, differing only slightly in nitrogen and sugar- content, gave no increase of nitrogen and there was only a slight decrease in the amount of sugar. Another experiment is mentioned but here the gains and losses are so slight as to seem within the limits of experimental error (p. 133). The nitrogen is not all locked up in the organisms; a portion is soluble and will dialyze (about j in a flask culture of 100 cc. diluted to 800 cc. with distilled water, i. e., 8.5 mgr. out of 32.04 mgr.). In media containing very minute quantities of combined nitrogen the root-nodule organism makes a feeble growth and does not fix free nitrogen. He got no increase of nitrogen in 50 cc. flasks of bean broth containing as little as ^.t, mgr. of combined nitrogen. This agrees with Beyerinck's results, and contradicts Frank's, Prazmowski's, and Laurent's. Legumin is a good source of nitrogen. Nitrates are better foods than ammonia salts. In ammoniacal bouillon cultures 30 days old there was no increase of nitrogen and very little diminution of the saccharose. The nodule bacteria grew also in sterilized soil free from nitrates, but with no increase of nitrogen (3 months) : One experiment only and believed to be insufficient. He states that he did not succeed in isolating from the soil a bacillus capable by itself of producing nodules. Saccharose and dextrose attract these bacteria. Water of germination repels them. They are sen- sitive to acids. The only chemotactic substances emitted by the roots of legumes are carbohydrates. Laurent states that the maximum temperature for growth of Bad. radicicola is 30° C, but Maze found it grew very well on agar at 350 C, especially after a few transfers. The branched forms are due to vegetation under harmful conditions as shown by growth in acid media and at 350 C. During the first few transfers at 350 C, and especially at the end of the first 24 hours, they are abundant. In successive transfers as the bacteria become accustomed to this temperature the branched forms disappear entirely. If the branched cultures are diluted with bean bouillon they give rise to unbranched rods. It is impossible to fix in the breeder's sense the branched forms by any method of culture. Growth in bean bouillon is prevented by the addition of a small amount of tartaric acid (1 : 1000). By sowing very copiously, growth was obtained on slightly alkaline agar to which 1 : 1000 tar- taric or oxalic acid had been added and here pear-shaped forms were found. The pear-shaped and branched forms found in the nodules are ascribed to the injurious action of the acid cell-sap of the host. The bacteria as isolated from the nodules do not liquefy gelatin. Later Maze obtained from some of his cultures round forms believed by him to be part of the life cycle, and these liquefied gelatin rapidly. Maze states that the round and rod-shaped forms, which he believed at first to be two species, but later forms of one, inoculated separately do not give nodules. Those roots inoculated with mixtures of the two organisms gave numerous nodules. Maze states that the active rod-shaped form is unable to form nodules. In this he is clearly wrong. He is probably wrong also as to the relationship of the round organism, and this throws more or less doubt on all of his paper. Many of his conclusions 108 BACTERIA IN RELATION TO PLANT DISEASES. seem to me doubtful. I think he was experimenting with mixed cultures. Especially do I think his theory of alternation of generations, in which Oospora is one stage, and an endospore bearing bacillus another stage, not well supported. Possibly, therefore, the nitrogen stored in his flasks may have been due to some other organism than Bad. leguminosarum. In no part of his paper are the details of his experiments so stated that one could reproduce them. Apparently he did not make use of poured plates, but depended for isolation on streak-cultures, made in tubes of slant gelatin. He states that the nodule-producing organism is pathogenic for some species of animals, e. g., rabbits, but this also seems to me not well established by his experiments, since he obtained Oospora and an almost round form of very small diameter from the rabbits inoculated with a supposed pure culture of the nodule organism. Abscesses formed, locally, in the inoculated animals. In 1899, Maze published a fourth paper on the bacteria of leguminous root-nodules in which he reviews the methods of Salfeld and of Nobbe for inoculating the soil with these bacteria, and in addition gives some of his own experiences. Concerning Nobbe's work he says: To justify the method which he recommends, Nobbe starts out with the following hypotheses: There exist in the soil, neutral forms, capable of forming tubercles on the greater part of the Leguminosae, and forms adapted to definite species. In general the infection of plants takes place by the former, especially in uncultivated soil or in soil which has not borne Leguminosae in a long time. The neutral form is modified profoundly by a passage through a leguminous plant becoming, in this way, incapable of infecting other species. Bacteria thus adapted constitute a definite race : Thus the species Bacillus radicicola (Beyerinck) or Rhyzobium pastcurianum (Laurent) comprises a certain number of races each possessing the ability to infect particular species of Leguminosae. Sometimes a race is able to attach itself to different plants, closely related botanically, but it is not able to utilize atmospheric nitrogen upon these inappropriate hosts. Maze then raises the question, not mentioned by Nobbe, as to how these races pass over from one season to another. He says : "May we conclude that they retain after months and years the ability of their ancestors to live incapable of attaching themselves to plants of other species than the one which previously sheltered them? Nothing would be less justifiable than such an assumption. It has long been known in bacteriology, that all species of bacteria are subject to the influence of the medium on which they live. More than any others, the bacteria of the Leguminosae possess this adaptability which assures the dissemination and preservation of a species." Mazd claims that forms living in the soil lose, little by little, the characteristics which made them easily identified when taken directly from the nodules. A dilution of soil applied to plants growing in nutrient solutions caused nodules after 15 days. The same dilution inoculated on a series of agar tubes, made for the purpose of obtaining isolated colonies, did not give any forms which corresponded either morphologically or physiologically to the typical bacterium of the nodules. By a long series of passages with all the species obtained from these cultures he states that finally two forms were obtained which he identified by inoculations as the root-nodule organism. From this he draws the conclusion that forms isolated from the soil acquire gradually, when subjected to a medium containing the proper carbohydrate and nitrogen, the ability to elaborate the mucilaginous substance and to fix atmospheric nitrogen. He thinks, therefore, that this ability is very unstable with the bacteria of the Leguminosae. They acquire it in the nodules and lose it in the soil. He gives the following experiment as proof of this: He sowed the nodule bacteria on both sterilized and unsterilized soil kept saturated during the whole experiment. On the unsterilized soil, conditions favored the growth of soil bacteria. At the end of 8 months it was impossible to obtain colonies resembling those which supplied the bacteria sowed. On sterilized soil the bacteria removed from competition with other soil bacteria, retained their initial characteristics after S months. He states also that the characteristics of these races of bacteria at the moment of isolation from the nodules are far from being as distinct as Nobbe claims. Thus, for example, a bacterium coming from one leguminous species is capable of attaching itself to certain other species. Nobbe admits this but thinks that while able to form nodules the bacterium is no longer able on these strange plants to fix nitrogen and so it becomes a parasite which is frequently injurious. Maz£ does not agree with this last statement: Bacteria from any of these plants will fix nitrogen if they have sugar and enough initial nitrogen. The plants all offer that, and he says that the only condition requisite to nitrogen fixation is their ability to penetrate them. This ability, he thinks, depends on the alkalinity or acidity of the soil. He found that lupins inoculated with bacteria from furze and broom formed just as many and as large nodules as those inoculated with bacteria from the lupin, while the checks showed no trace of nodules. The furze and broom came from uncultivated ROOT-NODULES OF LEGUMINOSAE. 109 land which probably had never borne lupins. The only satisfactory explanation which he finds for this is the long adaptation of the bacteria to soil having the same reaction. Those which live in alkaline soils are capable, he thinks, of invading all plants indigenous to such soils while those living in acid (non-calcareous) soils attach themselves indifferently to the lupin, the furze, and the broom. These facts led him to undertake new experiments. He says that if the reaction of the soil is the essential reason for the existence of two great physiological groups of nodule bacteria it should be sufficient to accustom to acid media a bacterium from alkaline soil, in order to render it capable of producing nodules on the roots of the lupin. This he says he succeeded in demonstrating : Bacteria cultivated 8 months on media of gradually increasing acidity produced nodules on all the lupins inoculated with them. The nodules appeared on the first lateral roots. There were none on the tap roots. Five checks gave only 1 nodule. The same experiments on white lupin gave negative results when grown in mineral solutions, but results were positive on plants normally developed in sand. In explanation of the influence of the acidity of the soil on the penetration of roots Maze says: " If the soil is alkaline, the acidity of the secretions of the roots is neutralized to a certain depth in the tissues. The bacteria, very sensitive to the action of acids, penetrate this layer, attracted by the diffused sugar, but are not able to go farther into the roots." There must, therefore, he thinks, be forms especially adapted to acid soils. From his results he concludes that Nobbe's hypothesis is not confirmed either by cultural experiments, or by the physiology of the bacteria. He says that the bacteria which are free in the soil may be grouped according to the reaction of the soil, into two great categories, and that the forms which are found in acid soils are capable of invading only those plants which avoid alkaline soils such as the lupin, furze, and broom. Concerning the prevalence of the bacteria he says that when they do not manifest themselves by the production of root-nodules it is not because they are absent from the soil but because the conditions for their development are lacking. These conditions are obtained by proper treatment of the soil, and certainly no one would attempt to inoculate with pure cultures a soil that had not been so ameliorated. Hence he thinks that the use of pure cultures does not greatly aid agriculture. This opinion applies to the nodule bacteria of the Leguminosae. It remains to be seen whether the bacteria of alinite are of as little value. In 1899, Maria Dawson contributed an interesting paper on nitragin and the nodules of legu- minous plants. Her investigations were carried on in England in the laboratory of H. Marshall Ward and were suggested by the commercial introduction of nitragin by Nobbe and Hiltner. Her studies were confined principally to Vicia hirsuta and Pisum sativum. Each showed pal- mately branched nodules within 1 4 days of sowing. Various fixatives were tried. The most satisfactory results were obtained by using Flemming's more concentrated solution or absolute alcohol. Hand sections served better than microtome ones for examination of the bacterial filaments within the cells. She found abundant evidence of the parasitic nature of the organism. In fresh material the infection tubes were made visible by treating with Eau de Javelle or potash. In all cases a bright spot of infection was seen either at the tip or at the side of the hair, accompanied by a bladder-like swelling of the hair at the point of attack. Hand sections of fresh material treated as above showed the course of the infection-tube across the cortex and its branching into the deeper cells of the nodule. Trumpet-like swellings where the tubes cross the cell-walls and numerous spherical or pear-shaped swellings on the tube within the cells (previously described by Marshall Ward) were clearly seen, as well as breaks in the tube, each portion ending in a fine point, the points directed toward each other (tig. 36). This also has been seen by others. The author next attempted to obtain a reagent which would stain the filaments but not the bacteroids. Gold chloride (0.5 per cent) used on fresh material gave some help. Fresh material en masse was left in the stain from 1 to 24 hours for microtome sections, while hand sections were stained for 10 or 15 minutes. In either case the material was quickly washed with water and trans- ferred to a solution of formic acid (0.25 per cent) in the dark for 24 hours, or for the same time to water acidulated with acetic acid, in the light. The sections were then washed well in water and placed in formic glycerin, or if intended for imbedding, the material was transferred gradually to absolute alcohol and thence to paraffin. By this method the contents of the filament stained deeply and had the vague appearance of being made up of numerous short rodlets. The limiting layer remained colorless. This hint as to the nature of the filament was successfully supplemented by the use of Stras- burger's method for differentiating fungous hyphae in the tissues of the host. She found the best method of treatment to be as follows: Sections hardened in alcohol (best without previous treatment with chromic or osmic acid) are placed for about two hours in alcoholic potash (one part 5 per cent potash to three parts absolute alcohol) and then passed into Eau de Javelle for 10 minutes. From no BACTERIA IN RELATION TO PLANT DISEASES. this solution they are transferred to the dye which is prepared by mixing an alcoholic solution of aniline blue with orseillin, drop by drop, until a violet solution is obtained. The mixture is acidu- lated with a few drops of glacial acetic acid. The sections remain in the stain for two hours and are then transferred to dilute glycerin and finally mounted in glycerin. By this method the rodlets were plainly differentiated. Where swellings on the filaments occur these rodlets are very numerous and finally the tube bursts and the rodlets are liberated into the cell-cavity. The bursting of the filaments, or tubes as Miss Dawson calls them, is a normal phenomenon. A transverse section of the nodule showed a filament crossing the cell-wall, the figure given resembles an ordinary sieve-plate, but the relation of the bacteria to the plate is rather obscure. She thinks the rodlets actually pierce the wall, absorb- ing only the middle lamella. Further confirmation of the general results was obtained by staining with methvl violet and fuchsin, though the former method was the more successful. Fig. 36.* In some cases the filaments were in close contact with the nucleus but she did not find this relation constant. She says: "In sections of older tubercles the thicker filaments crossing the cortex are no longer to be seen but those in the main tissue of the tubercles persist until decay has set in." She says further: "The tube, therefore, is actually formed by the parasite as it grows down the hair, and do^s not arise from the plasma of the host plant." A variety of opinions exist as to the presence and constitution of a membrane bounding the filaments. This author maintains that her results confirm Marshall Ward's claim that a membrane is present, but she failed to detect in it cellulose or chitin. The presence of mucilage she considers doubtful. She used Wisselingh's method for the detection of chitin. This is as follows: .Sections of alcholic material are heated in concentrated potash to 160° C. for two hours. After cooling they are *FlG. 36. —Longitudinal section (9) of root-nodule of Pi\:i'n sativum stained with aniline blue and orseillin, showing rodlets of Bact. leguminosarum within filaments. Longitudinal section (10) of root-nodule of Pisum sativum stained with methyl violet and fuchsin, showing liberation of rodlets from filaments. After Maria Dawson. ROOT-NODULES OF LEGUMINOSAE. 1 1 1 washed in 90 per cent alcohol and then stained with iodine and sulphuric acid. If chitin be present a beautiful pink stain is given to the hyphae while the cells of the host take on the usual blue color of cellulose. Staining the infection tube within the root-hair shows it to consist of a chain of rodlets like those found in the filaments within the nodule. The growing point of these filaments is, as Frank asserted, a diffuse open end. In fresh material this open end generally shows a rosette of refringent granules, suggesting the exudation of a ferment by the contained organism. In plants growing in ordinary soil only one infection tube was found entering each nodule, while among those experimental!}' infected, several tubes from as many root-hairs often entered the same nodule. The bacteroids of all species examined were the same in character, consisting of small straight X -shaped or Y-shaped rodlets which stain very readily. At the close of the vegetative period the older nodules are empty sac-like bodies, devoid of bacteroids, but containing a few straight rodlets and some proteid bodies. This observation she thinks supports the theory that the bacteroids have been absorbed by the plant along with any nitrogen contained in them. The characters thus far determined are opposed to the view that this organism is one of the higher fungi. The mean size of the rodlets is given as o . 99 X 3 . 3/z. Experiments to determine the life history of the rodlets were undertaken by Miss Dawson. She secured pure cultures and by dilution isolated them in drop cultures for continuous microscopic investigation. To secure pure cultures large nodules were washed with mercuric chloride and alcohol, then with distilled water and cut across with a sharp razor. Streak cultures were then made on slant tubes of gelatin with a sterile platinum needle which had pierced the cut surface. From such streaked cultures unmixed cultures were obtained by a series of plates, and slant tubes were then infected for future use. The multiplication of the rodlets by division (2 to 4 hours) was successfully followed in hanging drops, but in no case was the formation of bacteroids seen. The organism is aerobic. An attempt was made to grow the organism on dead roots. Pea seeds were germinated between layers of cotton wool till the radicles were an inch long, then dropped into sterile tubes containing wet plugs of cotton and steamed in a water bath for ten minutes. After cooling the roots were infected with nitragin and kept in the dark. In 10 days good growth was obtained, seemingly of the organism sought, but attempts to get pure cultures failed because of the presence of liquefying bacteria. Tests were then made of the ability of nitragin to produce root-nodules. Inoculations were made according to directions, both by rubbing the seeds with the nitragin rubbed up in water and by pour- ing such water over the soil where the seeds were to be planted. Sterile water and utensils were used. No attempt to sterilize seeds before sowing was made and check experiments, she says, justified this, showing that the Leguminosae are not hereditarily infected with the nodule organism. Results from inoculations were in all cases positive. In 4 out of 6 experiments the controls remained free from nodules, while in one of the inoculated sets the entire 20 plants developed nodules. The nitragin from Pisum and Vicia was apparently identical in action and the latter when applied to seeds of Lathyrns aphaca produced a considerable increase in positive results in comparison with untreated plants. A similar increase resulted from the use of the nitragin supplied for Onobrychis and Lupinus upon seeds of Vicia hirsuta. The appearance of nodules on controls illustrates, she says, the difficulty of keeping soil or sand free for many weeks from this ubiquitous organism. [In all cases an attempt at least should have been made to sterilize the surface of the seeds. If controls become affected how then are we certain what caused the effects produced in the inoculated plants?] Miss Dawson agrees with Zinsser that the bacteroids do not occur in the aerial organs of the plant or elsewhere in it, except in the nodules. Zinsser attempted direct infection of the roots under conditions which could be observed, that is, by injecting the organism into the tissues and by stroking the rootlets with needles dipped in the inoculating material. His results were in both cases negative. Since infections in nature occur always through the root-hairs Miss Dawson used external appli- cations only. Seedlings whose roots were infected by drops of water containing nitragin or were dipped entirely into the solution grew vigorously but gave negative results. This suggested that either the organism must pass through the soil, or that infection is impossible after the root has grown beyond a certain stage. Further experiments showed that the second hypothesis is the correct one, since placing the bacteria on the radicle shortly after germination gave very positive results, in one case fully 27 root-hairs side by side showing infection tubes. In all cases infection resulted within 12 days of inoculation. The question is undecided as to conditions regulating the entrance of these organisms, since full- grown hairs often show tubes just beginning growth, while infection of the root-hairs is perfectly easy and certain if the organism is placed on roots that have not yet formed hairs. She thinks that the 112 BACTERIA IN RELATION TO PLANT DISEASES. infection often takes place in the root-hair just as it is emerging from the root. In one case she observed several very small root-hairs, scarcely larger than the cell from which they arose, pene- trated bv infection tubes which had already reached and entered the outer layer of the root-cells. Further experiments were made to determine the possible inhibiting effect of carbon dioxide collecting about the roots when seedlings were grown in tubes : These were negative. Caustic potash was introduced into some of the tubes to absorb the carbon dioxide. Plants flourished equally well with potash and without potash, the only positive result occurring on a seedling in a tube without potash. To test the effect produced by changing conditions under which plants were growing, seed- lings grown one week in sterilized sand with nutrient salts were removed, infected with nitragin and fixed in tubes. Others were germinated in the same manner, but returned to fresh sand after inocu- lation. Others were germinated in tubes, and when the roots were two inches long were inoculated and again fixed in tubes. Three weeks after inoculation the most positive results were where condi- tions had remained as far as possible unchanged. Examination of the nitragin and of fresh subcultures therefrom showed it to " consist of immense numbers of very minute bodies, scarcely longer than broad, all non-motile and similar in size and shape. No trace is found of the variety of shapes exhibited by the bacteroids." In 1900, Maria Dawson published "Further Investigations on the Nature and Functions of the Nodules of Leguminous Plants," from which I abstract as follows: Phascolus shows no nodules for at least 3 weeks after germination, and these are confined almost entirely to small lateral roots. Large nodules contain a considerable quantity of starch. Situated from one to three cells below the surface of the nodule is a layer containing large crystals of calcium oxalate. A similar layer was found in Desmodium. Bacterial filaments strictly comparable to those in Pisum, were found in small nodules, never in those larger than a pin's head, and only once was an infection tube seen within a root-hair. In this genus root-hairs are few in number. These results suggest that in Phascolus the germs in the absence of root-hairs can enter the host directly across the piliferous layer, and that within the root they can continue their growth for a while with or without the formation of a filamentous structure. Acacia agrees with Phascolus in having filaments in very young nodules but not in the older. In this as well as in Phascolus it is possible that we have an intermediate stage in the adaptation of the parasite. A detailed study was made of the nodules of Desmodium gyraus and pure cultures were made of the organism concerned in their formation, since this was of an unusally large size (1.3X3 to 7/x). Similar large forms occur in Acacia, Flemingia, Carmichaclia, Corouilla, and Psoralca. In section, the nodules of Desmodium resemble those of Lupinus and Phascolus, but a new feature was noted. This was the presence in material hardened in absolute alcohol of bright, apple-green bodies, one, as a rule, in each cell. The nature of these bodies has not as yet been determined. They are promptly and completely soluble in 5 per cent potash. These bodies occur also in the nodule-cells of Robinia pseudacacia and in both cases digestion in gastric juice caused the green color to become more con- spicuous. Gastric juice was also found useful for rendering the bacterial filaments conspicuous in sections of Pisum, Vicia, and Robinia. Upon Cassia roots she did not observe the formation of nodules. The older roots of this genus are jet black, contrasting strongly with the pale, greenish-yellow root tips. The author discusses further the biology of the bacteroids. The time required for the growth of a lateral branch in hanging drop cultures averaged about 1.5 to 2 hours. She isolated the nodule organism by a series of separations on tube and plate cultures, and from pure cultures on gelatin microscopic preparations were made. A triple series of these cultures was kept under observation and referred to in her descriptions as A, B, and C. They were: (A) Organisms from sub-cultures of commercial nitragin for Pisum sativum (B) Organisms cultivated directly from the nodules of Pisum sativum. (C) Organisms cultivated from the nodules of Desmodium gyrans. Gelatin plates made from nitragin yielded what appeared to be pure cultures, the colonies looking alike. Further studies led the author to consider the "Nitragin" examined by her as a bacterio- Iogically pure culture. She says: " The general characters of the three organisms are alike, though small differences are noticeable in aggregate cultures. They all grow readily on gelatin, or agar, containing a decoction of pea stems and lr;i\ es, asparagin, and a small percentage of sugar, and giving a very faintly acid reaction." ( In bioth-agar no growth occurred at 200, or at 250 to 300 C. On broth-gelatin at 200 C. growth was extremely slow. Beyerinck also states that his Bacillus radicicola grows very slowly in meat- juice peptone gelatin. No change occurred in milk kept for 3 weeks at 15° C. either in consistency or litmus reaction. root-nodulHs of leguminosae;. 113 "For the purpose of a close comparison of 'nitragin' with organisms direct from the nodule, grown on gelatin, a double series of tubes (gelatin ioper cent; asparagin 0.25 per cent; saccharose 1 per cent ; pea extract) were infected with cultures A and B respectively and kept under the same conditions at ordinary temperatures. From these at intervals of 24 hours, preparations were made, and stained with carbol fuchsin." The nodule bacteria grow most rapidly on gelatin at 150 to 18° C. and on agar at 300 to 350 C. The microscopic characters of the organisms in both cases were quite similar, but those of A (nitragin), after 24 hours' growth, had enlarged to nearly twice their former size. After 48 hours' growth, the maximum size was reached with a few X and Y forms present. The size gradually diminished until after 5 days the original size was reached. No X and Y forms were seen in the last preparations of either type. In drop cultures of the organism from Desmodium, colonies 8 to 10 days old consisted of numerous small rodlets, with some long rods and intermediate stages. In all three types, the formation of a typical colony was observed in a hanging drop of 5 or 2.5 per cent gelatin. Within 5 days the colony reached its maximum size (28/* in diameter). From this time it slowly disintegrated when it was obvious that many X and Y forms were present, the latter predominating. Some rods were curved, others straight. Several individuals were in turn observed for the formation of bacteroids. The fact that X and Y forms arise as a distinct branching of the rods was repeatedly demonstrated (fig. 37). In 14 days the branched forms had disappeared from 2 45pm 3 30pn ft S 45pm 20'C 11 am I 545pm S »,»*. dh U 3 50pm (j £%/ 4 30pm Jj S40pn Jl 11 30 a 1 c 0 9 pm 77°C 11 30pm 0 tfomm ; /mm 11 * occ 2 May 10 \ S-30am 10 am n '130 IS 'C " 4 30pm 15 'C H « // 30 aw 50pm e ^ <^ threads, individual joints of which are readily demonstrated by the use of fuchsin. He states that he observed threads from 10 to 40^ long. Most of these were not straight but wavy and bent. They form commonly an interwoven, felt-like layer. The growth of Leptothrix threads due to unfavorable conditions usually precedes spore formation. In such threads there is a row of spores, while in the single vegetative cells which do not grow out into threads, there are always only two spores in a cell, one at each end (fig. 39). The spore formation begins with the appearance at each end of the cell of a small, bright dot, which gradually increases in size, becomes bounded by a sharp contour and is finally converted into a true spore. These spores are always round and their diameter never exceeds the thickness of the cell. The figure borrowed by von Freudenreich represents not spores but germinating spores. The shortest cell observed with two polar spores measured 3^. Most of them were 6M long. The longest seen was 20^1. He was never able to find any cross-wall separating the two spores, not even when he used Hartnack Imm. X. He, therefore, concludes that the two spores are certainly inclosed in one cell. He could not make out in the vegetative cells whether the spore formation was brought about by free cell-formation or by cell-division. On the contrary, in the Leptothrix threads he found a plain cell-division. The round free-lying spores reach a diameter of 1 ii. The germinating spores swell up to a diameter of 1 .6^. He was able to observe the germination and has figured it, but it is not perfectly clear from his statements whether these germinating spores were those from the Leptothrix threads, or those from the motile organism or from other non-motile short rods or whether they really had anything to do with the organism concerned in the kefir sym- biosis. After considerable discussion of the views of earlier writers on the systematic position of the bacteria, he describes his organism, Dispora caucasica as follows : "Vegetative cells in the form of short cylindric rods, 3.2^ to 8mXo.8m. In zoogloeae condition the cells form white compact elastic clumps of considerable size (up to 5 cm.). The motile vegetative cells have at one end a thin thread-like, wavy nagellum. The spores are round. Lying in the cells they do not exceed the breadth of the latter. When they are free they are 1 n in diameter. The round spores are always arranged two in a cell, one at each end." The kefir clumps do not appear to lose power of growth by drying. They shrink considerably, become dirty brown and stone hard, but are able again to resume their activity when thrown into milk, and are preserved by the mountaineers in a dry condition for a long time. The author himself preserved them in an air dry place for two months, and after a few days, when thrown into milk these could not be distinguished from fresh clumps nor was there any perceptible difference in their power of fermentation. Under the microscope the dry clumps showed a considerable number of changes. Many of the yeast-cells were dead and those which remained alive were principally spherical. These dried ones contained no spores. The Dispora when in spore condition is said not to be destroyed by boiling for an hour. The foregoing is the substance of Kern's paper in the Bulletin de la Societe Imperiale des Natural- istes, Moscow, 1SS1. He sums up his conclusions as follows: (1) The little clumps, the ferment of the Kephir, afford an interesting example of a symbiotic life — commensualism (?) — of yeast-cells and bacteria. (2) The yeast-cells are to be considered as the ordinary beer-yeast, Saccharomyces ccrevisiae Meyen. (3) The bacteria, in the vegetative condition scarcely to be distinguished from Bacillus subtilis Cohn, may, on the ground of very peculiar spore formation, be set off into a new genus, near the genus Bacillus — Dispora caucasica, nov. gen., nov. sp. (4) A distinct cell-membrane can be distinguished on the vegetative cells of the Dispora. (5) The motile cells of the Dispora have a thin, thread-like, wavy flagellum at one end. (6) Moreover, the little clumps, but especially the vegetative cells and the spores of the Dispora, are very resistant to unfavorable influences. This paper is followed by two tables of figures. From them I have borrowed the two figures 38 and 39. *Fig. 39. — Spore formation in Dispora caucasica, and also mature and germinating spores (25). After Kern. In his fig. 2$ at p, p, are masses of protoplasm which he states he observed to break up into two spores. In his figure 24 are a group of vegetative cells provided with a spore at each end and destitute of any cross-wall between them. In his fig. 23, one, two and three are said to be stages following each other in spore development. Fig. 39.' 158 BACTERIA IN RELATION TO PLANT DISEASES. Lewton-Brain and Deerr have published a figure strikingly similar to Kern's figure here listed as 39 (The Bacterial Flora of Hawaiian Sugars, Bull. 9, Exp. Sta., H. S. PI. Asso. Honolulu, 1909, p. 21, fig. 17). This was drawn from their Bacillus D, a sugar destroying species. Is is called by them a curious bipolar effect produced by carbol fuchsin staining. [This] consists of a faintly stained central part, with a very brightly stained circular body at each end. * * * It seems possible that the central body represents the spore, the two brightly stain- ing bodies the degenerated protoplasm of the remainder of the cell, while the walls of the cells have swollen up and become confluent with those of other cells lying close at hand. This peculiar effect was not met with in the other bacteria stained in the same way, and was always obtained with D, so that it would appear to have some diagnostic value. Beyerinck devoted a paper to this subject in 1889. Kefir is composed of a yeast and a schizomycete in symbiotic relationship, the result of their combined action on milk being alcohol, carbon dioxide and lactic acid. He distinguished the yeast as a new form, Saccharomyces kefyr and described the schizomycete as Bacillus caucasicus. These to- gether produce small plates which grow by the formation of local excrescences that increase in size and fuse at their base. In sour milk the kefir grains remain alive for a long time. He figures the yeast as occurring in a uniform thin layer on the outside of the grains. He also states in the text that the yeast occurs almost exclusively on the surface of the grains, the bulk of the mass being bac- teria. In some instances, however, he states that he saw chains or layers of the yeast in the interior of the mass. In the bacterial mass Beyerinck distinguished a cortical layer, and a so-called pith, with a central cavity partly or fully occupied by zoogloeae masses. The yeast is oval. The measurements given by Beyerinck are 3 to 6m- It is easy to cultivate. It is able to convert milk-sugar into alcohol and carbon dioxide. After a long time it liquefies neutral or feebly alkaline lactose gelatin. This yeast tolerates a large amount of lactic acid (up to one-half normal), acetic acid on the contrary is very injurious to it. The yeast is said to invert milk-sugar by means of an enzyme, lactase. That the inversion pre- cedes the formation of alcohol was shown by the use of poured plates containing fish bouillon, 3 per cent sea-salt, and 7 per cent gelatin, to which was added some milk-sugar and then sown with his Photobactcrium phosphor escens. After 2 days the gelatin became luminous. Later, as the food was exhausted the luminosity diminished, this organism being unable to use the milk-sugar present in the gelatin. If then the kefir yeast was placed on portions of the plate these portions again became luminous, indicating the liberation by the yeast of simpler, assimilable sugars from the milk-sugar. The bacillus produces lactic acid. It is mixed with other bacteria which are to be regarded as impurities. Some of these impurities are readily distinguished, i.e., the bacilli producing carbon dioxide and hydrogen, or lactic acid milk ferments in the form of diplococci, Oidium lactis, and foreign yeasts. Other rod-shaped bacteria are less easily distinguishable, such as those producing acetic acid or lactic acid. The advantage of the symbiosis to the yeast is freedom from acetic acid and from putrefactive bacteria, which are assured to it by the presence of the lactic acid ferment. The advantages to the bacteria from the presence of the yeast are less clearly evident, but are believed by Beyerinck certainly to exist. He offers some hypotheses, but no actual facts. A good culture medium for Bacillus caucasicus is gelatin with serum of milk either neutral or feebly acidified with lactic acid. When such a medium is sown with water in which a fragment of kefir has been crushed, there appear after two or three weeks, between the yeast colonies, very minute and extremely slow-growing colonies of this lactic ferment. The most favorable temperature for the growth of these minute colonies is said to be 450 C, and then, of course, agar plates must be used. Bacillus caucasicus does not liquefy gelatin. It forms neither lactase nor invertase, but trans- forms milk-sugar, cane-sugar, maltose and glucose directly into lactic acid. The optimum tempera- ture for this change is 40° to 45°C. There is no special action on starch or casein. The fermentation and formation of lactic acid take place equally well in the presence or absence of free oxygen. "Under all conditions the ferment keeps the form of rods or filaments, which often remain united into chains and may become very long. I have never observed the least indication of spore formation, nor of motility." Beyerinck was not able to obtain the kefir synthesis by using a mixture of the yeast and lactic ferment on lactose gelatin. In 1891, Mix published a contribution upon an American kefir. The work was done at Harvard University on material obtained from two different sources, i. e., New Jersey and Ontario. Both specimens were lobed and fissured and of a dirty brown color resemb- KEFIR. 159 ling dirty gum arabic. That from New Jersey had been dry for more than two years but revived when placed in a nutrient fluid and most of the studies were made with this New Jersey form. The yeast in this kefir Mix states to agree with Saccharomyces kefir, and to differ from Saccharo- myces cerevisiae. It agrees with Beyerinek's form in measurements, in being associated with a rod- shaped schizomycete in a granular mass, in being able to ferment lactose, but not saccharose, and in not producing spores. "Although I cultivated it in saccharose solutions of all strengths, it never caused a trace of fermentation." It ferments milk. The cells of the yeast were of various sizes and shapes, from spherical to elliptical, the spherical ones measuring 3.2^ to 6. 4m in diameter, the elliptical ones varying from 3. 2m to 9.6m in the major axis by 3. 2m to 6. 4m in the minor axis. The bacteria are described as short symmetrical rods, varying from 8. 5m to 4. 5m by o.8m, precisely agreeing with Kern's measurements. "The cells increase by splitting perpendicularly to the long axis, the resulting cells being some- times joined together, thus producing leptothrix-like threads of all lengths, even to 120m, and some- times completely separated. Many of the isolated cells possess the power of motion, but after repeated efforts I was unable to demonstrate the presence of cilia." He states it is not easy to induce these bacteria to produce spores, but that he was able to observe spore formation by placing a clump of the yeast [Kefir grain] in a watch crystal with a little water and covering the whole with another crystal. In 24 hours the threads began to form and within 36 to 48 hours the spores appeared. " It will be remembered that Kern gives two distinct methods of spore formation — one occurring in isolated cells, the other in the leptothrix-like threads. * * * "My investigations on the North American form have led to results diametrically opposed to those of Kern. First, I found but one method of spore formation; secondly, I found this method occurring only in the leptothrix-like threads, although I sometimes found isolated threads bent or curled in such a manner that spore formation was well simulated. Spore formation in the lepto- thrix threads takes place as follows: At each end of each cell of the thread a small bright dot appears. It becomes brighter, larger and much more highly refractive than the rest of the cell until finally it assumes a well-defined spore wall and develops into the mature spore. Each cell has produced two spores, one at each end, and each originating independently of the other. In no case did I see two spores formed, as Kern states, by the division of a single agglomerated mass of protoplasm into two portions." With this kefir-like substance Mix obtained alcoholic fermentation of milk with the formation of carbon dioxide and lactic acid, and the production of a fermented milk closely resembling the descriptions of kefir. "The milk does not sour in the ordinary sense, for it does not coagulate in large masses; still it is acid, contains some carbonic acid gas and alcohol, and is by no means unpleasant to the taste." Mix further states that the North American form of kefir causes (1) alcoholic fermentation of milk sugar; (2) the alcoholic fermentation of dextrose, and (3) that it does not cause the fermenta- tion of cane-sugar. He thinks that the alcoholic fermentation of the milk takes place in the following manner : "The Bacillus acidi lactis begins the process by forming some lactic acid, which in turn, assisted by the bacillus itself, inverts the milk-sugar to galactose and dextrose. The galactose is further acted upon by the Bacillus acidi lactis, and converted into lactic acid; the dextrose is acted upon by the yeast, and converted into alcohol and carbonic acid gas. In the kefir drink, therefore, we should find plenty of lactic acid, a little milk-sugar, not inverted, the amount depending upon the duration of fermentation, some alcohol, and carbonic acid gas — precisely what is found." In 1896, Ed. von Freudenreich published a paper on kefir, of which the following con- densation includes the most important statements: From his own experiments which are in harmony with those of all previous observers he concludes that Beyerinck did not have the kefir yeast because this yeast is unable by itself to ferment milk- sugar. From Beyerinek's drawings and from the trouble he had in isolating the Bacillus, it seems probable to von Freudenreich that Beyerinck had under observation the same bacterial organism that von Feudenreich has studied and which he is inclined to consider Kern's Dispora caucasica, with, however, a considerable number of reservations as to its morphology. He thinks that the motile one-flagellate bacteria described by Kern probably had nothing to do with the Dispora. Other observers he states have come to the same conclusion, for example, Adametz. The Bacillus subtilis which frequently occurs in the kefir grains has nothing to do with the kefir fermentation according to l6o BACTERIA IN RELATION TO PLANT DISEASES. Essaulof, with which conclusion von Freudenreich agrees. Essaulof believed that only Bacillus acidi lactici and the yeast were necessary to the formation of the symbiosis. Von Freudenreich was unable to obtain conclusive evidence on this point. He found other lactic acid bacteria and suggests that possibly the same micro-organisms do not always occur in kefir. "The yeast and Kern's Bacillus are always present but the lactic acid bacteria may possibly be different if only they can bring about the splitting up and fermentation of the milk. If one sums up the results obtained hitherto we have in kefir an example of the symbiosis of several micro-organisms, among which is a yeast that according to most authors is not able to ferment milk-sugar, as well as probably a lactic acid ferment and a bacillus hitherto only cultivated by Beyerinck, which appears to be identical with the bacillus present in the kefir grains and described, but not cultivated bv Kern. On the other hand, up to this time the role of these particular micro-organisms in kefir fermentation has not been clearly made out." His own experiments began in 1892 and were continued with interruptions up to the time of the publication of his paper. In his preparations he found especially yeast-cells and long, mostly bent bacilli, very much resembling Kerns' pictures, but also shorter rods — younger stages of the bacillus — and furthermore, coccus forms, the latter, however, much more rarely. Often he found that the bacillus stained only at the two poles, a phenomenon which he thinks led Kern into error as to the presence of spores in his Dispora. He thinks that the sporogenous organisms occurring in Kern's cultures were only potato bacilli and similar bacilli. "I have never observed spores in the bacillus of the kefir grains, Dispora caucasica, therefore, I would write Bacillus caucasicus." When the kefir was clean, von Freudenreich found four different micro-organisms in it, namely, yeast cells, large coccus forms arranged in chains, smaller cocci and bacilli. The larger streptococcus and the yeast grew readily on gelatin plates, sometimes also the smaller streptococcus, but not the bacillus. Only once did he obtain colonies of the latter on an anaerobic gelatin plate. On the surface of milk serum agar plates at 35° C. one readily obtains the smaller streptococcus along with the larger one and also very small colonies of a bacillus believed to be identical with Bacillus caucasicus. Nevertheless he says that such cultures do not always succeed. Sometimes its colonies were entirely absent without any reason therefor being apparent. When he made streaks on slant milk serum agar, using kefir itself as a substance for inoculation and keeping the tubes at 220 C. he states that he obtained masses of growth containing the four organisms mentioned, and he figures the microscopic preparations of such mixed cultures, but these figures are not very conclusive as to any symbiosis. He says also that Bacillus caucasicus may be isolated by stab-cultures in deep layers of agar, these cultures being kept at 350 C. In the stab then appear mostly only the bacillus and the small strep- tococcus. He describes the yeast as follows. Saccharomyces kefir is obtained readily in plate-cultures where it produces very small, coarse-grained pale colonies. On milk serum gelatin plates the colonies are said to be round and yellowish and better developed than on ordinary nutrient gelatin. The granulations on the edge of the colony are coarse. On the less thickly sown plates, the superficial colonies are well formed and whitish, finally yellowish. The buried ones are a yellowish color. The center of the colony is dark brown. Stab cultures were distinctly visible in 24 hours. Development on the plates also was rather prompt, the colonies being visible for the most part after 2 to 3 days. Beef-broth kept at 200 C. clouded in 24 hours, also milk-sugar bouillon. The growth, however, is not so vigorous as in beer-must. In the latter medium there was an abundant growth. Gas-pro- duction occurs but is less abundant than in case of beer yeasts. Maltose is fermented by the yeast. The yeast ferments grape-sugar with production of alcohol. It does not cause any fermentation of milk, but develops well in it with the formation of a peculiar taste, which is different from that due to beer yeast. To the eye the milk remains unchanged. On potato the yeast grew with the formation of a yellowish patch. The optimum temperature is about 220 C. The yeast will growat 280 C, but not at 35° C. This yeast consists of oval cells of variable size, on an average 3 to 5^X2 to 3^. Single cells are roundish, especially in potato cultures. The cells stain readily with all the ordinary aniline dyes, also by Gram. There is ordinarily a vacuole. In the protoplasm there are one or more shining granules. The yeast is unlike Saccharomyces pastorianus I, II, and III, and also unlike Saccharomyi es < erevisiae and Saccharomyces ellipsoideus, with which he compared it. There is never any pellicle formed by the kefir yeast, something which always goes with the other yeasts mentioned. He could not discover any ascospore formation. A temperature of 50° C. for 5 minutes sometimes sterilizes the culture, and a temperature of 55° C. always sterilizes it. He found the yeast also very sensitive to dry air. It endured 2 and 3 days' exposure, but not 4 or more days when taken from fluid cultures and exposed on filter paper. I omit descriptions of the two forms of streptococci because most observers are agreed that they only occur accidentally in the kefir. The photomicrograph of his larger streptococcus shows an organism with a long diameter nearly double the short diameter and makes one think that very likely the organism figured is not a streptococcus at all. KEFIR. l6l Bacillus caucasicus is described as follows: On ordinary gelatin plates it does not grow at all. Only once, as already stated, did von Freudenreieh obtain it on a gelatin plate exposed to anaerobic conditions according to Miquel's method. Other times, using the same method, he did not obtain it. Also on milk-sugar gelatin plates he never observed it. Having once obtained it, it grows in stab- cultures even in ordinary gelatin but then first after a long time. On milk-sugar gelatin plates he often had no growths; at other times microscopic colonies. On the surface of milk agar plates, on the contrary, he often obtained colonies. Upon this it produces small, flat, grayish colonies which appear circular to the naked eye. With a weak magnification they are seen to have irregular contours and are not uniformly circular; they also appear whitish and granular. This granulation is produced he states, by the irregular arrangement of the bacilli, plain to be seen on the edge, out of which the bacillary forms project. In ordinary nutrient bouillon he could not obtain any growth, not even at 350 C. In milk-sugar bouillon there was a slow growth at 220 C. — nothing to be seen for the first 3 days, but at 35° C. the growth is faster. The reaction was acid. It produces no coagulation in milk although the reaction becomes somewhat acid. The taste of such milk was slightlyacid and astringent similar to that produced by the smaller streptococcus when grown in milk. There was a moderate gas formation. There was no growth on potato. In milk-sugar bouillon it appears ordinarily as a straight bacillus with rounded end, often with a shining point at each end. This appearance corre- sponds well, he says, to the phenomenon interpreted as spores by Kern. Their slight resistance to heat, however, shows that they are not spores; also when exposed to staining media the bacillus stains in toto which would not be the case if these bodies were spores. The organism stains easily with the common aniline dyes, and also by Gram's method. The breadth of Bacillus caucasicus is about 1 /j, the length 5 to 6/x, but long forms are also found which are then crooked. It is very feebly motile. A good photomicrograph of thisorganism is shown in his fig. 5, table 1 (fig. 40). Its resistance to external influences is slight. It endured drying 1 day. It was regularly killed by a drying of two or more days. Nevertheless, it lives a long time in the kefir grains, which he thinks is explain- able by the fact that it is protected from the action of the air. It was killed, as already stated, by 5 minutes exposure to 550 C, while 2.5 per cent carbolic acid killed it in 30 seconds. Corrosive sublimate 1 : 1000 killed it in one experiment in i, 2, and 60 minutes, but not after 5 and 15 minutes. This contradiction is attributed to dissimilar resistance of individual bacilli. The acid was estimated in terms of lactic acid, but I find no statements concerning its determination. When inoculated separately into milk he could not obtain with cultures of these organisms anything corresponding to kefir. Moreover, with two organisms alone he could not obtain kefir. The yeast and the large streptococcus caused the milk to coagu- late with a small amount of gas formation but no further change. Fig. 40.* The small streptococcus combined with the yeast and inoculated into milk produced gas formation and a sour taste, but no kefir fermentation. The amount of gas formed was variable. This he attributes to decrease in the virulence of the streptococcus. Finally, with the yeast and the Bacillus caucasicus he could not obtain kefir. Also when he inoculated all four of the micro-organisms together the experiment miscarried regularly at the beginning. The lactic acid ferment took place, the milk coagulated, but nothing further happened. Von Freudenreieh states that throughout his studies he had many failures but that finally he frequently obtained good kefir by inoculating milk with mixed growths of the organisms obtained by rubbing kefir grains on slant agar. Usually the first flask of milk inoculated did not yield kefir, but when transfers were made from this to a second flask of milk good kefir was often obtained. He seems to have been more successful in using this source of inoculation than pure cultures of the separate organisms mixed, although on the second or third transfer from milk to milk he states that he also obtained kefir from these. He states that sometimes he obtained a drink which could scarcely be distinguished from kefir by use of the yeast and the two streptococci. He was never able to produce the kefir grains synthetically, and he considers that the role of the Bacillus caucasicus is still involved in a good deal of uncertainty. Its presence seems absolutely necessary to the symbiosis, but just what its function is in the fermentation he does not know. Podwyssotsky (French edition of 1902) says nothing is known respecting the origin of kefir. There are various hypotheses current among the natives of the Caucasus respect- ing it: (1) The kefir grains are the direct gift of God through his Prophet Mohammed, *Fig. 40. — Bacillus caucasicus. From a photomicrograph by von Freudenreieh. x 1000. 1 62 BACTERIA IN RELATION TO PLANT DISEASES. and hence called "Millet of the Prophet;" (2) the grains were found very long ago in a bush on the high mountains near the eternal snow; (3) the first grains appeared in a dirty milk receptacle (outre). "Cette derniere version populaire se rapproche beaucoup, a notre avis, de la verite." According to Podwyssotsky, the kefir grain is composed of three organisms : The kefir bacterium proper, the yeast, and a third schizomycete which produces the lactic acid. He states that Stanghe" was the first to call attention to the presence of a third organism in kefir. The yeast cells are on the outer face of the grains. The deeper layers consist of a fibrous stroma of bacteria. This author states that healthy kefir grains never contain streptococi nor staphylococci. He is inclined to consider the the kefir bacteria as related to Bacillus sublilis (descended from it). He states that the yeast re- sembles Saccharotnyces cerevisiae in its action. Moreover, if the alcoholic action of the kefir in milk is not proceeding properly it may be hastened by the addition of ordinary beer yeast. Podwyssotsky also refers to the fact that there appear to be two types of kefir grains ; a coarse large form which comes to the top of the milk during fermentation, and a smaller grained form which lies at the bottom. These have the same action on the milk. The grains which occur at the bottom of the milk break apart more easily when pressed between the fingers and are not as elastic as those which float. The kefir grains, especially as brought into the market dry, are often attacked by other organisms e. g., Oidium laclis, Penicillium glaucum, coccus forms, and various rod-shaped bacteria. A cursory inspection of these grains is often sufficient to show that they are diseased by these extraneous organ- isms, the surface of the dry grains being covered with white spots. If the grains are dried slowly in a moist and shady place, they often become very moldy and exhale a characteristic and very dis- agreeable odor. He recognizes especially two diseases of the kefir fermentation: (1) Muciiication of the grains due apparently to the multiplication of foreign bacteria, the yeast cells being destroyed, and spherical and long filamentous bacteria becoming abundant. This is believed to be a contagious disease since, if a single affected grain occurs in a mass of grains, there will be many others after some days. (2) A butyric acid fermentation which may be readily detected by the peculiar penetrating odor, resembling that of rancid butter. Moreover, microscopic examination of a drop of the fer- mented milk shows the presence of a great number of bacteria with swollen ends while the yeast cells have here also disappeared. On account of the prevalence of extraneous molds and bacteria, kefir grains designed for sale dry should be washed thoroughly in several waters, i.e., until the water comes away clear, and then dried rapidly in the sun on linen or filter paper. Podwyssotsky states that the kefir ferment may be obtained in various places in Europe in the form of tablets and powders. These are not so efficient as the kefir grains, but by several transfers through milk the kefir ferment may often be obtained from them in an active condition. The first product is usually too acid and does not contain enough carbon dioxide and alcohol. Much of the kefir on the markets in Russia is contaminated by butyric acid organisms and is of very inferior quality. Many of the kefir grains offered for sale are also of this character. The best temperature to obtain a suitable kefir fermentation of milk is stated to be 150 to i7°C. At temperatures of 250 to 30° C. the lactic acid fermentation is too intense and only insignificant quantities of alcohol and carbon dioxide are produced. Frequent agitation of the receptacle con- taining the fermenting milk is considered to be very desirable; more so, even, than in the case of Kumys. The author states further that the inoculated milk should be left open to the oxygen of the air for the first 6 to 8 hours, then closed tightly and the fermentation allowed to continue for 1 or 2 days. The finished kefir should contain about 0.7 to 0.9 per cent lactic acid, a small amount of pep- tone, 1 .5 per cent cr less of alcohol and considerable quantities of carbon dioxide. Kefir more than 5 days old should never be consumed. A drop of good kefir 2 days old under the microscope should contain some yeast cells, considerable numbers of kefir bacteria, numerous minute lactic acid bacteria, a fine deposit of precipitated casein, and fat drops of various sizes. Kefir grains moistened and rubbed upon a slide should show under the microscope yeast cells, large bacteria (the specific kefir organism), and smaller lactic acid bacteria. In a drop of kefir 8 days old, the lactic acid bacteria are very abundant and the yeast cells have entirely disappeared. THE GINGER-BEER PLANT. The following account of the ginger-beer plant, and the organisms composing it, is condensed from H. Marshall Ward's long paper in the Transactions of the Royal Society of London. THE GINGER-BEER PLANT. 1 63 When seen in the fresh state, as it comes from flasks or other vessels, the ginger-beer plant pre- sents the appearance of solid, white, semi-translucent, irregular, lumpy masses, not unlike pieces of soaked sago or tapioca; these lumps are brittle, like firm jelly, and their size varies from that of a pin's head, or smaller, to that of a large plum, or larger. Opacity and brittleness vary, even in the same lump. Fresh-dried lumps do not dissolve in water, even if boiled. When thoroughly dry they are often hard and horny. Fresh moist specimens are usually distinctly acid, though in varying degrees. The most striking characteristics of these lumps of ginger-beer plant become evident only when they are placed in saccharine solutions. After some days in a closed soda-water bottle three- fourths full of Pasteur's fluid, a lump of ginger, and a few lumps of the ginger-beer plant, kept in a warm place, the liquid is found to be very turbid and more or less viscous. The fermentation goes on rapidly. Much gas is produced and the container may explode if tightly closed. In time the viscosity increases, and it sometimes happens that the liquid becomes so thick that the gas-bubbles rise slowly. Viscosity is not due to the mere presence of yeast-cells, because they fall to the bottom, but to the presence of innumerable swollen or slimy vermiform bodies distributed through the mass of the liquor. Myriads of rod-shaped bodies (bacteria) are also observable. The increasing deposit below is also found, in later stages, to consist of bacteria, swarming amongst the yeast-cells. The "ginger-beer" is distinctly acid, as well as viscous. As time goes on, the surface of the liquid usually becomes covered with a dense scum, unless very well corked and protected. The problems then which present themselves are: What is the yeast which so rapidly spreads in the earlier stages of fermentation? What are the slimy vermiform bodies in the liquor? What species of Schizomycetes are present? What does the scum consist of?, and finally, What have all, or any, of these organisms to do with the ginger-beer plant, and the con- version of the saccharine liquor into "ginger-beer"? In an effort to solve these problems, almost two thousand separate cultures, each extending over periods of from several days to months, and even in some cases to two years, were made. These cultures were of three kinds: (1) Large cultures in flasks, usually liquids, sometimes solid gelatin; (2) smaller cultures in tubes; and (3) cultures in hanging drops, made in sterilized cells under the microscope. Every piece of apparatus was heated in a hot-air chamber to at least 140° C. for 2 hours, and everything was lifted by forceps, similarly treated. Various Organisms Found in the Ginger-Beer Plant. It was apparent from the start that the ginger-beer plant is a body, consisting of several organ- isms, or, at least yielding more than one definite organism. Investigation has shown, however, that two specific cryptogams constitute the ginger-beer plant proper, and are necessary for its formation and peculiar action, while the rest are merely accessory or foreign organisms. Of the two essential forms, one is a new species of Saceharomyces, the other a new and very re- markable species of schizomycete. Of two non essential forms, found in all the specimens examined, one is a yeast-like form, Mycoderma ccrcvisiae Desm., while the other is the vinegar organism — Bacterium aceti Kutz. The intruders most commonly met with are species of Saceharomyces, Bacillus, Micrococcus, Oiilium, Torula, Dematium, and one or two ordinary mould fungi, of which Pcnieillium is by far the commonest. The new yeast Saceharomyces pyriformis which resembles 5. ellipsoideus, is the most important one met with in this investigation, being constant in every specimen examined, and undoubtedly the yeast principally concerned in fermentation of ginger-beer. It induces active fermentation in sugar-solutions, either cane-sugar, or glucose [no statement respecting lactose in this place, but elsewhere it is said that milk sugar can not take the place of cane sugar or glucose], resulting in a copious evolution of carbon dioxide gas, and in the formation at the bottom of the flasks, tubes, etc., of a voluminous white pasty deposit, consisting of characteristic colonies of budding-yeast cells. Pure cultures were readily obtained, both by the dilution method, and by growth on gelatin media; cultures were obtained from single cells grown in hanging drops. The single cell is globoid, or more commonly ellipsoid, or ovoid in shape, colorless and translucent, and measures from 6 to 7 m long X 5. 5m broad, though smaller and larger cells are found. The pyriform cells occur in the surface film. There is no limit to the size and shape of the colonies. In very active, vigorous cultures of this yeast, the protoplasm gives a strikingly clear glycogen reaction — on adding iodine dissolved in an aqueous solution of potassic iodide, the cells turn dark sienna red, or red-brown. Ascospores occur repeatedly, and with singular distinctness. They are formed on moist gypsum blocks and also on the surface of gelatin. This species is named Saceharomyces pyriformis from characteristic pear-shaped aerobian cells. 164 BACTERIA IN RELATION TO PLANT DISEASES. Chief characters: A low, or bottom-fermentation yeast, which inverts and ferments cane-sugar. Ordinary cells ovoid or globoid, ranging from 5 to o,/* in diameter, though smaller and larger ones occur. Ascospores formed in from 2 to 4 days, at 25° C. and lower. Aerobian forms, as films, of pyriform, or sausage-shaped cells are developed in wort in 21 days. Occurs in " home-brewed ginger- beer, " and is the predominant form in the so-called "Ginger-beer plant. " Whenever the fermentations were carried on or finished with access of air, a dense wrinkled skin formed at the surface; and, since this occurred when the air had to filter through sterilized plugs of cotton wool, there can be no doubt as to the origin of the fungus from the inoculating material. This growth was identified as the very polymorphic M ycoderma cerevisiae of Desmazieres. It is not a true Saccharomyces for it does not form ascospores, and differs in several respects from the true yeasts, in the narrow sense. It is distinctly an aerobian form; is unable to invert cane-sugar or to bring about its fermentation; it is apt to appear on lager beer even in cold cellars; cells are elongated, less translucent than active Saccharomyces cells; average size, 6 to 8^ long, by 2 to 4^ broad. A pink or rosy yeast Cryptococcus glutinis Fres. ?, also occurred. This organism is not a necessary part of the ginger-beer plant, and is not always present in cultures of this "plant." A small yeast, the cells of which are very nearly spherical, averaging about 2.3 to 3. 7m in diameter [Yeast D) was frequently met with, but is not a normal constituent of the ginger-beer plant. Not sufficient knowledge of its characters to warrant any specific name. The ordinary beer-yeast {Saccharomyces cerevisiae) and two other forms, not identified with cer- tainty, one of which is probably S. apiculatus, were found occasionally. These are not concerned in the formation of the ginger-beer plant. The new schizomyeete called Bacterium vermiforme is a peculiarly vermiform organism, enclosed in hyalin, swollen, gelatinous sheaths, and imprisoning the yeast-cells of Saccharomyces pyriformis, etc., in the brain-like masses formed by its convolutions. It is a constant and essential form — essential because the ginger-beer plant can not exist as such without it. It is the swollen sheaths of this organism which constitute the jelly-like matrix of the "plant. " Fig. 41.* Various attempts were made to isolate this organism — a difficult thing to do at first, because then it was not known what relation the naked forms bore to the sheathed cocci, rodlets, filaments, etc. Two distinct phases of this organism exist — the vermiform, sheathed stage found in acid saccharine media saturated with carbon dioxide, and the motile naked filaments, rodlets and cocci met with in neutral bouillon and other incomplete nutritive media containing oxygen. It was neces- sary to take the precaution of cultivating both forms side by side, under exactly similar conditions, varied one by one similarly for each. In the presence of oxygen the bacteria promptly escape from their sheaths. If these unsheathed rods are grown in the presence of yeasts, i.e., carbon dioxide pro- ducers and oxygen consumers, the sheaths form again. When a small piece of the gelatinous form, i. e., the compacted coils of sheathed filaments, rod- lets, etc., of Bad. vermiforme is put into a test-tube of suitable nutritive fluid (e.g., Pasteur-bouillon, beet-solution, boullion +5 per cent of sugar, etc.) and kept at 150 to 18° C, the usual course of events is as follows: The liquid becomes more and more turbid after 48 hours or so; then a whitish film begins to form above, and a deposit at the edges of the level of the liquid, while a similarly whitish, granular or cloudy looking deposit falls to the bottom. In from 7 to 14 days the rapidly increasing deposit becomes more and more gelatinous, and at length assumes the consistency of a sort of jelly. This gelatinous cushion at the base consists of the sheathed coils so often referred to; the film and ring at the level of the liquid, and the turbidity throughout the body of the same, are chiefly due to the free filaments and rodlets already described as escaping from the sheaths. The preliminary turbidity of the liquid is due to the motile forms of these filaments and rodlets. Flagella could not be demon- strated. The sheaths may grow end on (fig. 41) or sidewise (fig. 42). *Fig. 41. — Two stages of Marshall Ward's Bacterium vermiforme in a hanging drop of Pasteur bouillon stiffened with gelatin. I have omitted intermediate stages b and c. The stage d was drawn 21 hours later than a. After Ward. THE GINGER-BEER PLANT. I65 " It must, therefore, be concluded that this schizomycete is able to live and grow in an acid sac- charine" solution, with suitable minerals and nitrogenous materials, not only in an atmosphere totally deprived of oxygen, but in one of vapor which is so attenuated that it is practically a vacuum so far as permanent gases are concerned — and that only forms its gelatinous sheaths if carbon dioxide is present." (See figs. 43 and 44.) A ginger-bacillus (Schizomycete No. 2.) is frequently met with in fermentations to which lumps of unsterilized ginger were added, but is not essential to the formation of the ginger-beer plant. It occurs on ginger rhizomes. Cultures of Schizomycete No. 3 (Bacterium aceti Kiitz.), demonstrated not only that this bacte- rium is not a normal or necessary constituent of the ginger-beer plant, but also that it can not be induced to form a submerged commensal growth with any of the yeasts. Synthesis of the Ginger-Beer Plant from Pure Cultures. The most conclusive proof of the accuracy of the foregoing studies is afforded by the re-consti- tution of the ginger-beer plant as such, by bringing together pure cultures of the organisms com- posing it, and showing that the specimens so produced act like the original specimens. The other forms mentioned above were tried in various combinations, but only the two essential ones, Sac- charomyecs pyrifortnis and Bacterium vermiforme were successful. The relations between this yeast Fig. 43.t MfftH'fil W!i//'7 Fig. 42/ Fig. 444 and bacterium are those of true symbiosis. The ginger-beer plant forms only in acid media and in closed vessels, or in other conditions in which the free oxygen has been removed and carbon dioxide substituted, e.g., under Mycoderma pellicles. The nutritive substance must also contain carbo- hydrates. Cane-sugar is better than glucose; milk-sugar will not do ; sterilized ground rice may be used in place of ginger — the starch in the ginger being apparently the essential. The schizomycete will grow in the absence of these substances but no sheaths form. The origin of the ginger-beer plant is involved in obscurity, but there is evidence to show that the yeast is introduced from the grocer's shops attached to the ginger and brown sugar employed in ordinary practice, while the bacterium is introduced with the ginger. *Fig. 42. — Splitting of sheaths in Marshall Ward's Bacterium vermiforme: a, condition when fixed under the microscope in a drop of ginger-gelatin; b, 6 hours later; c, 18 hours after b. The corresponding parts are designated by x's. After Ward. fFiG. 43. — The ginger-beer plant, a compact mass due to growth of Bacterium vermiforme with yeast in a suitable saccharine medium. This figure shows appearance after 15 days in Pasteur-bouillon. "The filaments and rodlets ensheath themselves as soon as the carbon dioxide is in excess, and entangle the well-developed yeast-cells in the coils of the gelatinous matrix. The mass becomes denser and denser, and at last forms the hard, brain-like lumps of the ginger-beer plants." After Ward. |Fig. 44. — Bacterium vermiforme and Saccharomyces pyriformis grown in an unsuitable medium in which no sheaths appear on bacteria and no symbiosis takes place. Drawing made after 15 days in ordinary bouillon. "The yeast buds slowly and for a short time only. The Schizomycete grows out into filaments which rapidly break up into very short rodlets (bacteria) and cocci." After Ward. 1 66 BACTERIA IN RELATION TO PLANT DISEASES. In addition to large quantities of carbon dioxide, ginger beer contains during eaily stages of fermentation traces of alcohol and acetic acid, while relatively large quantities of an acid resembling lactic acid, if not identical with it are formed. Just what each organism gains from the combination was not made out clearly, but the yeast seems to do better in the presence of the bacterium than where separated from it. The following pertinent paragraphs from Ward's paper may close this review: "Everything points to the view that the relations between the yeast and the bacterium are those of true symbiosis, because every attempt to feed the schizomycete with dead yeast-cells or decoctions of such, or detect it embracing such cells in a dead or feeble condition has failed. " It is significant that the synthesis of this dual organsim — which is so strikingly like the lichen that we may compare it forthwith with one of the gelatinous forms — was most easily brought about by adding the yeast-cells to already advanced cultures of the bacterium, both having been grown in the same medium and under like conditions. * * * "The schizomycete is favored by obtaining some substance or substances directly they leave the sphere of metabolic activity of the yeast-cells ; it can benefit by the presence of these substances, even apart from the living yeast, though to a less extent. "The yeast, on the other hand, benefits by these substances being removed and destroyed, hence its renewed and continued activity — as evidenced by the steady and copious evolution of carbon- dioxide for weeks, and the corresponding increase of the yeast-cells by budding— when the symbiosis is established. "For the present this can only be regarded as a hypothesis. It might be objected that I have inverted the order of things — that, since the schizomycete is able to evolve small quantities of carbon dioxide daily from saccharine solutions, it may be that its powers are enhanced by the yeast removing inhibiting substances of its activity. The objection is possibly valid, but I think the former hypo- thesis explains most of the facts : How, for instance, is it to be explained that the schizomycete slowly and steadily converts the whole of the liquid sugar-solution into a solid gelatinous mass, if the organism excretes such inhibiting substances?" THE SO-CALLED "BEER-SEED." The writer has had but one opportunity of seeing kefir-like grains. These were sent from Missouri to the Department of Agriculture under the name of "Califor- nia Beer-seed." They were acid to litmus paper, and had a feebly acid, rather agreeable, ester-like odor. The roundish gelatinous granules (fig. 45) consisted of several kinds of bacteria, and of one or more kinds of yeasts, together with oblong, large, jointed threads which stained like yeast and were interpreted as Oidinni lactis. The majority of the yeast-cells were round or roundish and the long oval or elliptical ones appeared to be of a different sort. Morphologically there appeared to beat least three kinds of bacteria in the grains, but no coccus forms were detected. Quite unlike the kefir figured by Beyerinck, the yeast-cells were not in a uniform thin layer on the surface, but were distributed through the grains in little clumps, most of the clumps, however, being in the outer parts of the grains (fig. 46). The greater portion of the bacterial mass consisted of rather thick short threads. Long filaments were exceptional. The appearance of the more common yeast and schizomycete as crushed out in water is shown in figure 47. The general relation of yeasts and bacteria in the substance as determined by thin sections of fresh grains and by microtome sections of embedded material is shown in fig. 48. The writer did not observe any bacterial forms corresponding to the end-splitting ones figured and described by Mar- shall Ward, but prolonged study was not given to the subject. Poured-plates made from the granules on ordinary +15 beef-bouillon agar yielded a considerable number of small, slow-growing white colonies which proved to be a yeast. *Fic. 45. — Grains of "California beer-seed." Received in 1508 from Missouri. BEER-SEED. 167 No bacteria were obtained on these plates. This yeast fermented grape-sugar in peptone water readily and also cane-sugar and maltose, but not lactose. This yeast grew on potato in a thick whitish layer. It grew readily in the thermostat tem ,,-,.■ . ."s * life . -^ - . ">'v"" -"■ ■::.'''■ "'■'-■' - - ' "'■ ■:■"-'''• Fig. 46.* 1 m V® ijt*9fr •MC ' Fig 47-t i ftoosat Fig. 48. % at 280 C. On agar streaks it sporulated freely after some days, the protoplasm, rounding off into two or more parts. Often three spore bodies and sometimes four or more in a cell were observed (fig. 49). The appearance and behavior of this yeast sug- gested Saccharomyces cerevisiae. Sterile litmus milk inoculated by putting some of the grains into it did not redden, did not coagulate at once and did not produce any gas- bubbles. Gradually the litmus was reduced and remained reduced for a very long time. The milk also finally curdled. It was blue after the reduction ceased. It would appear, therefore, that some kefir-like grains contain a yeast incapable of fermenting lactose, while grains from other sources contain a yeast capable of fermenting lactose. The relation of kefir to the ginger-beer fermentation remains to be determined. Fig. 49. § *Fig. 46. — Free-hand unstained section through one of the "beer-seeds," surface of grain being at left. The uniform gray structure represents bacterial zoogloea?, the darker masses the undisturbed position of imprisoned clumps of yeast. fFic. 47. — A small portion of "beer-seed" crushed in water and drawn after staining with carbol fuchsin. The yeasts buried in the zoogloeae were not budding, most of them were globose or nearly so. tFic. 48. — Another crushed-out water mount of "beer-seed," showing both round and elongate yeasts, the con- tents omitted, x and y stained alike with fuchsin and both forms were full of coarse granules. §Fig. 49. — Spores or spore-like bodies in yeasts on agar sub-culture (streak) made from an agar-plate poured from crushed "beer-seed." Age 16 days. Stain, carbol fuchsin. 1 68 BACTERIA IN RELATION TO PLANT DISEASES. Bacteria with Fungi. In 1903, Zederbauer published a paper on the Myxobacteriaceae, in which he claims that he has demonstrated that certain of these organisms are a symbiotic combination of bacteria and fungi. His first investigations were made on a reddish form, which he calls Myxococcus incrustans, growing on sponges used for wetting gummed paper. This form, which looked like a slime mold was made up of bacteria, chiefly, with occasional fungous hyphae, and chains of small round bodies which he thinks were conidia. He succeeded in obtaining pure cultures of the bacteria (some of them, at least) on peptone gelatin. At room-temperature, in the light, they grew rapidly, forming a film of branching radiating chains of bacteria on the surface. In rather old colonies these floated for a time on the liquefied gelatin, then sank to the bottom. Growth occurred also below the surface, but was not so luxuriant. All cultures, in repeated experiments, produced the same bacterium. The color of the Myxococcus is not shared by the bacteria. The colonies are dirty white. Spores were formed in old cultures. No cilia were demonstrated, nor was the movement like that of ciliated forms. On agar, growth appeared similar to that on gelatin, but the agar was not liquefied. The bacteria did not grow on sterilized bread or potato. The fungus was also cultivated separately. Spores taken from the so-called Myxococcus ger- minated in the usual manner, forming several celled hyphae, which soon produced conidia, like the original spores. Mycelium also developed from oidia which were formed by the breaking up of filaments. Although Zederbauer thus cultivated both fungus and bacteria separately, he did not succeed in reproducing the original mixture, the Myxococcus, by growing them together. Cysts, he states, like those described by Thaxter, were found. These were composed of bacteria and chains of conidia surrounded by a common envelope, probably composed of hardened slime secreted by the bacteria. On germination, this bursts and the new organism begins growth. The color of his Myxococcus is not constant. It may be red, pale yellow, and sometimes black. A form which he calls Chondromyces glomeratus was found in several localities growing in groups upon the cut surface of beech stumps which had not begun to decay. The slimy red outgrowths 4 to 5 mm. high, resemble Tremella. This form he says was also composed of bacteria and fungous hyphse. The long slender hyphae, rising from spores at the base, intertwined and formed at the surface a thicker layer of conidiophores bearing chains of conidia. The very small rod-like bacteria which swarmed in the interior were actively motile. Several conidia were germinated in a moist room, forming hyphae, which, however, did not produce conidiophores. The bacteria, which stained with methylene blue and fuchsin, grew on gelatin and agar. On gelatin plate cultures at room temperature or in the thermostat at 2o°C, small dirty white drops were formed which united and liquefied the gelatin in hollows. In cultures kept near the window, growth was strongest near the light. Streak cultures behaved very much like plate cultures. The bacteria did not liquefy agar but formed over the whole surface a dirty white layer, starting from small round flecks. All cultures were fluorescent. The flagella, attached to all parts of the body, were stained with Van Ermengem's stain as modi- fied by Hinterberger. In some cases they were ten times the length of the bacteria. Spore formation was not observed. In gelatin cultures when hyphae were brought in at the point of inoculation they took on irre- gular shapes and seemed to form conidia. Chlamydospores were also formed in such cultures. Zederbauer claims that the origin of these demonstrates that they are fungous spores and not bacteria as claimed by Thaxter. The first part of this paper is devoted to speculations on whether Thaxter's statements and figures can be interpreted as indicating the presence of mycelium in the Myxobacteriaceae. There seem to be a good many uncertainties connected with his own researches. Dr. Thaxter's comment on this paper is as follows: "This treatment of the group, though novel, seems somewhat hasty; especially in view of the fact that the figures and descriptions given in this paper show very clearly that its author is as yet unacquainted with any member of the order he discusses, having been misled by fancied resemblances and influenced no doubt by an exaggerated notion of the difficulties associated with the differentia- tion of rod-like bacteria from Oedocepkalum, Torula, and similar hyphomycetous types. A specimen BACTERIA WITH FUNGI. 1 69 of Myxococcus incruslans (Torula myxococci-incrustantis n. sp. X Bacterium myxococci-incrustantis n. sp.), which Dr. Zederbauer has kindly communicated to the writer, serves further to confirm this impression. An examination of this specimen shows it to consist of a horny incrustation which at least closely resembles a dried up mouldy plasmodium, blackened by the abundant fructifications of a toruloid hyphomyeete; and from the fact that the bulk of the mass consists of calcic carbonate, one might perhaps venture the suggestion that it may be related to the Physareae. That a number of organisms are associated in this lichen can scarcely be disputed; yet whatever it may prove to be, either as a whole, or in detail, it surely has no connection with any of the Myxobacteriaceae, as this group is at present understood. * * * "Since the present paper was sent to the Gazette for publication a mounted preparation con- taining sections of authentic material of Ckondromyces glomeratus has been received from Dr. Zeder- bauer and proves to be the conidial condition of Coryne sarcoides (Jacq.) Tul., to which the name Tremella sarcoides was given by Fries. It need hardly be remarked that this fungus is a widely distributed and very common form, well known to mycologists, having no connection either with 'lichens' or Myxobacteriaceae." In 1903, Molliard succeeded in obtaining normal perithecia of Ascobolus in cultures on nutrient media by introducing into such cultures certain bacteria. The following is a brief account of his work. He easily obtained the germination of ascospores of Ascobolus gathered aseptically. A luxuriant mycelium was formed with abundant arthrospores, like those found by Brefeld. The white myce- lium growing from these filled the culture-tube but produced very few perithecia. These appeared only after 4 to 6 weeks and did not mature. In some other cultures perithecia appeared after 10 to 15 days and matured normally. In such cultures the perithecia arose on a few filaments, rising into the air above the rest, which were wet with the liquid which bathed the substratum. A microscopic examination showed that these cultures were contaminated with bacteria. As the original asco- spores were taken from colonies on cowdung, Molliard assumed that the bacteria were carried over from that source. To test this assumption he cultivated A scobolus on sterilized cowdung. A vigorous mycelium was produced which remained indefinitely sterile, i.e., did not produce perithecia. When, however, the mycelium was cultivated on sterile cowdung and then contaminated with bacteria isolated from the tube cultures, only a few filaments appeared on the exterior of the substratum, but these produced, in the course of 20 days, numerous large perithecia. From these results Molliard concludes that the bacteria are in some way responsible for the formation of perithecia, and that we have thus presented a method for obtaining perfect forms of many coprophilous and humicolous fungi. Bacteria with Myxomycetes. Pinoy speaks of a symbiosis with bacteria as necessary for obtaining cultures of myxo- mycetes. In his first paper he says that various workers have had bacterial contamination in their myxomycete cultures and raises the question: "Can one obtain a pure culture of myxomycetes?" He obtained cultures of two species of myxomycetes with bacteria on solid media. The bacterial species was identified as Bacillus luteus Fliigge. The myxo- mycetes were Chondrioderma difforme and Didymium effusum. The only evidence given that bacteria are necessary to the growth of these myxomycetes is that, when transfers of spore material were made after flaming the sporophores, some tubes remained sterile and others gave mixed growths. According to Potts there is no symbiosis. The Diet v. mucoroides profits by the presence of the bacteria, but the latter grow equally well when the myxomycete is absent. When the Dictyostelium fruits on bacterial colonies it causes them to become transparent. A large proportion of the bacteria forming such colonies are dead (98 per cent in one case, 83 per cent in another), and when such colonies are examined under the microscope they are seen to contain many involution forms and remnants of bacteria, but few or no normal bacteria. The bacteria are killed and then consumed. Dictyostelium was cultivated in combination with four different bacteria: Bad. fimbriatum, Bacillus megaterium, Bacillus subtilis, Bad. fluorescens liquefaciens. It can 170 BACTERIA IN RELATION TO PLANT DISEASES. not make use of the by-products of these bacteria, but nourishes itself at the expense of the bacteria themselves. "Die enorme Zahl von Bacterien, die Dictyostelium mucoroides verdaute, und seine geringen Anforderungen an die Ernahrung erklaren es, dass Bacterien den Hauptbestandtheil seiner Nahrung bilden und fur ein massiges, wenn nicht reiehliehes Wachstum geniigen." In a second paper Pinoy states that he freed spores of Dictyostelium mucoroides from the presence of Bacillus fluorescens liquefaciens by exposing them to a temperature of 500 for one hour. Under these conditions the bacteria were killed but the spores of the myxo- mycetes were not killed. Such spores germinated readily in the presence of various bacteria e. g., Bacillus fluorescens liquefaciens, Microbacillus prodigiosus, Bacillus coli, etc. They did not germinate when sown by themselves. The development of the myxomycetes was more or less abundant according to the bacterial species used in connection with it. When fluorescent bacteria were used the Dictyostelium became a yellowish color. On the contrary when the Bacillus prodigiosus was used the spore-heads were white with a very slight tint of rose color. Vuillemin has also published a note on this subject. He states that he cultivated Dictyostelium mucoroides in test-tubes, cotton plugged, on agar containing 0.5 per cent peptone and 0.2 per cent maltose, at laboratory temperature, protected from the light. The sowings were made from white spore heads and these often contained bacteria which he states to have been a fetid fluorescent species. All the tubes which showed any growth of Dictyostelium also contained the bacteria. The fruiting pedicles appeared the third day. If the sowings had not developed any bacteria there was no visible growth of the myxo- mycetes, although the microscope showed amoeba-formed bodies to have issued from the spores. To obtain growths of the myxomycetes from these apparently sterile tubes it was only necessary to introduce a culture of the bacillus. When pyocyanic bacteria were substituted the results were negative. Nadson believed that the bacteria rendered service by producing an alkaline substratum. Vuillemin states that the bacteria do not act indirectly, by modifying the substratum, but that they serve directly as food for the amoeboid bodies of the myxomycetes. Nadson, whose paper is earlier than any of those already referred to (1899), states that he obtained absolutely pure cultures of Dictyostelium mucoroides, but that these were weak dwarfed forms, giving generally no proper conception of the species. He also speaks of a symbiosis and says that the ordinary companion of this myxomycete is Bacillus fluo- rescens liquefaciens Flugge. In 1905 Pinoy published a paper on the role of bacteria in the devolpment of Plasmo- diophora brassicae, the myxomycete occurring in hernia of the cabbage. He found in pieces of young tumors of cabbage obtained by experimental infection, that some cells invaded by the parasite contained also masses of bacteria (forms of cocco- bacillus occurring either singly or in pairs). He followed this microscopic work by cultures as follows : The surface of large tumors, showing no trace of decay was burned deeply with a hot iron, and portions removed by means of flamed pipettes. The spores of the parasite contained in great num- bers in this material were sowed on the ordinary media and produced numerous colonies of bacteria. He thinks, therefore, that the bacteria were introduced into the root of the cabbage by the parasite. What role do they play? Pieces of healthy young turnip were removed aseptically by means of a sterile punch (Borrel's), placed in flamed tubes and sowed with the spores of the fungus. The tubes were then sealed in the flame and placed in the thermostat at 220 C. During the first days scattered colonies of aerobic bacteria arose, which, however, ceased to grow when the oxygen was exhausted. Five days after sowing, the cells of the turnip contained Plasmodiophora in various stages. Many cells were filled with spores. When the same experiment was carried on in tubes plugged with cotton, i.e., exposed to the air, the aerobic bacteria which accompanied the spores developed more abundantly and brought about the decay of the turnip. When anaerobic bacteria were accidentally introduced the growth of the myxomycete was stopped. BACTERIA WITH MYXOMYCETES. 171 The presence of aerobic bacteria seems to be necessary to the life of the myxomyeete outside of the host cells. Thus among the great number of tubes of agar sowed with the spores, the greater part of those containing bacteria gave at once a development of the fungus (formation of the amoeboid individuals which soon perished) while in two containing no bacteria, the spores, though perfectly- preserved did not germinate. Pinoy thinks, therefore, that the bacteria introduced with the parasite contribute to the decay of the tumor of the cabbage when the conditions are favorable to their multiplication. Pinoy continued his interesting studies, publishing a monograph in 1907 wherein he considered at length the relations of various bacteria to several species of Myxomycetes in the Group Acrasieae ; then to several species of the endospore-bearing forms, and finally to Plasmodia phor a brassicae. I summarize as follows: Pinoy, like Nadson, found Bacterium fluoresceins liquefaciens associated with Dictyostelium mucoroides. On agar poured plates the spores of this slime mold germinate only in the presence of bacterial colonies. Elsewhere the spores do not germinate. But those spores which do not germinate can not for that reason be assumed to be bacteria-free. They also bear bacteria which can be resusci- tated by putting them into bouillon, which clouds after 6 to 8 days. Nadson, Potts, and Vuillemin did not take into account these tardily developing bacteria. B. fluoresceins is killed by exposure to 500 C. for 1 hour. On the contrary, 80 per cent of the spores of Dictyostelium mucoroides are still able to germinate after such an exposure. It is possible also to purify the spores by heating them for 2 minutes at 560 C. Spores about 8 days old are most resistant. Spores thus purified will never germinate on any culture medium whatsoever, unless suitable bacteria are added. Potts' statement that he was able to grow Dictyostelium mucoroides on the dead bodies of his Bacterium fimbriatum is regarded as illusory, i. e., the purity of his spores is questioned. Pinoy showed the need of living bacteria quite clearly as follows: Into a flask of culture medium he plunged a collodion sack attached to a projecting tube. After proper sterilization, the medium inside the collodion sack was inoculated with purified spores of D. mucoroides, that outside with Bact. fluoresceins. The spores of the Myxomyeete germinated but the amoebae soon rounded off and died. The dead bodies of Bad. fluoresceins (killed by heat, ether and chloroform) were also placed in the collodion sack without result. " En resume, le D. mucoroides ne peut vivre qu'en association avec une Bacterie vivante. Toutes les Bacteries ne conviennent pas egalement." The culture medium exerts a marked influence, e. g., on potato sowed with mixtures of Diet v. mucoroides and Bad. fluorescens the slime mold does not develop. The same is true on this medium whatever bacterium isused. With most bacteria this is also true on peptone agar, or meat broth agar. The best culture medium found was flax-seed agar. On this the bacteria grow abundantly and the harvest of Dictyostelium mucoroides reaches its maximum. Bacteria that stain by Gram are not suitable for such cultures. In general, bacteria which do not stain by Gram allow the Dictyostelium mucoroides to grow. The Dicty. mucoroides will not grow in the presence of pure cultures of Bacillus megaterium, but if this organism is added to Dictyostelium mucoroides with Bad. fluorescens growth may be had on beef agar. The growth of the Dictyostelium mucoroides may be regarded as a parasitism on the bacteria. They are absorbed into the vacuoles and digested. Pinoy confirms Metchnikoff's observation that the liquid in the vacuoles is acid. Neutral red is recommended as a stain for the bacteria in process of digestion. It does not stain the living bacteria nor kill the Myxomyeete unless too strong. Vesuvin may also be used. For details see paper. Potts did not find bacteria in the interior of the vacuoles of Dicty. mucoroides because his technic of fixation and staining was insufficient, he therefore formed the erroneous hypothesis of an extra-cellular diastase. Grown with Bacillus coli, the enzyme isolated at the end of 40 hours liquefied gelatin. It acts in neutral or feebly alkaline liquids. The acidity of methyl orange inhibits. It is therefore related to trypsin rather than to pepsin. It is destroyed at 550 C. Its maximum of activity is about 380 C. It has scarcely any action on fibrin or on albumen coagulated by heat. " Acradiastase " does not act on bacteria killed by heat, but readily dissolves those killed by ether or chloroform. "The best bacterial test is also B. coli which is not self-autolytic, and a chloroformed emulsion of which remains cloudy. 172 BACTERIA IN RELATION TO PLANT DISEASES. "Let us take such an emulsion and add some drops of it to two tubes : one containing the normal diastasic liquid, the other the same quantity of this liquid boiled. They are put into the thermostat at 380. Of these two tubes, equally cloudy, the check after some hours remains cloudy, while the other has become almost completely transparent." There is no precipitate and therefore the clearing can not be ascribed to agglutination. "Ainsi Dictyostelium mucoroides ne peut se developper qu'avecdes Bacteries; il est parasite des colonies bacteriennes ; ses myxamibes ingerent les Bacteries et les digerent dans leurs vacuoles a l'aide d'une diastase dont Taction est assez semblable a celle de l'amibodiastase." Similar results were obtained with other species, i. c, Dicty. purpureum, and Polysphondylium violaceum, showing that these also are bacterial parasites. These Acrasieae are strictly aerobic. As soon as a tube is sealed growth ceases. The amount of humidity greatly influences the morphology of the sporophores. The optimum temperature for growth is between 220 and 250. Above 280 there is no development. They will grow at a tem- perature as low as 8° but then very slowly. The morphology and the color of the Myxomycete are both changed by changes in the sub- stratum, e. g., if Bacillus subtilis is added to mixed cultures of Dictyostelium mucoroides and Bad. fluorcscens, the sporophores are longer and branched forms are frequent. Under some circumstances bacterial pigments are absorbed by the living Myxomycetes. The author holds that certain Acrasieae described as distinct from Dicty. mucoroides on account of their variation in color are only the same species associated with different chromogenic bacteria. These bacterial pigments therefore have a taxonomic importance in the Acrasieae. Grown with Bad. fluorescens the young fructifications of Dicty. mucoroides are fluorescent and the old ones are color of a dead leaf; grown with B. coli the fructifications of this species are pure white and remain so. When Polysphondylium violaceum, which has a pigment of its own, is grown in the presence of Bacterium violaceum its color becomes paler, the pigments of the two being chemically dissimilar bodies. Similar results as regards necessity for living bacteria were obtained with Didymium difforme and D. cjfusum. So far as the writer has observed bacteria always occur in the club-root of crucifers along with Plasmodiophora brassicae. One Bacterium with Another. This subject is a very large one and no attempt has been made to cover it either in the text or bibliography. According to Beyerinck and Van Delden their Chroococcum assimilates nitrogen only when it enters into symbiosis with other bacteria — Granulobacter, Aerobactcr, etc. In 1906, Keding published his Weitere Untersuchungen. He found Azotobacter not only on the surface of Fucus and several other salt water algse, but in dune sand near the roots of strand plants, and in all investigated soils, except moor soil. Azotobacter is able, he says, to assimilate the nitrogen of the air in pure culture, and this ability was not increased by growing it in combination with other bacteria. The sea forms of the organism can grow in the presence of 8 per cent salt. According to Bottomley " PscudoHicmas radicicola and Azotobacter, together make a powerful combination for the fixation of free nitrogen." These are both said to have been isolated from the algal zone of the root-tubercles of cycads. He inoculated oats, barley, hyacinths (galtonia), and parsnips with mixed cultures. Best results with oats which were nearly doubled in weight. Thomas F. Manns (The Blade Blight of Oats; a bacterial disease. Agr. Exp. Sta., Ohio, Bull. 210) has stated that a widely prevalent disease of oats is due to a symbiotic relationship between two species of bacteria. SYMBIOSIS. 173 LITERATURE. Bacteria with Yeasts. [For the earlier literature see Vol I, p. 214.' 1892. Ward, H. Marshall. The "ginger-beer plant," and the organisms composing it: a contribu- tion to the study of fermentation-yeasts and bacteria. Proe. of the Roy. Soc. of London, London, 1892 (possibly late 1891), vol. L, No. 304, pp. 261-265. A preliminary note. 1892. Ward, H. Marshall. The ginger-beer plant, and the organisms composing it. Philos. Trans, of the Roy. Soc. of London, 1892, vol. 183, Series B, pp. 125-198, pis. 11-16. 1902. Podwyssotskv, Wladimir. Le Kephir (Fer- ment et boisson therapeutique prepares avec du lait de vache). Histoire, preparation, composition de la boisson, morphologic du ferment, ses maladies; valeur physiologique et therapeutique du kephir. Traduit d'apres la cinquieme edition russe, notablement modifiee et agumentee, par Mile. S. Broido et Mme. P. Eliacheff. avec preface de M. G. Hayem. Published by C. Naud, Paris, 1902. 4 figs, pp. x, 76. There is also a German trans- lation by Schmidt from the 4th Russian Ed. Contains a bibliography of seventy-five titles. 1910. Doidge, Ethel M. The Flora of Certain Kaffir Beers "Leting" and "Joala." Agri- cultural Science Bulletin No. 5, Transvaal Dept. of Agri., Pretoria, 1910, 31 pp., 8 pis. Bacteria with Fungi. 1903. ZEDERBAUER, E. Myxobacteriaceae.eine Sym- biose zwischen PilzenundBakterien.Sitzungsb. der K. Akad. der Wissenschaften, Jahrgang 1903, 1 Abt., cxii Bd., iv bis vn Heft. Wien, 1903, pp. 447-482, 2 Tafeln. Paper of doubtful value. 1903. Molliard. Role des bacteries dans la produc- tion des peritheces des Ascobolus. C. R. d. se' d. l'Acad. d. Sci., Paris, 1903, Tome cxxxvi, pp. 899-901. 1904 Thaxter, Roland Contributions from the Cryptogamic Laboratory of Harvard Univ. LVI. Notes on the Myxobacteriaceae Bot. Gazette, vol. xxxvii, 1904, pp. 406-408. Bacteria with Myxomvcetes. 1899. Nadson, D. A. Des cultures du Dictyostelium mucoroides Bref. et des cultures pures des Amoebes en general. (Extr. des Scripta Botanica. Fasc. xv, 1899, 8°, 38 pp. St. Peters- burg.) Resume Just's Botanischer Jahres- bericht.Siebenundzwanzigster Jahrgang (1899), Leipzig, p. 86. 1902. Pinoy. Necessite de la presence d'une bacterie pourobtenirlaculturede certain myxomycetes. Note preliminaire. Bull, de la Societe Mycolog. de France, Aug., 1902, Tome xvin. p. 288. 1902. Potts, George. Zur Physiologie des Dicty- ostelium mucoroides. Flora oder Allgemeine Botanische Zeitung, Bd. 91, Oct. 4, 1902, pp. 281-347. Bibliography of 38 titles. 1903. Pinoy. Necessite d'une symbiose microbienne pour obtenir la culture des myxomycetes. Paris. Compt. Rend, des se. de l'Acad. des Sei. >9°3. Tome cxxxvn, pp. 580-581. 1903. Vuillemin, Paul. Une Acrasiee bacteriophage. Compt. Rend, des se. de l'Acad. des Sci., Aug. 10, 1903, Tome cxxxvn, pp 387-389. 1905. Pinoy. Role des bacteries dans le developpe- ment du Plasmodiophora brassicae, Myxo- mycete parasite produisant la hernie du ehou. Compt. rend. soc. biol., T. lviii, No. 22, pp. 1010-1012. 1907. Pinoy, Ernest. Role des bacteries dans le developpement de certains Myxomycetes. Ann. Inst. Past., vol. 21, No. 8, pp. 622-656, pis. 13-16, Aug. 25; No. 9, pp. 688-700, Sept. 25. Paris, 1907. Bacteria with other Bacteria. 1902. BeijErinck, M. W. und van Delden.A. Ueber die Assimilation des freien StiekstofTs durch Bakterien. Centralbl. f. Bakt., 1902, 2 Abt., Bd. ix, pp. 3-43. 1906. Keding, Max. Weitere Untersuchungen iiber stickstoffbindende Bakterien. Wissenschaft- liehe Meeresuntersuchungen herausgegeben von der Komm. z. wiss. Unters. d. d. Meere in Kiel und der Biol. Anst. auf Helgoland. Neue Folge, Neunter Bd., Abt. Kiel, 1906, PP- 273-308. 1907. Belonowski, G Uber die Produkte des Bacterium coli commune in Symbiose mit Milchsaurebacillen und unter einigen anderen Bedingungen. Biochem. Zeitsehr., vol. 6, Berlin, 1907, pp. 251-271. 1908. Musgrave, W. E. The influence of symbiosis upon the pathogenicity of microorganisms (the evolution of parasitism). Phil. Journ. Sci., B, Med. Sci., vol. in, 1908, pp. 77-88. 1908. Proca, G. Sur quelques particularites du Bacille fusiforme (Vincent ) cultive en symbiose. Compt. Rend, de la Soc. Biol., Paris, 1908, T. I., pp. 771-772. 1908. Crithari, C. Etude sur la symbiose du Bacille bulgare et du Bacille butyrique. Compt. Rend, de la Soc. Biol., Paris, 1908, T. I., pp. 818-820. 1909. Bottomley, W. B. Some effects of nitrogen- fixing bacteria on the growth of non-leguminous plants. Proc. Royal Soc., Series B, vol. 81, No. B. 584, Biological Sciences, pp. 287-289. 1910 Seliber, G Sur la symbiose du bacille buty- rique en culture avec d'autres microbes an- aerobies. Compt. Rend, des se. de l'Acad. des Sci., T. CL-, Paris, June6, 1910, pp. 1545-1548. ARE ANY BACTERIA KNOWN TO CAUSE DISEASE IN BOTH PLANTS AND ANIMALS? EVIDENCE FROM INOCULATING PLANT PARASITES INTO ANIMALS— EVIDENCE FROM INOCULATING ANIMAL PARASITES INTO PLANTS— DO PLANTS HARBOR ANIMAL PARASITES? Theoretically, this subject is of great importance. Actually, very little of positive value has been developed by the studies thus far undertaken, i. e., the results in general have been negative. Most bacterial plant parasites are unable to grow at blood-heat, and for this reason may be regarded as harmless to man and the domestic animals. Most animal parasites are more or less delicately balanced to the conditions prevalent in animal bodies and not to those occurring in plants, although when inoculated into certain plants some of them have remained alive in the vicinity of the wound for a consider- able period. The chief danger to health would appear to lie in the ingestion of plants whose surfaces have been contaminated by animal pathogenic organisms, i. c, in the use of raw vegetables and salad plants, particularly those grown on lands fertilized with untreated sewage. Sewage should be sterilized before it is passed into streams or flooded upon agricultural lands. Vegetables grown on lands manured with night-soil or with untreated sewage should not be eaten raw. It would be entirely proper to prohibit altogether the sale of such vegetables. The principal studies, so far as known to the writer, are summarized in the following paragraphs. ANIMAL PARASITES INOCULATED INTO PLANTS. Grancher and Deschamps (1889) experimented on seedling radishes and carrots grown in special boxes and watered repeatedly with typhoid cultures diluted in water (20 cultures in 10 liters of sterilized water). The experiment was begun April 9 and finished June 6. Nine gelatin plates were poured from the inner tissues with negative results, the plants being wiped and flamed, and the pulp removed under sterile conditions. Tests were also made by them of radishes and carrots from the garden of the hospital and of radishes, carrots, and asparagus from the municipal garden at Gennevilliers, 46 tubes of peptone-gelatin and 20 flasks of bouillon being inoculated. Part of these cultures were kept in the thermostat and the rest held at room temperature. All were negative. Conclusion: Le Bacille typhique et les microbes communs du sol ne penetre pas dans la pulpe des legumes sains. One of the hospital radishes yielded a common organism but its pulp was probably already invaded through a scratch on its surface. In 1890 Lominsky* published his paper in the Russian Medical Journal Wratch. It is believed that a rather full account of this paper will be welcome to English readers. The author approaches this problem from the standpoint of a physician. If plants are capable of nourishing a single organism causing animal disease, to know it is a matter of great importance. It has long been known that disease-producing microbes can grow on dead vegetable matter, espe- cially some culture media, e.g., cooked potato. Whether they will grow on living plants is quite another matter. Up to this time living vegetables have been considered very unfavorable media for the growth of bacteria. The experiments of Buchner, Lehmann, Fernbach, Miquel and Grancher lead to one conclusion, viz., that vegetables, seeds and plants do not contain microbes. "And, therefore," says the author, "I had in view to investigate whether the animal-pathogenic bacteria are able to find in the tissue of a living and growing plant a favorable soil for their existence. " *Spellcd also Lomnitzky, Lominskago, Lommitzky, etc. "74 ANIMAL PARASITES INOCULATED INTO PLANTS. 1 75 The paper is stated to be a preliminary one. For his experiments the author took three bacteria, viz, the bacillus of the Siberian plague, the bacillus of typhoid fever, and Bacillus prodigiosus. The plants inoculated were Triticum vulgare, Agapavthus, Polygonum fagopyrum, Trifolium pralense, Sambucus, Hyacinthus, and Tulipa. Heinz's results with his B. hyacinthi-scpticus were known to Lominsky. Two ways were chosen for investigating this subject : (1) Seeds were planted on infected soil; (2) plants were inoculated by puncture, especially on the leaves. The surfaces of the seeds were sterilized by washing in soap and water and then in mer- curic chloride 1 :iooo. They were afterwards left for half an hour in 1 15000 HgCL or for one hour in 1 :io,ooo HgCL. Seeds of wheat thus sterilized on their surfaces were treated in two ways. In the one case, they were plunged into colonies of the above named bacteria, then laid in a tin box on sterilized soil, and covered with about an inch of soil. This little box was then put on a glass plate and covered with a bell-jar, the upper opening in which was covered with cotton. From time to time sterile water was added. The seeds germinated in 5 to 30 days, the room temperature being 250 to 2 70 Celsius. In the other case, the germinations were sometimes made on moist sterilized cotton, but more often on boiled potatoes prepared as for bacterial cultures, except that to them was added a little of the following solution: water, 1,000; potassium nitrate, 1; potassium sulphate, 0.25; monopotassium phosphate, 0.25; magnesium sulphate, 0.25. A little ferrum phosphate in powder was also added. After sterilizing this medium the wheat was put in. This method enables one to decide whether the seed and substratum have been properly sterilized. Then, after a few days, the microbes were introduced on the end of a platinum needle. The inoculations on the leaves were made into very young plants and into older ones. The very young plants were germinated in soil or in cotton, and when the green parts had reached a height of 2 to 5 cm. the specified microbes were inoculated by means of a needle. On adult plants the inocula- tions were made with a platinum needle shoved flat-wise between the upper and lower surface of the blade, after first washing the leaves in 1 :iooo HgCl2 and drying them in sterilized cotton. The sur- face of the wound was covered with collodion. Leaves of plants were also dipped into sterilized water to which the microbes had been added. After 3 to 42 days the inoculated leaves were examined microscopically. Cultures were also made from them and inoculated into animals. Leaves to be examined were hardened in alcohol. Bacteria in the growing tissues were stained by the methods in use for animal tissues. Three hundred experiments were made. The author's conclusions are as follows : (1) Disease creating microbes [animal pathogenic bacteria] may find conditions for their existence in tissues of the higher plants. (2) The uninjured cuticle prevents the entrance of bacteria. (3) Mechanical injuries of the leaves and stems of growing plants afford an opportunity for the entrance of bacteria into the tissues. (4) The bacillus of Siberian plague [Aplanobacler anlkracis], the bacillus of typhoid fever, and the B. prodigiosus can multiply and form colonies inside of living plants. (5) In artificial inoculations these three bacilli multiply not only at the point of inoculation but spread into the neighboring parts. (6) Although these three bacilli spread from the point of inoculation to adjacent parts of the tissue, they do not extend widely, that is, the whole organ or the whole plant is not infected by artificial inoculation. (7) The part of the leaf injured by the microbes sometimes may be identified macroseopically. The injured spot in the leaf differs from the healthy part by being lighter green. Sometimes on the part injured by Bacillus pro- digiosus brick-red spots or stripes are noticed along the track of the injury. [Probably a host-reaction.] (8) The disease-creating microbes spread in the plant by way of the intercellular passages, and the size of the microbe is of great importance in this regard. The smaller it is, the easier it spreads in the tissue. For this reason B. prodigiosus spreads farther than the others. (9) The author could not observe that motility in any way favored its spread. (10) The walls of the cells do not absolutely prevent the entrance of the microbes into the cell. (11) The protoplasm of the cell may afford a medium for the growth of the microbes. (12) The dead and dry tissue does not afford a good medium but the dead and juicy cells afford a very convenient soil for their development. The microbes were also alive in the living cells but preferred the dead ones. (13) The bacillus of the Siberian plague multiplies vigorously during the first few days in the leaves of Agapanthus and grows out into a thread. At the end of the first week there is an inclination to form spores, which in course of time becomes more distinct. Many spore chains were visible on the eighteenth day, together with separate spores and nonsporiferous threads. Filaments and spores occur not only at the point of inoculation but also between the healthy cells of the spongy tissue of the leaf, and in the cells themselves. Some of these threads stain well with gentian violet, others do not. Those which do not stain with gentian violet, stain afterward with carmine (double stain) and are refractive. Still others stain only in parts, or are not stained, and look like bright, pale, drawn threads. Slides made 42 days after inoculation still showed numerous vegetative forms of the plague bacillus, together with spores and spore- bearing threads. 176 BACTERIA IN RELATION TO PLANT DISEASES. (14) In Agapanlkus leaves 26 days after inoculation the bacilli and filaments of the Siberian plague were found showing great changes. On the unstained slides their color was somewhat yellowish and they were noticed on account of their refraction which was like glass or galena. These threads were two or three times the diameter of normal fila- ments and they were constricted, i. e., they had lost their normal filamentous form. (15) Sowings of infected parts of the leaves of agapanth on nutrient gelatin or boiled potatoes on the sixteenth and forty-second day after inoculation gave a typical culture of the Siberian plague bacillus. (16) The inoculation into mice of portions of similar leaves 16 and 42 days after the inoculation of the plant caused the death of the animals from the Siberian plague. (17) The typhoid fever bacillus multiplied in the leaves of wheat and agapanth only during the course of the first days after inoculation, gradually dying out. This dying out was shown (a) by its not taking Loeffler's or Ziehi's stain, (b) by the presence of involution forms, (c) by its failure to grow in cultures. (18) Of all the microbes investigated B. prodigiosus multiplied most energetically and after the manner of the Siberian plague, that is, in the intercellular passages and in the living cells adjacent to the point of inoculation. (19) The dying out of B. prodigiosus was not noticed even at the expiration of 32 days from inoculation. On this date transfers from inoculated parts of the leaf into nutrient gelatin and boiled potatoes gave a typical culture. (20) Plants in the course of their growth may mechanically throw out the microbes from shallow layers to the surface. (21) When wheat is grown on soil infected by disease-creating microbes many of the microbes may enter into the root-system, the smaller ones getting in the easier. (22) When wheat is grown on soil infected by a mixture of microbes all of them may be found in the tissues cf the root. (23) The passage of the microbes from the infected roots of wheat into the stems and leaves was not observed. Russell (p. 6) states that he could not confirm Lominsky's results with Bacillus prodi- giosus, to wit, the production of red spots and stripes in the injected plants, but inasmuch as he did not experiment with the same plants as Lominsky, his experiments can not be considered as a refutation of Lominsky's statements. Russell also states that he failed to verify some of the results obtained by Lominsky with animal parasites injected into plants, but here again his experiments are not strictly comparable since he used different plants. He also obtained different results from watering the soil with "dilute infusions of the different germs." In this case Russell does not state what plants he used, but presumably not wheat, from a remark on the following page — "This result would have been much more convincing had he [Lominsky] used larger plants than wheat." Russell found Bacterium pyocyaneum present in large numbers at the point of inoculation in begonias after 69 days; in geranium after 32 days, and in Paithorum after 36 days. The anthrax organism was absent from geranium (Pelargonium) at the point of inoculation after 38 days, and was only sparingly present in lima bean after 1 1 days and in Echinocaclus after 5 days. Staphylococcus epidcrmidus alius was not recovered from the point of inoculation in geranium after 40 days. Staphy- lococcus pyogenes aureus was also dead in geranium after 42 days, but was recovered very sparingly from lima bean after 13 days. B. cholerae galliuarum was moderately abundant in geranium after 18 days. The organism of Schweineseuche was present in large numbers in geranium after 17 days. The diphtheria organism was not found in geranium at the point of inoculation after 10 days. He has the following paragraph on the result of his experiments with Bacillus amylovorus and Bacterium avenae. ' 'The pear-blight germ grown in a begonia-plant for 30 days showed at end of that time large numbers at inoculation point, but not distributed throughout the plant. The same result was found when injected into Phaseolus vulgaris for 30 days, also in Ph. lunatics for 16 days. In Tradcseantia alba, no trace could be found at the end of 60 days' incubation in this tissue. Bad. avenae was injected into tissue of begonia, onion, corn, wheat, and squash, but in no ease was any pathological change macroscopically observable. The bacilli were not killed out in the plant-tissue, however, as they were isolated from begonia and squash in large numbers, after 30 days' incubation in these tissues, but their presence was confined to the tissue contiguous to point of introduction." Concerning the general conclusions to be drawn from his own observations, Russell has the following: "The results of the foregoing inoculation experiments made with various forms of micro-organ- isms, saprophytes as well as parasites (both for animals and vegetables), show that these germs in many cases are able to live in the plant-tissues for a considerable length of time. A number of the different forms, particularly saprophytes, are able to grow and spread throughout the plant to a limited extent. Of the parasitic species tested, very few showed any tendency to thus spread. Even ANIMAL PARASITES INOCULATED INTO PLANTS. 1 77 those forms that are natural parasites of certain higher vegetable species showed no power to spread in plants which were not their natural hosts, but they were able to live at inoculation-point for a con- siderable time. * * * "The distribution of the micro-organisms in the plant-axis, as determined by culture experiments, always took place in an ascending direction. This distance varied from 30 to 50 mm. from point of introduction, but in no case were bacteria found more than 2 to 3 mm. below inoculation-point." Charrin (1893) injected Bad. pyocyaneum into Pachyphyton bracteosum, a plant of the family Crassulaceae. The inoculations were made into the fleshy leaves. This organism lived for some time and multiplied (mostly in the intercellular spaces). The leaves of the plant, after the germs had grown in them 15 to 30 days, became wrinkled, lost color, and fell off; the acidity of the leaves lessened in proportion as the bacteria multiplied. Even when large numbers of the organism were introduced (0.25 to 0.5 cc. of a liquid culture) it remained alive a short time only (2 to 3 weeks, more or less) and when only 1 to 2 drops were inoculated 8 to 12 days usually sufficed to kill the organism, especially if a weakened germ was used. An interesting paper, especially the last half dozen paragraphs on immunity. Kasparek and Kornauth (1896) experimented with Aplanobacter anthracis on oats, barley, wheat, rape and maize. Their method of procedure was as follows: Sterilized flower-pots were filled half full of sterile soil and each watered with 10 cc. of bouillon containing large numbers of spores and threads of the anthrax organism. Sterile seeds of the above mentioned plants were then embedded in this moistened earth and the pot filled nearly full with additional sterile soil. The surface of the seeds was sterilized by soaking for a short time in 2 per cent mercuric chloride solution which was removed by washing in sterile water, alcohol and ether. Then they were put for 24 hours into nutrient bouillon at blood temperature to be sure that their surface was actually sterile, and the seeds in those tubes which clouded were rejected. The germi- nating power of seeds thus treated was not injured. After two months in the first series, and after three months in the second series the experiment was broken off and the earth and plants examined microscopically and bacteriologically. A microscopic examination of the soil showed that anthrax spores were abundant not only in the part which was watered, but also in the upper layers of the soil even to the surface, but no anthrax threads or rods could be found. Samples of this soil pro- duced typical anthrax when inoculated into white mice and also gave numerous anthrax colonies when sowed in plate cultures. The unwashed underground parts when inoculated into mice also produced typical anthrax, but when these roots and stems were first washed with mercuric chloride, alcohol and ether, the same as the seeds had been, they did not produce any disease in mice when inoculated, and the agar plates also remained sterile. Moreover, a microscopic examination of sec- tions made through the roots and other parts of these plants grown in anthrax infected earth likewise showed complete absence of the anthrax organism in the tissues of the plants. Kornauth (1896) continued and extended these experiments, including Streptococcus pyogenes with anthrax in order to have a very small Schizomycete for comparison with the large one. The results were the same. Kornauth used for his experiments seeds of maize and peas. These were washed in 2 per cent mercuric chloride water, alcohol, and ether, then put into sterile bouillon in Petri dishes and incubated for 2 days at 370 C. If the bouillon remained clear then it was inocu- lated with either A planobactcr anthracis or the streptococcus. The cultures succeeded admirably and appeared on microscopic examination to be pure. After about 3 weeks when the seedlings had reached the length of about 2 cm. the experiment was broken off. After washing the seedlings in mercuric chloride water, alcohol and ether to render the surface sterile, they were crushed with anti- septic cautions, and inoculated into mice (those used with anthrax) and also put into sterile bouillon. Sections of these seedlings were also prepared for microscopic examination. The result of the experi- ment was that the mice did not contract the disease, the bouillon remained clear, and the sections showed no trace of any bacteria. His conclusion, therefore, is that the plant under normal conditions is a perfect filter for bacteria, that only a few species of bacteria can penetrate the uninjured plant tissue, and these only under very special conditions. He next undertook to determine whether the organisms would multiply in injured tissue, i.e., in wounds, as stated by Lominsky. For this purpose he selected the following 178 BACTERIA IN RELATION TO PLANT DISEASES. bacteria: Micrococcus cinnabareus, Aplanobacter pneumoniae (Weichselbaum), Streptococcus pyogenes, Bacillus coli, Bacillus prodigiosus, Mycobacterium diphtkeriae, Bacillus typhosus, Aplanobacter anthracis (spores and filaments) and Actinomyces. For each one he used two specimens of onions and hyacinths well provided with leaves, and three sorts of cactus. The places where the wounds were to be made were first painted with mercuric chloride, alcohol and ether to destroy the surface organisms, then with a pair of sterile shears the wound was made and through the opening by means of a platinum loop the culture was inserted. The wound was immediately closed, the exuding excess of culture removed with a sterile knife and the wound fastened together with collodion. As a rule the wound healed well. After 8 days the infected spots were sampled with a flamed corkborer, their infectiousness tested on animals, and the presence of the organism determined both by cultures and by examination of sections. With the next larger corkborer, a cylinder-mantle was also removed and transferred to bouillon. In the cylinders that were used for inoculations and sections the following organisms were found living: Aplanobacter anthracis, Bacillus prodigiosus, B. coli, and .1/. cinnabareus. The following were dead : Streptococcus pyogenes, Mycobacterium diphtkeriae, Bacillus typhosus and .4 plano- bacter pneumoniae. Actinomyces transferred to agar also failed to grow. The cylinder-mantle, which was about 5 mm. thick, left the bouillon clear. Anthrax was found only in spore form in the plant tissues These wounds were all superficial. The author then tried whether inoculations into deeper wounds would have any different results. For this purpose he used a needle with which he made deep punctures introducing into them the organism. The general method of procedure was the same, sterilizing the surface and finally covering the wound with collodion. Tests were then made after various periods. Again the anthrax organism was found to have sporulated and the spores were fullv infectious at the end of 4 months. The diphtheria organism and the pneumonia germ were noninfectious at the end of 48 hours. The typhoid bacillus at the end of 5 days was non-infectious, and Bacillus coli at the end of 8 days. Micrococcus cinnabareus and Bacillus prodigiosus dried out but remained alive a long time. Streptococcus pyogenes also remained alive, but showed on the ordinary culture media only a slight or weakened growth. The sections showed that the diphtheria organism had taken on involution forms, while the Actinomyces had fallen into a granular detritus. These experiments were made in warm and dry air. This author also tried some experiments in moist air, using for this purpose Aplanobacter anthra- cis (spores and threads), and Micrococcus cinnabareus. The inoculations were into buds and the plants were kept in a moist chamber. After some days fungi appeared on the inoculated places and these seem to have quickly killed the bacteria, as the latter could not be recovered in poured-plates. Kornauth's conclusions therefore, are just opposed to Lominsky's: "The bacteria introduced into the living plant under favorable conditions as to warmth, exclusion of foreign organisms, etc., have never shown any multiplication and just as little any staining of the inoculated spot (through chromogenic bacteria) or a loss of color of the surrounding tissues." Zinsser's paper appeared in 1897. It deals mostly with the root-nodule organisms of the Leguminosae, but there are also detailed experiments with other bacteria. The work was done in Pfeffer's laboratory. Zinsser gives a very interesting table showing the behavior of various bacteria when introduced into plant tissues. In one instance Bacillus prodigiosus yielded cultures after remaining in a bean stem 96 days. In general, Bacillus sublilis, B. megaterium. and B. prodigiosus were most resistant, being frequently found alive in the stems and leaves of various plants after 14 to 48 days. Other organisms were destroyed more speedily. Zinsser also inoculated animal pathogenic forms into plants. He experimented mostly with Aplanobacter anthracis. This lived in various plants such as beans, Cyclamen, Abutilon, Scmpcrviium, and Barbacenia from 14 to 28 days. In several cases it was dead after 27 to 28 days. Even the more resistant species did not multiply extensively or behave like parasites. All lost their ability to grow after a longer or shorter period and perished. Concern- ing their spread in the tissues the author says: " Now and then according to the microscopic appearance the injected bacteria appeared to have multiplied for a short time, and they were able also to penetrate into the neighboring tissue, but this power of translocation is not great, for even a few centimeters away from the point of inoculation I could not afterwards demonstrate bacteria." The following notes are from Hartleb's paper: Hartleb, who worked in Stutzer's laboratory in Bonn, states (1898) that he experimented with the bacteria of the foot and mouth disease, using their organism and Siegel's organism. The inocu- lations were made into the stems of beans, Yicia (aha, potatoes, and peas. He also inoculated pods ANIMAL PARASITES INOCULATED INTO PLANTS. 1 79 of beans and peas. His method was to wash the surface with mercuric chloride water, then with sterile water, after which a longitudinal incision was made with a sterile knife, the wound pried apart a little and the platinum needle carrying the infectious material introduced. Unlike Kornauth and Kasparek he did not cover the wound with collodion but left it exposed to the air so as to have the conditions as nearly as possible like natural conditions. He states that the organism remained alive in wounds for a long time and multiplied. When the seeds were wounded most of them passed over into a slimy rot which contained for a long time extremely well developed living bacteria of the form introduced. Bacteria from these wounds produced the typical disease when inoculated back again into animals (white mice and finally guinea pigs). In the dried parts both of stems and pods, the organism in its resting form was found to be alive and infectious to animals at the end of 6 months. "The inoculation cuts had for the most part a rusty brown appearance [host reaction], but not rarely in some cases there was also a slight accumulation of the slimy substance to be observed which was found to be an aggregation of bacteria into a slimy mass. " His conclusions are as follows: (1) Our bacterium cultivated upon acid nutrient media can develop further in living plant parts. (2) It is capable of remaining alive not simply for a short time, but even upon dead plant parts with the help of its resting forms, perhaps to maintain a prolonged capability of growth, without actually penetrating directly into the cell-tissue and multiplying inside of the same. (3) That, in consequence of this preponderating parasitic manner of life, it is able, when carried back to animals to cause the infection and death of the same. It is stated that the organism was also cultivated on carrots, both sterilized and unsterilized, where it formed at first a slime, and afterwards a dry layer, and in one instance a guinea pig was infected by feeding upon such carrots. The inoculated parts of the plants either healed over with more or less callose, or else there was a direct death of the cell-tissue in the vicinity of the wound. Generally the organism appeared in unmixed growths in the wounds, sometimes in vegetative forms as rods, and sometimes in the resting form. Only scattering wild yeast-cells accompanied it. Hartleb states that the cultures he used had lost most of their power to destroy animals and consequently this passage through plants increased their virulence. Just what he worked with is not apparent, as the cause of the foot and mouth disease is still believed by most pathologists to be undetermined and probably ultra microscopic (Vide Kolle und Wassermann, Jena, 1904, Bd. IV, 2ter Teil, p. 1325). Laurent (1899) maintained that he was able by special treatments to cause Bacillus coli to become a plant parasite, but this claim is still disputed. In this connection see page 4 2 . John R. Johnston working in the writer's laboratory on the coconut bud-rot of the West Indies has obtained evidence that it is due to an organism indistinguishable from what ordinarily passes for Bacillus coli (Phytopathology, Vol. I, No. 3). Ellrodt's paper (1902) relates to the possible transmission of human and animal parasites by means of plants. He states that Bacterium pyocyaneum can enter plants through wounded roots, but not through sound ones. This conclusion rests on the follow- ing experiments. In a series of flower pots, oats, beans, vetches and peas were planted. When these were about 20 cm. high, the earth was watered with a suspension of Bacterium pyocyaneum. The same experi- ment was undertaken with Viola odorata, Paconia officinalis, and Iris sibirica. Cultures were made the next day by cutting out a piece of the plant with sterile knives, thrusting a platinum needle into the exposed tissues and streaking the sap on agar and glycerin agar. As the cultures gave absolutely negative results, the earth was again wet with a bacterial suspension of the organism, and after 4 or 5 days streaks were again attempted on agar in the same manner. These likewise remained sterile. vSamples of the soil, on the contrary, yielded numerous colonies of Bacterium pyocyaneum. Erlenmeyer flasks containing a culture fluid of the following composition: distilled water 1000; asparagin5; sodium acetate 5 ; potassium phosphate 2 ; sodium chloride 2 ; magnesium sulphate 0.1, were now inoculated with Bad. pyocyaneum, and incubated for 5 days, during which time there was luxuriant growth and appearance of the characteristic blue-green color. Bean plants taken from pots were now introduced into this fluid. After some days leaves were cut off with a sterile knife, a platinum needle was thrust into the leaf -stalk, and streaks were made on glycerin agar. These remained sterile. Some days later the cultures were repeated, being taken this time from the interior of the stem at 15 cm. from the roots. One plant yielded a pure culture of Bad. pyocyaneum. Those l8o BACTERIA IN RELATION TO PLANT DISEASES. plants which yielded no bacteria at this level were now cut off just above the roots and a third set of tubes inoculated from the inner tissues, all of which now yielded this organism. As this positive result was not in accord with the previous experiments and might perhaps be attributed to injury of the roots during removal from the pots, the following experiment was undertaken. Beans were planted in a nutrient fluid consisting of: water 1000: potassium nitrate 0.5, potas- sium phosphate 0.2, magnesium sulphate 0.2, ferrum sulphate 0.1. They grew well in this fluid and when about 20 cm. high a culture of Bad. pyocyaneum was added. The roots of some of the plants were purposely injured, while the others remained sound. After some days cultures were made from the interior of the plants. These showed the presence of Bad. pyocyaneum in all the injured plants, and in none of the uninjured ones. Ellrodt's conclusion, therefore, is that bacteria can not penetrate sound roots, but may enter through broken ones, and that since root-injuries are common occurrences in soil, it is not yet certain that pathogenic bacteria can not enter the plant from infected soils. Clauditz criticises Ellrodt for not telling in what tissues the bacteria occurred. Fur- thermore, inasmuch as he does not state that he washed or otherwise removed the bacteria from the surface of the plants, he may really have got his results from surface organisms, which were dragged into the tissues. Certainly, the surface of his plants, particularly the parts near the roots and consequently near the bacterial fluid, should have been flamed or otherwise disinfected. In the last mentioned experiment, however, he probably did not get his results from surface organisms because his checks were sterile. Clauditz (1904) made a series of experiments with the typhoid bacillus to learn whether infection through plants is possible and especially whether this organism can penetrate into the interior of plants. In certain respects his statements also are vague. Clauditz used the plants which are commonly eaten raw, viz., radish, cress, and lettuce. The soil was taken from the yard of the Hyg. Institute of the Royal University of Berlin. To imitate as nearly as possible the conditions of the soil in the sewage fields several glass tubes were thrust into the earth a depth of 8 cm., and 24-hour old bouillon cultures of the typhoid organism were poured into these tubes every other day. After 8 days, repeated attempts were made to recover the organism from the soil, but these failed in spite of renewed infections with a fresher isolation. These cultures were made both from the surface and from 4 to 6 cm. down. He states that it is difficult to isolate the typhoid organism from the earth because in most cases this organism quickly perishes when brought into competition with the bacteria of the soil. Following Rullmann's advice he mixed the infected soil with double its quantity of sterile bouillon and incubated at 370, but always the soil organisms got the advantage and the typhoid bacillus was not to be recovered in this way. He then tried to accustom the typhoid organism to the soil bacteria in bouillon cultures by adding to sterile bouillon a loop of a 24-hour bouillon culture of B. typhosus and 2 loops of soil and exposing for 24 hours at 370 C. From this tube 3 loops were then transferred to a second tube which was incubated for the same time and at same temperature, and so on for 10 tubes. After 5 days the typhoid organism was not demonstrable in the first tube, and not after 24 hours in the second, while in all the others the results were negative. The strain isolated from the second tube was designated "Typhus Erde I." With this a second series of 10 tubes was inoculated in the same way as before. The results from this set of tubes were all positive, and even after a half year the typhoid organism was easily demonstrated in tube 10. Along with it were present a variety of other bacteria, Bacillus subtilis, Bad. fluorescein liquefaciens, cocci, etc. A second set of soil inoculations was undertaken with this strain, "Typhus Erde II," the bouillon cultures being now poured into the soil after dilution with sterile distilled water. It was now easy to demonstrate the bacillus in the soil and a strain so isolated was called " Typhus Erde III." The latter was now used for all the subsequent experiments. After the organism had been isolated from the soil and when the plants were 5 to 8 cm. high, they were cut off close to the earth with a sterile knife, washed one-half hour in sterile water, bruised in a sterile mortar with sterile bouillon and then incubated for 24 hours at 370 C. Streaks were then made on Drigalski-medium, one out of four being positive. The experiment was repeated with the precaution first to put the plants in a 1 :ioo solution of mercuric chloride (time not stated) and then wash them thoroughly in sterile water. All the tests were now negative. To avoid the objection that the mercuric chloride may have penetrated the plants and killed the bacteria, the experiment was repeated, the surface of the plants being sterilized this time (so far as regarded B. typhosus) by ANIMAL PARASITES INOCULATED INTO PLANTS. I»I dipping them for 10 to 20 seconds in hot water (900 to 700 C.) All the results were again negative. Some soil bacteria appeared in all the plates which ever way treated. The above mentioned plants had uninjured roots. In additional experiments the roots of the plants were now injured, but the results of the cultures [time intervening not stated] were still negative. New experiments were planned in which the bacterial cultures were let into the subsoil by rubber tubes having side openings. The soil and drains were in large flat pans; on these, earth was laid and this was sowed with lettuce, radish, and cress. When the plants were about 4 cm. high, the infection of the soil through the drains was begun, the fluid being uniformly distributed in all parts of the dishes. The roots of some plants were injured, others not. The results of the subsequent cultures were negative in both cases. It is not stated whether the pans (Schalen) were zinc or copper. An experiment was now made with peas. The conditions of the experiment were as before, except that the soil was infected with the typhoid bacillus before the peas had sprouted. The roots were injured when the plants were 4 cm. high. When they were 10 cm. high the plants were cut off close to the earth, washed in sterile water, crushed in a mortar and streaks made at once on Drigalski- media or Endo-media. The results were as follows: 1 (direct streak) — positive; 2 (Ficker-Hoff- mann's method) — positive; 3, 4 (direct streaks) — negative. Five stems were now examined by culture in their upper and lower parts. Three gave negative results, two showed the bacteria in the lower end (pieces 3 cm. long), but not above. The bouillon in which the stems were washed yielded the bacillus when the interior of the stems did not. There can at least be no doubt, therefore, that bacteria were brought up out of the soil on the surface of the plants, since the plants were watered with the bacillus entirely from below. Another experiment was now made, the soil being first wet from below after the plants were 10 cm. high. Some davs later the roots were broken, and a few days afterwards [scant time allowed] the plants were examined in the same way as before. Only negative results were obtained. The conclusion reached is that the typhoid organism occurs on the outside of the plant and sticks so fast that it can not be washed away. Radish and pea plants were now wet with a suspension of the same culture of the typhoid bacillus for comparison. After 14 days the bacillus was found abundant on the leaves and stems of the peas in spite of direct sunlight, but had disappeared from the radish by the fourth day. The radish leaves appeared to be an unfavorable surface. The surfaces of radish roots grown in infected soil were now tested for the presence of the typhoid bacillus after they had been washed until all visible dirt was removed. In all cases the typhoid bacillus was found in abundance on the surface of such roots. PLANT PARASITES INOCULATED INTO ANIMALS. In 1895 Ostrowsky reported as moderately pathogenic to rabbits a short rod-shaped Schizomycete said to have been isolated from the browned interior of grape-stems. The rabbits were inoculated intravenously (quantity not stated) a slight fever super- vened, there was rapid loss of weight (400 grams in 8 to 10 days) and on autopsy there were sometimes small miliary abscesses in the spleen or liver. The organism is not well described. It liquefies gelatin; the colonies are moist, soft and whitish on agar. It is apparently aerobic. In old gelatin cultures there is a brown stain. No evidence is offered in support of the statement that it is parasitic on the vine, and as a matter of fact Viala and Ravaz state that it was not, i. e., no disease of the vine could be induced with it by means of inoculations. In 1899 Charrin and Viala state that when the microbe of gelevure, otherwise known as Mai nero, Gommose bacillaire, Maromba, Maladie d'oleran, etc.. was first inoculated it caused at most the death of some fish, but by repeated inoculations into rabbits it caused frequently a slight enteritis with some fever and loss of appetite, etc., ending in recovery in the greater number of cases. The germ is said to require education. It is not described. Owing to its speculative character and the absence of all details as to the exact nature of the experiments, the paper does not tend to win the confidence of the reader. There are many opportunities for error. Moreover, the etiology of Mai nero itself is still in doubt, with the probabilities against its being of bacterial origin. According to Dr. V. A. Moore Bacillus cloacae, supposed to be the cause of a disease in maize, has no effect on experimental animals except when injected into the blood stream in large quantities. 1 82 BACTERIA IN RELATION TO PLANT DISEASES. Harding inoculated Bact. campestre into animals with negative results. Harrison inoculated Bacillus solanisaprns into guinea-pigs and rabbits without positive results (see Vol. Ill, Basal Stem-rot of Potato). The following experiments with fishes in water containing bacteria of plant diseases were undertaken for me in 1905 by M. C. Marsh, bacteriologist of the Bureau of Fisheries, in Washington. I prepared the cultures myself. The summary is his: Bacillus aroidcac. — About 25 liters Potomac tap water in glass aquarium jar, with constant flow of air in small bubbles delivering at the bottom. Two sunfish, 1 small goldfish, and 1 mummichog were used with 14 cc. of a well-clouded 2-day bouillon culture of B. aroideae introduced on first day, February 17, 1905; March 5, 15 cc. of a 5-day culture added. March 13, after 26 days, all fishes alive and in good condition. March 5, injected largest of the above sunfish behind eye with about 0.5 cc. of a 5-day well- clouded bouillon culture of B. aroideae; producing great exophthalmia. Sunfish remained in jar containing B. aroideae mixed with water. March 13, fish aiive, eye normal; time 8 days. Bacteria of carnation leaf spot, February 19, about 25 liters Potomac tap water in glass aquarium, aeration as above. One small black bass (5 inches), 2 sunfish, 1 mummichog. February 20, added two 20-day bouillon cultures of carnation bacteria, one 5-day bouillon culture, and one 20-day agar slant culture; February 28, added three 9-day bouillon cultures. March 13, after 21 days, all fish in good condition. Temperature of water 14. 50 to 220 C. No change of water during experiments. Fishes fed very sparingly. In transmitting the above report Mr. Marsh made the following comment : I send you herewith a statement of the effect of the plant bacteria on fishes, from which it appears that the effect is nil. I did not make a direct inoculation with the carnation rot, on account of the result with the presumably more dangerous calla rot. It is not likely that these organisms would harm any fishes, though I was unable to try trout. The eye inoculation should have taken if there was any pathogenicity about the calla rot for fishes. Inoculations of Bact. tumefaciens into Fish and Frogs. In the spring of 1908, the writer made fourteen sets of inoculation experiments on fish and frogs with pure cultures of Bact. tumefaciens derived from tumors on the hothouse daisy (Chrysanthemum) to determine whether this organism would induce similar abnor- mal growths in cold-blooded animals, experiments on warm-blooded animals being con- sidered unnecessary because of the low maximum temperature of the organism (about 36.5°C). These inoculations were carried on in Washington in houses belonging to the Bureau of Fisheries with trout and roach kindly placed at my disposal from the stock tanks and with frogs bought from a Washington dealer. With exception of those used in Experiment IX, the trout were 8 to 10 inches long and were ordinary brook trout (Salvelinus fontanalis) . The roach (Abramis chrysoleucus) were about 6 inches long. There were no checks on the roaches or frogs. The checks on the trout consisted of a school of about 100 fish of the same age and condition, and of which those I took had previously formed a part. These were in one of the ordinary exhibition tanks of the Bureau along with some rainbow trout. They were not checks in the strictest sense of the word because they were not wounded in any way. Inoculations of March 20, 1908. — These were made from four agar streak cultures 48 hours old. None of them were hypodermic injections. I. Two trout. Each two needle-oricks in the eye-socket. II. Two trout. Each three needle-pricks in the region of the anus. III. Two trout. Each three or four needle-pricks in the fleshy fin (adipose). PLANT PARASITES INOCULATED INTO ANIMALS. 1 83 IV. Two trout. Each two shallow pricks in the throat outside near the junction of the gill arches. V. Two trout. Each inoculated in the eye-socket. This time the skin was cut with a scalpel and a 2-mm. loop of the white bacterial slime was inserted. VI. Two trout. Each inoculated inside of the mouth at the base of the tongue by means of several needle-pricks. VII. Four roach. Each inoculated in the eye-socket. The skin was cut and a 2-mm. loop of the bacteria inserted. VIII. Two roach. Each received several needle-pricks in the vicinity of the anus. IX. Fifty salmon trout fry (2 to 3 cm. long). They were put over night into 2 liters of water into which the remnant of the 4 agar cultures had been washed. The next morning they were trans- ferred to running water in the ordinary shallow hatchery boxes. Inoculations of April 4, 1908. — For these, 5 slant agar cultures 3 days old were used. The copious growth was washed off into 20 cc. of distilled water, making a milky suspension. All of the inoculations were made with the hypodermic syringe. X. Three leopard frogs. Each received !cc. of the very cloudy fluid. This was injected into the muscles of the right thigh. XI. Three leopard frogs. Each received 'cc. under the skin on the abdomen. In one of these it was thought that the needle entered the abdominal cavity and for this reason it was kept separate, but the result was not different. XII. Two leopard frogs and 3 green frogs (Ran a clamitans). Each received Jcc. in the right eye-socket. XIII. Three brook trout. Each received ?cc. in the eye-socket. XIV. Three brook trout. Each received |cc. of the very cloudy bacterial suspension. This was injected into the peritoneal cavity, the needle being set in just behind the ventral fins. XV. The virulence of the cultures was determined by making inoculations on four young daisy plants. These promptly contracted the disease. On June 1 these plants bore, where inoculated, tumors which were over an inch long by 0.50 to 0.75 inches broad and thick. Results: The results so far as tumor production is concerned were either negative or uncertain; all the fish have not been sectioned. The experiment was complicated by the discovery after the inoculations were begun that some of the fish were suffering from car- cinoma of the tongue, gills and thyroid region. Thenceforth, I examined each fish and inoculated only such as appeared to be sound, but nevertheless some of them may have then been about ready to develop such tumors as subsequently appeared. On March 25, 1908, I caught and examined 50 of the 100 check fish and found 3 with carcinomatous throat tumors. Additional cases of this disease appeared in the checks, especially as the season advanced. Consequently the tumors which developed on the inoculated fish may have been due altogether to this disease, at least under the circumstances I could not be certain that they were not so caused. The frogs proved very resistant. None developed tumors. Most were finally chloro- formed at the end of the experiment. The few that died earlier gave no plain evidence of being in any way injured by the bacteria. Of the roaches one died the day after inoculation. The rest died in from 17 to 32 days. None of the latter developed tumors. All were more or less inflamed, both roaches and trout. The inoculated trout (except the fry which showed no signs of disease attributable to the bacteria) died off faster than the checks in the main tank. They were, however, not under altogether the best conditions, i.e., they were rather too crowded at the beginning of the experiment and the water was several degrees too warm toward the close of the experi- ment, but at the same temperature as that given to the check fish. If I were to do over this experiment in a climate like that of Washington, I would begin in the autumn, so as to allow the experiment to run for at least six months in cool weather. 1 84 BACTERIA IN RELATION TO PLANT DISEASES. Of the inoculated trout one died at the end of 15 days, died in from 30 to 40 days from the time of inoculation, inoculated and check trout is shown in the following table : The rest (with one exception) The relative rate of death of Inoculated Trout. Table Showing Date of Death and Svmptoms Observed in Trout. Check Trout. Date. No. Remarks. 1908. Apr. 4 Apr. 22 Apr. 23 Apr. 25 Apr. 28 Apr. 29 VI. 1 II. 1 VI, 1 I, 1 XIV, 1 II, I IV, 1 V, 1 IV, 1 XIV. 1 Apr. 30 IV, 1 May 1 III. 1 May 2 I, 1 May 4 XIII, 1 May 5 May 7 V, 1 Congestion in region of anus, base of tongue, liver, heart. Eye-socket inflamed. Liver diseased. Hard yellowish rough tumor on tongue (the only hard tumor). Eye- socket inflamed. Walls of lower intestine inflamed, ulcers on inner belly wall and externally at root of pectoral 6n and on outside below anal fin. Throat sound. Anaemic. Ulcerous tongue and inner belly wall. Anaemic. Eye-sockets inflamed. Vicinity of thyroid inflamed and slightly swollen. Eye-socket inflamed. Liver white-mottled. Spots on gills and region of the thyroid inflamed. Stomach and intestines much inflamed. White patch on liver. Back part of throat inflamed, i. c, below the tongue. Inflammation in throat where needle-pricks ended; None outside. Marked inflammation of abdominal wall where needle entered. Needle wound healed externally. Inflamed spot at base of tongue. Lower intestine much inflamed. Bloody patches on lower intestine. Ulcerontail. Several smalt ulcers on surface. Throat sound, liver diseased, in- flamed patches on inner wall of abdomen. Small abrasion on surface, also a Saprolegnia patch (3 sq. in.). Gills, stomach and intestine congested. Eye-socket badly inflamed. Eye on inoculated side is white-clouded, swollen inflamed tissue at base of eyes, especially on inoculated side. Marked congestion of the inner wall of abdomen in lower part; tongue somewhat swollen and red. Small sore spot at roots of the tail. Throat sound. Membrane cover- ing the intestine and eye-socket highly inflamed. Also inflamed places on the inner lower belly wall. No tumor on the adipose (where inoculated) . Base of tongue swollen and inflamed outside and inside. From Mar. 20 to Apr. 29. no deaths among the 100 check fish. During this period 10 deaths among the 18 inoculated fish. On Mar. 25, 3 showing throat tumors were separated from the rest of the checks. Apr. 30 1. in throat. The first check to die. Tumor May n XIII, 1 Throat and gills sound, eye-socket inflamed, inflamed. Inner belly wall badly inflamed. viscera swollen and May 19 XIV One alive on this date when experiment was abandoned. May 6. Two, with cancerous growths in throat. May 8. 1. Cancerous thyroid. May 9. 1. Mouth sore. May 11. 7. Two have sores in the mouth. May 14. 3. One ancemic and with an ulcer on the gills. Weather hot for last three days. May 15. 1. Cancer in thyroid region May 16. 4. One has a cancerous throat. May 19. 7. Not dissected. Since the above paragraphs were written the most hopeful portions of the inoculated trout have been infiltrated, sectioned, stained, and studied, with the following results (the figures in parenthesis refer to the preceding table). 519 (April 28, V). Ulcer on inner wall of abdomen. Proliferations too regular for sarcoma. Probably not malignant. 520 (April 25, XIV). Hypodermic. Ulceron inner belly wall near entrance of needle. Sarcoma(?). Very suspicious. 546 (April 23, IV). Proliferations — not malignant. 549 (April 25, II). Tumor on tongue: Adenocarcinoma. 621 a (April 23, I). Portion of tongue. Probably not malignant. Some of the cartilage has an abnormally large number of cells in it (chondroma?) . 622 (April 12, II). Inflamed part of inner wall of abdomen. Proliferations but nothing definite. 642 (April 19, XIV). Sections of small swellings on inner belly wall where inoculated. Mav be accounted for as simple inflammation. Later: There arc giant cells in it. 625 a (May 4, XIII). Hypodermic, eye-socket. Ulcer at base of eye. Typical giant cells, but possibly foreign body giant cells. DO PLANTS HARBOR ANIMAL PARASITES? In a paper published in 1889 in the Comptes Rendus of the French Academy, Dr. Domingos Freire undertakes to show that roses and various other common flowers harbor bacteria, some of which are pathogenic to man. All this paper really proves is that which PLANTS AS CARRIERS OF DISEASE. 1 85 was already well known, viz., that bacteria are commonly present in the air and conse- quently liable to be found on the surface of everything, not excepting the surface of the flowers with which the author experimented. Florists and flower lovers need give them- selves no particular uneasiness, since Dr. Freire's conclusions do not necessarily follow from his experiments. As to the correctness of his identification of species, the paper offers no means for judging. The author's conclusions are: (1) That animal pathogenic bacteria, Streptococcus pyogenes, Bacillus pyocyaneus, etc., are normally present in flowers which can, as he says, " noteworthily store up numerous germs, which may subsequently finish their development in the better adapted tissues of animals or plants;" (2) that there is some hidden relation between the colors of flowers and the bacteria found on them, e.g., the color of the colonies of Leptothrix ochraceae and the very pale rose color of the Rothschild rose, or the egg yellow color of Micrococcus crucijormis and the yellow of Hibiscus rosa-sinensis; (3) that certain bacteria, called by him osmogenes "reproduce odors analogous to those liberated by the essential oils of the flowers where they live. " Two new species are very imperfectly described, Micrococcus crucijormis from the anthers of Hibiscus and Bacillus gallicus from Rosa gallica (centifolia) . From Ipomoca quamoclit L. he isolated an organism having the characters of Micrococcus salivarius pyogenes and another identified as Spirillum plicatale. From the flower of the peach he also ob- tained something identified as Bacillus pyocyaneus. Uffelmann's experiments (1892) showed that the cholera vibrio might be disseminated on the surface of fruits and vegetables. -&v He moistened the surface of a ripe apple with some drops of liquid from a cholera stool which dried within 15 minutes. Then after 5, 10, and 20 hours he transferred small particles of this con- taminated skin direct to gelatin roll cultures and to tubes of bouillon for 24 hours at 35° C, after which gelatin rolls were made. In each case numerous colonies were obtained. After 24 hours, however, only a few colonies were demonstrable, and after 30 hours, very few. After 48 hours no colonies were obtained. From the surface of an apple treated in the same way except that it was kept under a bell jar, cholera bacilli were demonstrated up to the end of the fourth day. Similar experiments were performed on the leaf-stalk of a cauliflower. Two infections were made, one (I) on the exposed base of the petiole, the other (II) on the midrib where the blade of the leaf bent over so as to protect the spot from rapid drying. The spot (I) which was still somewhat moist, yielded numerous colonies after 24 hours, but none at the end of 48 hours, when it was fully dry. The spot II was tested at the end of 24, 48, 72, and 96 hours. Living cholera bacilli were present on it at the end of 24, 48, and 72 hours. They were not demonstrated at the end of 96 hours, and were sparingly present at the end of 72 hours. The experiments of Wurtz and Bourges (1901) were undertaken to determine whether the culture of certain vegetables should be forbidden on the sewage fields of Paris. Their technique was as follows: Pots were filled with soil and sowed with seeds of cress, lettuce, and radish. Immediately after, the earth was watered with suspensions of the given microbe, taken from agar cultures or potato cultures. Ordinary plate cultures gave no positive results, owing to the prodigious number of colo- nies developing from soil bacteria present on the stems. Only in case of three pathogenic bacteria did they obtain positive results from the tips of the leaves by the use of special methods. In three series of experiments with anthrax they obtained cultures from the leaves by first heating them for 3 minutes at 8o° C, and then making gelatin plates. The anthrax organism was recovered constantly and easily up to 3 weeks from the time of the sowing. The typhoid fever organism was recovered by placing the leaves in ordinary carbolated bouillon at 420 C. for 2 days, after which gelatin plates were poured. With pure cultures from these plates agglutination tests were made. The results were positive in 30 cases out of 30. The tubercle organism was detected by inoculating into guinea-pigs fragments of leaves bruised in sterile water. The first series gave 1 positive result out of 4. The second gave 18 positive results out of 30. Experiments with the colon bacillus miscarried. These experiments were carried on in the laboratory, at room temperature. The plants were exposed to the sun only a short time each day, and wTere not washed by rainfall. Consequently: 1 86 BACTERIA IN RELATION TO PLANT DISEASES. II y a en effet du danger a faire ces experiences portant sur des microbes pathogenes en pleine terre. To obtain conditions more nearly approximating those found in the fields they experimented with Bacterium violaceum and the red Kiel bacillus. Seeds of radish were sowed in the open field, were covered with about 2 cm. of earth and were watered with a culture of Bad. violaceum suspended in sterilized water. It was at this time very warm and dry. The radishes were watered very abundantly by showering with ordinary water. Fifteen days after the sowing the plants were lifted and a search made for the bacillus, but it could not be found either on the surface of the plants or in the earth where it had been put. The same experiment was repeated using the Kiel bacillus, but dividing the field into two parts, one of which was shaded, and the other exposed to the sun. The weather continued very hot and the artificial waterings were abundant. The Kiel bacillus was recovered on plate cultures from the sur- face of the plants grown in the shade, but not from those exposed to the sun, so that we must assume either that the sun destroyed the bacteria, or did away with their pigment production. In the above experiments [probably those with the pigment-forming bacteria] the plants were exposed to the infec- tious water after they had begun to develop. To test the ability of the plant to bring up bacteria out of the depths of the soil they made the following experiments. Potato tubers were wet on the upper surface w7ith a virulent, sporulating bouillon culture of the anthrax organism (about one culture for each three potatoes). They were then placed in boxes of earth and covered with from 5 to 10 cm. of vegetable earth. From March 2 1 to July 1 these potatoes were watered with large quantities of water, as well before as after the development of the shoots. During growth the plants were watered several times a week. In May and June the plants were exposed to sunshine from morning until 3 p. m. Fragments of the leaves and stems were taken with sterile instruments and washed in bouillon which was then heated for 5 minutes at 8o° C. This bouillon was then used for gelatin plates, 10 drops being put into each plate. In this way the anthrax organism was recovered 41, 93, and 101 days after the sowing, and at heights of from 4 to 30 cm. above the earth. The anthrax organism was constantly present, but usually only a few colonies developed, on an average 1 or 2 per plate. At the end of 3 months these anthrax spores had lost half their virulence. To reassure themselves of this and to do away with the effect of other soil organisms (Vibrio septique, tetanus, etc.) this experiment was repeated in sterile soil. Anthrax colonies taken from the tops of the stems of potatoes grown in this soil had lost their virulence. In spite of the action of light and of rainfall, there is, therefore, according to Wurtz and Bourges some reason for suspecting vegetables grown on lands devoted to the purification of sewage, even though theoretically the sewage be not deposited in a sheet on the surface of the soil, since plants, as well as animals, may serve to bring pathogenic bacteria to the surface of the earth. CROSS-INOCULATIONS. PLANTS AS CARRIERS OF DISEASE. I87 LITERATURE. 1889. Grancher, J. ET Deschamps, E. Recherches sur le Bacille typhique dans le sol. Archives de medicine experim. et d'anatom. patho!., 1899, ire. Ser. Tome 1, pp. 33-44. See epecially pp. 43-44- 1890. Lominsky, F. I. On the parasitism of some disease-creating microbes when introduced into plants. [Russian.] Wratch, 1890, No. 6, pp. 133-135. [Also written Vrach.) 1890. Lominsky, F. Ueber den Parasitismus einiger Krankheiten erzeugender Mikroben. 8vo. 76pp., with drawings and two plates. (Uni- versitats-Nachrichten der Universitat, Kiew, Jahrg. 1890, xxx, No. 10 [Russian], [Not seen.] 1892. Russell, H. L. Bacteria in their relation to vegetable tissue. A Diss, presented to the Board of University Studies of the Johns Hopkins University for the degree of Doctor of Philosophy, Baltimore, 1892, pp. 1-41. 1892. Uffelmann, J. Beitrage zur Biologie des Cholerabacillus. Berliner Klinische Wochen- schrift, 1892, xxix Jahrg., p. 1212. 1893. D'Arsonval ET Charrin, A. Action des microbes pathogenes sur la cellule vegetale. Compt. Rend. hebd. de la Soc. de Biologie, se. du 14 Janvier, Paris, 1893, p. 37. Describes the effect of "le bacille pyocyanique " on beer- yeast in test tubes with "l'eau sucree." At 370 C. the alcoholic fermentation was stopped, at io° C- it continued. (See also pp. 121. 237, 337) 1893. Charrin, A. Le bacille pyocyanique chez les vegetaux. Compt. Rend., des se. de l'Acad. des Sci. Tome cxvi. No. 19, pp. 1082-1085, Paris, 1893. Reviewed in Jahresbericht uber d. Fortschritte in d. Lehre von den Gahrungs- organismen, 1893, p. 104, and in Centralb. f. Bakt. etc., 1893, Bd. xiv, pp. 456-458. 1893. Russell, H. L. Non-parasitic bacteria in vegetable tissue. Bot. Gazette, March, 1893, vol. xviii, pp. 93-96. 1895. Ostrowsky. Bacille pathogene dans les deux regnes, animal et vegetal. Compt. Rend., hebd. de la Soc. de Biol., 1895, Tome 47 (10th Ser. T. 2), Paris, pp. 517 to 518. 1S96. Kasparek, Th. und Kornauth, K. Ueber die Infectionsfahigkeit der Pflanzen durch Milzbrandboden. Arch. f. die gesammte Physiol., des Menschen u. der Thiere. Bd. lxiii, 1896, pp. 293-300. 1896. Kornauth, Karl. Ueber das Verhalten pathogener Bakterien in lebenden Pflanzen- geweben. Centralbl. f. Bakt. etc., 1 Abt., Bd. xix, 1896, pp. 801-805. 1897. Zinsser, O. Ueber das Verhalten von Bacterien insbesondere von Knollehenbaeterien in leb- enden Pflanzliche Geweben. Jahrb. fiir Wiss. Botanik. Berlin. 1897, Band xxx. Heft 4, pp. 423-452. 1898. Hartleb, R. Ueber die Infectionsfahigkeit lebender Pflanzen mit dem bei der Maul — und Klauenseuche vorkommenden Bacterium. Centralbl. f. Bakt. etc., 1898, 2 Abt., Bd. iv, pp. 26-30. 1899. FreirE, Domingos. Les microbes des fleurs. Comptes Rendus des se. de l'Acad. des Sci. Paris, 1899, Tome cxxvm, pp. 1047-1049. 1899. Charrin, A., et Viala, P. Le microbe de la gelivure et la pathologic generate des deux regnes, animal et vegetal. Rev. de viticulture, 1899, No. 279, pp. 425-427. 1899. Laurent, E. Recherches experimentales sur les maladies des plantes. Ann. de l'lnst. Pasteur, Jan., 1 899, Tome xin . Also a separate. Inoculations with saprophytes. 1901. Wurtz, R. ET Bourges, H. Sur la presence des microbes pathogenes a la surface des feuilles et des tiges des vegetaux, qui se sont devel- oppesdans un solarrose avecde l'eau contenant ces microorganismes. Archives de med. experim. et d'anat. pathol., Paris, 1901, Tome xin, pp. 575-579. 1902. Ellrodt, Gustav. Ueber das Eindringen von Bakterien in Pflanzen. Centralbl. f. Bakt., 1902, 2 Abt., Bd. ix, pp. 639-642. 1904. Clauditz, H. Typhus und Pflanzen. Hygien- ische Rundschau, xiv Jarhg., Berlin, 1904, pp. 865-871. HYGIENE OF PLANTS. Nothing perhaps comes out plainer in a study of diseases of plants, bacterial diseases included, than the fact that such diseases are very often introduced on species or varieties brought in from other localities. No one, for example, would care to buy seed-potatoes from a field subject to brown rot or basal stem-rot, cabbage-seed from plants affected by black rot, sweet corn seed from fields subject to Stewart's disease, olive trees with olive- knot, pear-trees with pear-blight, plum-trees carrying with them the black spot organism, or peach-trees grown in nurseries subject to crown-gall, and yet without knowing it planters are doing these things all the time. Viewed in this light, the introduction of new things from all sorts of places is not an unmixed good. Many diseases are spread in this way. Especially is the importation in bulk and the immediate general distribution of all sorts of seeds and plants to be deprecated. It would be much safer to import seeds and plants in small quantities and multiply them for a year or two under strict Government supervision in experiment gardens before making a general distribution. The dissemination of many scales and other injurious insects would also be prevented by this method. Up to this time, with local exceptions, e.g., France, Germany, California, growers in all parts of the world have been allowed to import and distribute at will. The growers in the United States in particular do almost exactly as they please. The Department of Agriculture also has sometimes imported and distributed without proper inspection. The machinery of inspection is not properly organized in this country. To do such inspection thoroughly would require a small army of trained inspec- tors (entomologists and pathologists) distributed at at least a dozen different ports of entry, all subject to one efficient central inspection bureau. Universities are not turning out men fast enough to meet the demands for this sort of work, and at the present time there are not men available in this country to carry out properly any such system of inspection, important as it is to have it instituted speedily, nor even to meet the ordinary requirements of pathological research. As time goes on undoubtedly strict inspections will be required in all highly civilized countries, and the propagators and distributors of trees, shrubs, herbs, tubers, bulbs, seeds, etc., will be required to give some sort of guarantee as to where the plants were grown and under what conditions. Then seedsmen will not be permitted to sell seeds raised in infected districts and often harvested from infected plants, simply because it is convenient for them to do so, nor will nurserymen be allowed to sell stock known to be infected, for no better reason than simply because they have a large quantity on hand and wish to dispose of it. Meanwhile, in the absence of proper inspections, intelligent buyers will deal only with reliable firms, and will in addition seek for some specific assurance as to healthfulness. In case of large orders a visit to the plantations themselves before the trees are dug, or the seeds harvested, might occasionally prevent much subsequent vexation and loss. Some propagators of seeds and plants appear to be entirely indifferent to the welfare of the community, sometimes distributing things known to be infected, and there should be a severe law for such people.* The greater number, however, undoubtedly trespass through ignorance, and we can not hope for a general improvement of the seed and plant trade in these particulars until a knowledge of these diseases, particularly information as to their dangerous nature and the exact methods of their dissemination, is broadcasted through the trade, and made effective through the demands of the buyers for healthy stock. The buyer must be on his guard continually. He knows theundesirability of cocklebur, ragweed, and thistles, but in most cases he has not yet come to realize the greater undesirability of *The San Jose scale was disseminated through the eastern United States by nurserymen of this type. 1 88 HYGIENE OF PLANTS. 1 89 these microscopic pests. He should write them down in his list of "worst weeds, " and try to keep his fields free from them. There ought also to be some legal means of redress when a nurseryman or seedsman has taken hard-earned money and in return has infected a man's land and rendered his business unprofitable, but in most cases there is not. The ounce of prevention, therefore, is the thing to be thought of, and more and more the farmer must consider the advisability of suitably disinfecting trees and seeds before planting them, unless he knows their source to be a safe one. Many dealers who own propagating farms also buy large quantities of stock from other growers, so that the farmer seldom knows from what part of the country his plants have come. He may think he is buying from an uninfected region while in reality his trees may have come from diseased localities. In case of two kinds of seed sold extensively by the trade, viz., sweet corn and cabbage, it is notori- ous that they are propagated for seed largely in districts where no such seed should be grown, because the plantations often reek with disease, and the germs of these diseases (Stewart's disease of corn and the black-rot of cabbage) are liable to be distributed on the seeds. Excess of water undoubtedly renders many plants more susceptible to bacterial dis- eases. The evidence here is very good in a number of cases. It is now a well-known fact first observed, I think, by L- R. Jones, working in the writer's laboratory, that the soft-rot organisms need tissues filled with water in order to make rapid progress. In pear-blight slow growth is favorable to freedom from the disease and excessive moisture leading to rapid growth renders the plant much more susceptible to disease: This has been observed over and over again in many localities. Russell obtained black-rot of cabbage more readily on well watered plots Halsted observed the same thing in beans. The writer has seen the same thing in sweet corn and in tomatoes. It is, therefore, desirable that moist soil should be properly underdrained and that irrigation should be managed with great care. Lack of subsoil drainage is very favorable to the development of root rots of all sorts espe- cially in rainy seasons. Not infrequently an entire field of potatoes rots within a week from this cause. Excess of water on the foliage also leads to numerous infections as in the case of black spot of the plum, bean-spot, begonia-spot, etc. Where the plants are out of doors this can not be avoided, altogether, but under glass it can be modified by care in throwing water and by so arranging the houses that all parts shall have proper ventilation. Dis- eases often begin in ill-ventilated parts of the hothouse. Overcrowding may also be a cause of disease in some instances, especially if it leads to imperfect ventilation and the persistence of water upon the foliage. In one instance the writer observed a bacterial rot to be favored by excess of shade, namely the iris rot. Under his observation this was very serious one year in heavily shaded parts of a garden, and not at all present on the same grounds in clumps of the same iris exposed fully to the sun. Slow evaporation of water was probably the predisposing cause. Excess of manures, especially of those containing nitrogen, renders the plant more susceptible to cold and probably also to disease by throwing it out of physiological balance but I can cite no specific instances of bacterial diseases particularly favored by this in any way other than by the formation of rapidly-growing juicy tissues. Infected manures are to be avoided carefully. By this I mean barnyard manures with which diseased plants have been mixed. The writer saw a field of cabbage in Michigan, a small portion of which was much worse diseased by the black-rot than the remainder. This portion contained perhaps an acre or two, and the only difference between it and the remainder of the field appeared to be that it had received as a manure the refuse from a neighboring cab- bage storehouse, in which heads affected by the black-rot had been placed the preceding fall. The potato-rots are diseases likely to be transmitted in this way since diseased potatoes are often fed to stock or thrown out on manure piles. Stewart's disease of sweet corn is another disease likely to be distributed in this way since the farmer often feeds the stalks to cattle, and these are swarming with the organism which causes the disease. The bacterial spot of beans is another. 190 BACTERIA IN RELATION TO PLANT DISEASES. Negligent pruning in some instances may be responsible for the distribution of disease. There is not much doubt that pear-blight may be distributed from one tree to another by means of infected pruning shears. Since this sentence was written D. H. Jones has done it with a pruning saw. Probably olive-tubercle can be spread in the same way : The Italians think so. Apple gall is favored by carelessly made grafts (Hedgcock). Daisy knot fre- quently appears on the wounded end of cuttings. The wounding of the roots of plants in transplanting from the seedbed to the field is a fertile source of infection. The most striking examples of this are the distribution of to- bacco wilt and of tomato wilt. It would be best, if possible, to avoid transplanting into fields subject to these diseases. If such fields must be used, then the seeds should be planted where the plants are to be grown or the transplantings made with unusual care. Any wound- ing of the roots in such soils is highly detrimental because the plants seem to be able to keep out the soil organism so long as the roots are not broken, but when these are injured there is an open passageway into the vascular system of the plant which the parasite is not slow to find. It has been observed (Hunger) that this wilt disease is worst on fields occupied by plant-infesting nematodes, the direct loss due to these worms being a small part only of the actual injury since the wounds they make form an open passageway through which the bacteria enter the plants and destroy them. A great desideratum is some easy means of combating in the soil this type of nematodes. The early removal and destruction of trap crops is supposed to be a partial remedy. On small areas, e. g., in houses or under tents, they may be destroyed by the use of steam. Having said thus much about nematodes it remains to say a word about the destruc- tion of insect carriers of disease. We know, from Mr. Merton B. Waite's experiments, that pear-blight is commonly distributed by bees and flies. My own experiments have, I think, settled the fact that the wilt of cucurbits is distributed by Diabrotica vittata. Brenner plated Bad. campestre from an aphis allowed to puncture diseased veins of a cabbage-leaf. Jones has recently plated Bacillus amylovorus from aphides feeding on blighting shoots of the apple. The question is yet, perhaps, an open one whether aphides are responsible for the general distribution of many bacterial diseases. The same is true for various other bugs which have been incriminated, c. g., the squash-bug. Exact experiments are still wanting. On general principles, however, it is desirable to keep down the prevalence of biting and puncturing insects both in hothouse and field by the use of suitable insecticides. Change of crops is an important means of combating many diseases. This is par- ticularly true of bacterial diseases if we may accept Laurent's idea as in any degree repre- senting what really takes place in the soil. His laboratory experiments led him to believe that non-parasitic forms are often converted into transitory parasites by change of food, i.e., by growing for a time saprophytically on portions of the plant they become able to attack it parasitically. If, on the contrary, the organisms which have attacked plants are com- pelled to live for a time on other foods they lose the parasitic habit, i. c, become non-viru- lent. If this is true, fields which have developed a disease should be plowed so as to aerate the soil and thus hasten the decay of diseased roots and stems. Moreover, if possible the field should be put to other crops for some years. Rotation of crops is an old subject much written upon, but not yet sufficiently impressed upon the multitude of planters. To follow one crop with another of the same sort and that year after year invites disaster. We do not yet know much about the proper chemical treatment of soils to put them into the best condition to resist a particular disease, i.e., use of lime, acid phosphates, etc., but there is undoubtedly much to be learned. Of this fact I think there can be little doubt, namely, that bacterial plant parasites die out of some soils more readily than out of others. This chapter may well close with a pregnant sentence from Louis Pasteur: II est au pouvoir de l'hommc de faire dispairaitre de la surface du globe les maladies parasitaires. RECOVERY BY EXCISION. 191 RECOVERY BY EXCISION. It would be difficult to say when excisions were first practiced for the protection of plants. The method is probably as old as gardening itself. It is a common practice in China. It has been, however, purely empirical. The systematic practice of excision of plant-parts with a particular end in view, that end based on a scientific knowledge of the habits of the parasite to be overcome, is com- paratively modern. Wakker seems to have been the first scientific man to try this for a bacterial disease of plants. Knowing that the movement of the bacteria was downward in the vessels, he re- moved the leaves of hyacinths showing the beginnings of bacterial disease at their apex, and thus prevented the infection of the bulbs. Dr. Russell and the writer both showed that infection of cabbage-heads by Bacterium campestre could be prevented by removal of leaves only recently out of the water-pore stage of infection, the movement of the bacteria in this case also being downward. For details on excision experiments made by the writer to control wilt of cucurbits, see page 276. By the very prompt removal of affected leaves or branches valuable plants may sometimes be saved, but usually the wilt is not discovered until the bacillus has entered the main stem and then excisions are of no avail. The most successful example of protection by pruning known to the writer is the winter treatment of fire-blight of the pear, devised by Merton B. Waite. Mr. Waite discovered that Bacillus amylovorus winters over in pear-trees, but only in a small proportion of the whole number of the trees attacked, one tree, let us say, in fifty, or one in a hundred, the exact proportion is immaterial. From the gummy exudate out of patches of bark on these trees, through the agency of insect-carriers (bees, flies, etc.) the bacteria are distributed again the following spring to other trees in the orchard and the blight renewed — first, usually, as blossom-blight. Mr. Waite has not been able to find any other source of the early spring infection. These cases of "hold over blight," as he has called them, can be detected by sharp eyes and removed while they are in a dormant condition; and, theoreti- cally at least, if all were thus removed, there would be no source of infection the following spring, and consequently no new blight. The correctness of this hypothesis was first tested on a considerable scale in Georgia, where Mr. Waite succeeded in saving a large pear- orchard which had been threatened with complete destruction. Every visible vestige of the winter-blight was removed from these pear-trees several years in succession, and the disease ceased to be a cause of anxiety. In fact, the only cases that appeared thereafter in this orchard were scattering examples traceable to an occasional case of hold-over blight which was missed during the winter examinations, and to such infections as wandered into the borders of the orchard from the small neglected orchards of careless neighbors. The experiment was considered to be a brilliant success, both by the pathologist and by the owner of the orchard. In more recent years Mr. Waite and his assistants have endeavored to apply this method on a large scale to the pear-orchards of California. When he was urged to under- take the task, about one-fourth of the orchards of the State had already been destroyed by this disease, and the remainder were threatened with destruction, the disease being widely disseminated, although of recent appearance, and very aggressive. To be asked to do for a whole State what he had done before for a single large orchard might well appal any man. The loss of these orchards meant, however, a money loss to the State of California of per- haps ten million dollars. Mr. Waite, therefore, energetically undertook to save them. The task, however, was a gigantic one since it involved critical inspection of every pear-tree in the State, and the removal of all diseased parts during the winter season. It involved also the education of the pear-growers of a whole State, the combating of much ignorance, 192 BACTERIA IN RELATION TO PLANT DISEASES. the persuasion of many obstinate, doubtful individuals. For this purpose not enough money was available. As a result of the first year's work (winter of 1905-6) the disease was cheeked in many places, but the surgeons were too few and the cooperation of all the growers could not be obtained. The blight, therefore, still prevailed in many places. In succeeding years, by more vigorous efforts and a general campaign of education, many of the more intelligent pear-growers now assisting in the work of eradication, the disease has been so restricted as to afford a further demonstration of the immense value of systematically conducted winter-excisions.* Mr. O 'Gara has carried on a similar campaign of eradication against blight in the apple and pear orchards of Oregon (Rogue river valley) with the very best results in the protection of the magnificent orchards which there also were threatened with destruction. Even in localities in California where practically everything has been swept away by blight single growers of great energy and intelligence have saved their pear orchards by systematic removal of blighting wood. The pear orchard of 6,000 trees owned by Mr. Reed at Marysville is a good example. This is now almost or quite the only orchard of any size left in that region, the others having been destroyed by the blight. It is not too much to say, therefore, that excision is a remedy of prime importance in case of pear-blight, and that the disease can be held in check thereby. The principal difficulty lies in obtaining that unity of action throughout a community which is absolutely essential to full success. In case of the olive-tubercle equally vigorous treatment would probably yield equally good results. This statement is based on hothouse observations of inoculated plants metas- tasizing freely. In 6 or 8 months from the first appearance of a primary tubercle the organism causing the disease may pass downward through the stem in young and vigorous plants a distance of 18 to 36 inches or more, rupturing to the surface at intervals in the form of deep seated secondary tubercles. It is therefore of the utmost importance for the removal of this disease to prune early and severlyso as to reach beyond the unseen advancing bacteria. GERMICIDES AND INSECTICIDES. The surface sterilization of plants or parts of plants has been treated in Vol. I, so far as regards its use in connection with the isolation of pure cultures. To what has been said there may be added some remarks on the newer germicides and on the reasons for the occasional ineffectiveness of mercuric chloride. Respecting the latter it is a well-known fact that or- ganic substances of various sorts throw this salt out of solution fixing it in their outer layers. Consequently in many cases its penetrating power is slight, and even when a rather strong solution is used the deeper layers of a substance may not be reached and consequently some bacteria may escape. For this very reason it is well adapted to the surface disinfection of thin tissues such as leaves from the interior of which we wish subsequently to make a culture. The exposure, however, must be very short. It has also been shown in recent years that a killing quantity of mercuric chloride may be rendered ineffective by a subsequent soaking of the treated bacteria in water, i.e., the poison can be washed out of the outer layers of the bacterial membrane and sometimes growth will then take place. These facts must be borne in mind whenever mercuric chloride is used. The writer still uses it as described in Vol. I. It is not likely that it always fully sterilizes the surfaces treated, but long experience has shown that it does do so to an 'In .1 letter dated June 6, tyoS, from Auburn, California, Mr. P. J. O'Gara wrote to me as follows: "The work in California on pear-blight control is showing excellent results wherever there has been united effort in eradication. To show you how hard it is for me to get a culture of the organism, I traveled all one day through the besl pear section of Placer County and only succeeded in getting one case. Another interesting fact is that there are more pear trees growing in this county (Placer) and adjoining counties than there were three years ago. The local nurseries have not been able to supply the demand during the past year or two, all of which speaks very well for our I illice ," GERMICIDES. 193 extent sufficient to exclude rapid-growing surface organisms from participation in the subsequent poured-plate cultures, so that pure growths can be obtained without difficulty from deeper portions of the plants. So far as known to the writer no serious attempts have been made to try the newer germicides on plant parasites. A very few of these may be mentioned. Atoxyl, known chemically as arsenic-acid anilide, or arsanilic acid, has been used during the last few years for injection into the spinal cord in cases of sleeping sickness with a view to destroying the trypanosomes. Statements respecting its value are conflicting with the preponderence of the evidence in favor of the use of the germicide. It is undoubtedly only a makeshift till something better is found. Atoxyl is much less poisonous to the higher animals than white arsenic, but very deadly to some of the lower forms. Some of the more recently introduced related substances such as atoxyl acetyl or arsacetine, and especially arsenophenylglycin, or arsenophenylglycollate of sodium, are still more efficient against trypanosomes and at the same time less poisonous to the higher animals (mice, rats, guinea pigs, rabbits, dogs, horses). This latter salt, which, according to Ehrlich, is a little less toxic than atoxyl (but two to four times less toxic according to Roehl) is said to have a very high therapeutic value. A single dose is said to have cured an animal suffering from an experimental tripanasomiasis. The most-talked-about substance at present is the dioxydiamidoarsenobenzol from Ehrlich's laboratory, commonly called No. 606 and also salvarsan. A single dose of this substance if properly administered is said to be sufficient to destroy every vestige of Tre- ponema pallida in the body of a man. But here again conflicting statements are rife (191 1). Argyrol is an organic silver salt now much used by oculists and others. There are various other similar silver compounds possessing some of the germicidal properties of nitrate of silver without its caustic properties, e. g., protargol, argonin, albargin, largin, saphol (formaldehyd, nucleinic acid and silver). Some of these organic silver compounds are strongly germicidal without being very injurious to the animal body, e. g., argonin, argyrol, albargin. Persistent claims have been made by the manufacturers for the high disinfectant power of cyllin, and some of these claims appear to be borne out by scientific tests. Cyllin appears to be some sort of a phenol or mixture of phenols. It is said to be much less poisonous than carbolic acid, and twenty times as efficient, i.e., nearly or quite as germicidal as mercuric chloride. It is advertised in reputable journals as "non-toxic. " Aniodol, a French disinfectant, has been recommended recently as a substitute for iodoform. It is said to be highly germicidal and may be used as a powder or as a soap. It contains neither mercury nor copper. Cook's asepso soap has been recommended by the Journal of Tropical Medicine. This contains 3 per cent biniodide of mercury. One gram of the soap in 60 cc. of water is said to be equivalent to 1 : 2000 of the biniodide of mercury, and the lather of the soap is supposed to be still more effective. It is said to be a remedy for Favus. For notes on the newer germicides see back volumes of the Journal of Tropical Medicine and Hygiene, Bulletin de l'lnstitute Pasteur, and New and Non Official Remedies, 191 1, Press Am. Med. Asso., Chicago. The writer made several tests of the St. Laceleau soap (see Vol. I, p. 253) using spores of Bacillus sitbtilis. It appeared to be without sensible restraining effect on this organism, i.e., B. sitbtilis grew readily in bouillon after exposure to 10 per cent solutions for 30 minutes, and that, too, without a preliminary washing of the spores. Even when the soap was added to agar in the proportions recommended by Konradi it did not prevent the growth of bacteria. The particular cake tested had been out of the factory three years, but had been recently removed from the box and original tinfoil wrappings and was unchanged in appear- ance. 194 BACTERIA IN RELATION TO PLANT DISEASES. The absolute sterilization of the surface to prevent the plant from contracting disease, to prevent infection of the soil by suspected seeds, cuttings, etc., or finally to prevent the infection of menstrua and the consequent miscarriage of special pathological, physiological, or chemical experiments is sometimes a matter of considerable importance. Unfortunately our knowledge is still very incomplete. Often no specific advice can be given, preliminary experiments being necessary in particular cases. The best that can be done here is to make a few suggestions and record the results obtained in particular instances. The reader is also referred to the special cases mentioned under particular diseases. For holding solutions sterile for short periods without changing their composition the chemists have used chloroform, thymol, toluene and similar substances. To get the full restraining effect of these substances the material to which they have been added should be shaken continuously and shut away from the air; otherwise bacterial growths are likely to appear (see Vol I, p. 74). Glycerin is not germicidal to Bacillus subtilis see p. 149. Con- tinuous exposure at i°C. or at 8o° C. will often answer when the nature of the reaction desired, or of the substance to be isolated, will permit it. The growth of aerobes can be held in check by substituting for air some inert gas such as hydrogen or nitrogen. Surfaces can not be washed free from bacteria. Inasmuch as pathogenic bacteria are often transmitted from diseased plants to healthy plants by insects, the wholesale destruction of the latter often becomes imperative, and some comments are, therefore, added on the most efficient insecticides. In 1908, H. Chick and C. J. Martin published a paper on the standardization of dis- infectants in which they record the following results. It is generally recognized that in any method of standardization the temperature, and the com- position of the culture medium should be constant, while the numbet of bacteria per unit volume and the resistance of the test organisms used should be as constant as possible. The authors adopted 20° C. as the temperature most closely approximating the conditions of practical disinfection. Thirty minutes was found the most satisfactory time unit of exposure, as a shorter time was unfavorable to mercury and silver salts, while for phenol and emulsified disinfectants either 10 minutes or 30 minutes was satisfactory. It was found best to employ a sulphide to neutralize the traces of disinfectant carried over with the test sample. Mercuric chloride required an excess of sulphide to decompose a compound formed between the mercuric salt and the substance of the bacteria which prevented growth. The most satisfactory results were obtained with 0.2 cc. of saturated solution of hydrogen sulphide in distilled water to 10 cc. of broth. The efficiency of a disinfectant was found to vary with the organism used. In the case of spores, metallic salts were the most effective germicides. These were effective in very small concentration (1000 times less than phenol). With phenol, sporing forms were from 17 to 25 times more resistant than vegetative forms. Virulent strains were generally more resistant than non-virulent ones. Since in practice disinfectants are commonly used in the presence of organic matter it seemed desirable to introduce this factor into the Drocess of standardization. Experiments with this in view showed that 10 per cent blood serum reduces the efficiency of phenol about 12 per cent. A somewhat greater reduction occurs with emulsified disinfectants and a much greater with mercuric chloride. A solution containing 0.5 per cent of the latter was reduced from 0.6 to 0.06 of its original value as the concentration of serum was increased from 5 to 30 per cent. The presence of particulate organic matter (dust, animal charcoal, finely pulverized coagulated albumen, bacteria and faces) affects the germicidal value of emulsified disinfectants far more than that of phenol. Commercial cresols were reduced in efficiency 30 to 50 per cent by the introduction of such matter. Finer emul- sions are more seriously reduced than coarse ones. This reduction was shown to be principally due to adsorption of the emulsion upon the surfaces of the particles. In 1908, Dr. Rideal discussed the question of uniform methods of testing disinfectants. He states what every one knows that it is very difficult and in many cases impossible to compare the results of different workers because no definite standards have been adopted. He cites tests of Koch, Esmarch, Fraenkel, Geppert, Klein, Cash, Wynter Blyth, and Sternberg as examples. GERMICIDES. 195 Some of Rideal's comments (Journal of Tropical Medicine) are as follows: In 1897 Kronig & Paul recommended substituting small garnets of uniform size for Koch's thread method, since the garnets could be washed and treated with chemical reagents to neutralize the disinfectants. They tested the results of disinfectants quantitatively by making poured plates. Madsen & Nyman (Zeits. f. Hygiene, Bd. 57, p. 388) and Miss Chick (Journal of Hygiene, February, 1908) have shown that their results can be plotted in curves. These show that the velocity of disinfectants depends primarily on number of bacteria to be destroyed, and secondarily on their resistancy or age. When anthrax spores are used they may be regarded as of practically the same age and resistance. With ordinary cultures of non-spore-bearing organisms, he states there are individuals of different age having different resistancy, and then the velocity of the disinfectant is not so simple. He says that Miss Chick has shown that by taking cultures of short periods of 3 hours, and sub-culturing, the organisms present in such cultures become uniformly more resistant, and the velocity of dis- infection approximates to the same law governing disinfection of spores. Miss Chick has also found that temperature influences the process of disinfection, disinfection being more rapid in warm climates than in cold ones. Dr. Rideal states that Miss Chick's work has confirmed the method of disinfection recommended by Walker & Rideal in 1903. He says the elaborate method of the use of garnets "may be replaced advantageously by the drop method known under our names." They used carbolic acid as the basis of their tests. Next after the question of standardization of the disinfectants, he considers the culture broth of primary importance. The test organisms should be grown for 24 hours and then inoculated into primary test cultures always made up from nutrient broth of a definite composition. Rideal and Walker have suggested that the broth should contain 20 grams of lemco, 20 grams of Witte's peptone, and 10 grams of salt per liter of water, and should after boiling be neutralized with caustic soda, after which 15 cc. of normal hydrochloric acid are added per liter. "Having secured a standard broth, standard carbolic acid, and a standard culture, the only other conditions are those of temperature and sterility." Rideal says that these and other details are given in a little book by W. Partridge on the " Bac- teriological Examination of Disinfectants," published in 1907 by the Sanitary Publishing Co. He says further: " In view of the very great development that has taken place in our knowledge of the constitution of the derivatives of phenol, various refined and highly scientific products have been put upon the market by enterprising manufacturers, and it is, therefore, desirable that the medical man may have a weapon bv which these products may be identified and classified according to their germicidal efficiency, not only because these newer products are so valuable in preventing and eradicating disease, but in the comparatively new field of medical application for internal treatment, where, of course, the efficiency of the dose is of supreme importance. "Thus, for example, I understand that Dr. Wright and Dr. Morgan in working on cancer are using this test for determining the germicidal value of einnamic acid derivatives for internal use whilst Dr. Hartigan, in your own Journal in 1905, pointed out that a well known disinfectant with a high coefficient [cyllin, probably] could be used in sprue as an intestinal disinfectant when admin- istered in the form of palatinoids, and Fleet Surgeon MacNab has similarly found that it can also be used internally in treatment of Mediterranean fever. Captain Brodribb has also used the same disinfectant in cantonments in India as a douche for the treatment of gonorrhoea in women. " In a discussion of this paper Dr. Sommerville referred to the necessity of introducing organic matter into the cultures so that the laboratory tests are in a measure conformed to the actual con- ditions occurring in practice. He says: "The question of prime importance is the type and quantity of organic matter which should be introduced. From work executed a year and a half ago, it was found that less than 10 per cent of organic matter brought down the coefficients of all disinfectants to the same figure. Mr. Ainslie Walker and I have recently adopted a 1 per cent mixture of starch and gelatin." Professor Hewlett said: "I think that all those who have had anything to do with this Rideal-Walker test must agree with the beautiful simplicity of the method, and I think the greatest credit and congratulations are due to Dr. Rideal and Mr. Ainslie Walker for working out the test. It is really the first practical method that has been devised for comparing the germicidal efficiency of disinfectants. * * "Lastly there is the point which Dr. Scmmerville has raised; the test, though so beautifully efficient for so large a number of disinfectants, lacks, of course, in one point, namely, that one is acting upon naked germs, whereas in actual practice the germs are mixed with organic matter. There is a certain lowering of the efficiency of the disinfectant in the presence of organic matter, so 196 bacteria in relation to plant diseases. that we want to increase the efficiency of the test, if we can, by the addition of some form of organic matter, which will aid us in determining the real efficiency of disinfectants in the presence of organic matter. This point is especially important in connection with disinfectants which act by oxidation. If you try permanganate of potash on naked organisms you will find that it is very efficient, but if you mix with it a little dust you will find that its power has largely gone down." Dr. Schryver in the discussion also cited the physical action of organic matter held in suspen- sion as well as in solution and considered this to be important. GERMICIDAL TREATMENT OF SEEDS. Seeds with a thick impenetrable seed-coat offer no special obstacles to thorough dis- infection. All that is necessary is actually to wet every part with a strong germicide for a sufficient length of time. To insure this wetting there should be a brief preliminary wetting in alcohol that air may be driven out of all the minute crevices where otherwise the germi- cide would not penetrate. The case is quite different, however, with seeds having delicate and easily permeable seed-coats. These have to be treated with great care, and often it is not possible to dis- infect their surface thoroughly without at the same time destroying the embryo, at least in a large proportion of the seeds. Sometimes in such cases we may reach the end desired indirectly, i.e., by not allowing the seeds to become infected, since they are always originally sterile inside the unopened pods. To obtain sterile seeds it is suggested that the unopened seed-pods be collected with great care and their surface treated with germicides or fire, or both, after which the pods must be carefully opened (in still air), and the seeds removed by means of sterile forceps (see fig. 2, and p. 135). The writer found that 15 minutes' exposure of hard dry kernels of sweet corn to 1 :iooo mercuric chloride did not entirely sterilize the surface, although from the results obtained it must have come very near to doing it, so far as regards the death of the organism in question, i. e., Bacterium stewarti. Experiments made by the writer in the summer of 1909 with hybrid dent corns and sweet corns having a high germinating capacity showed that the dry kernels would stand exposure to 1 :iooo mercuric chloride water for 20, 30, 40, and 50 minutes with little injury, the kernels being placed at once in damp sand after preliminary rinsing in hydrant water or without rinsing. Nearly all germinated promptly and the seedlings looked as well as those from the untreated seeds. In the first series twenty seeds were planted in each pot and the following are the number of germinations per pot, the count being made on the sixth day: (1) U. S. P. B. No. 100 (field-corn): Checks, 20, 20, 20, 19, 19; Checks, — none. Mercuric chloride (20 minutes), 18, 20, 17, 19, 19; Mercuric chloride (30 minutes), iS, 19, 18, 19, 20. In a second series of tests, using 20 seeds and counting on the fifth day, the follow- ing results were obtained : (2) U. S. P. B. No. 120 (field-corn): Checks, 14, 18, 14, 17, 16; Checks, 16, 15, 17, 16, 18; Mercuric chloride (40 minutes), 18, 19, 15, 19, 12; Mercuric chloride (50 minutes), 18, 17, 16, 18, 20. There was no marked difference in the appearance of the seedlings. The experiments were repeated a few days later with this difference only, that the seeds were not rinsed, but dried promptly and planted at once with the mercuric chloride adhering to them. The results on the sixth day were as follows: (1) U. S. P. B. No. 100 (field-corn): Checks, 18, 20, 20, 19, 20; Clucks. [8, 19, 20, 19, 20; Mercuric chloride (20 minutes), 19, 18, 16. 19, 17; Mercuric chloride (30 minutes) 18, 20, 19, 20, 19. The treated seeds showed a slight retardation in germination. Checks 2 to 3 inches high ; treated 1.5 to 2 inches high. MERCURIC CHLORIDE ON MAIZE. 197 The experiments were then continued as follows: (2) U. S. P. B. No. 120 (field corn): Checks, 20 19, 20, 20, 20; Checks, 20, 18, 19, 18, 20; Mercuric chloride (40 minutes), 18, 18, 13, 18, 15; Mercuric chloride (50 minutes), 16, 18, 16, 18, 15. The most injury (retardation) was in seeds exposed 40 and 50 minutes. A few were just coming up on the sixth day. (3) Yellow Dent. — Examined at the end of seven days: Checks, 20, 20, 20, 20, 20, Checks, 20, 20, 20, 20, 20; Mercuric chloride (20 minutes), 20, 20, 20, 20, 20. Mercuric chloride (40 minutes), 19, 20, 20, 20, 20. Checks, 20, 20, 20, 20, 20; Checks, 20, 20, 20, 20, 20; Mercuric chloride (30 minutes) 20, 20, 20, 20, 20. Mercuric chloride (50 minutes), 20, 20, 20, 20, 20. Considerable retardation at first, especially in longer exposures. Scarcely any in the 20 minute exposure at end of experiment, and all making a good growth. (4) White Flint. — Examined at the end of eight days: Checks, 20, 20, 20, 20, 20; Checks, 19, 20, 20, 20, 20; Mercuric chloride (20 minutes), 20, 20, 20, 20, 20. Mercuric chloride (40 minutes), 20, 20, 20, 20, 19. Checks, 20, 20, 20, 19, 20; Checks, 19, 20, 20, 20, 19; Mercuric chloride (30 minutes), 19, 20, 19, 20, 20. Mercuric chloride, (50 minutes) 19, 20, 19, 20, 20, Considerable retardation at first, especially in longer exposures. Scarcely any in the 20 minute exposure at end of experiment, and all making a good growth, even the plants from seed exposed for 50 minutes. (5) Black Mexican, examined at the end of 7 days: Checks, 19, 17, 19, 18, 18. Checks, 20, 18, 15, 20, 16. Mercuric chloride (20 minutes), 16, 18, 18, 19, 17. Mercuric chloride (40 minutes), 19, 19, 18, 18. Checks, 17, 17, 19, 20, 19, Checks, 16, 18, 18, 15, 19. Mercuric chloride (30 minutes), 19, 16, 19, 19, 17. Mercuric chloride (50 minutes), 19, 18, 17, 18, 18. (6) Early Evergreen, examined at the end of 8 days: Checks, 18, 18, 19, 19, 20. Checks, 19, 19, 18, 20, 20, Mercuric chloride (20 minutes), 18, 19, 18, 20, 19. Mercuric chloride (40 minutes), 18, 19, 19, 18, 17. Checks, 17, 20, 20, 18, 20. Checks, 20, 19, 18, 19, 18. Mercuric chloride (30 minutes), 18, 19, 17, 19, 19. Mercuric chloride (50 minutes), 20, 19, 20, 16, 18. (7) Country Gentleman, examined at the end of 8 days: Checks, 19, 19, 18, 19, 18. Checks, 20, 18, 20, 19, 19. Mercuric chloride (20 minutes), 19, 20, 19, 19, 19. Mercuric chloride (40 minutes), 20, 20, 20, 18, 20. Checks, 16, 18, 19, 18, 20. Checks, 19, 20, 18, 19, 20. Mercuric chloride (30 minutes), 20, 20, 19, 19, 18. Mercuric chloride (50 minutes), 20, 18, 19, 17, 20. The retardation was about the same in 5, 6, and 7. It was most in those exposed for the longest periods. There was no serious injury in any. All made good plants. (8) Old Colony, examined at the end of 7 days: Checks, 19, 19, 18, 19, 20. Checks, 18, 19, 19, 19, 19. Mercuric chloride (20 minutes), 20, 19, 19, 20, 19. Mercuric chloride (40 minutes), 18, 20, 19, 18, 19. Checks, 20, 18, 20, 17, 19. Checks, 17, 17, 17, 20, 19. Mercuric chloride (30 minutes), 17, 19, 17, 17, 20. Mercuric chloride (50 minutes), 18, 16, iS, 17, 19. (9) Country Gentleman (another source), examined at the end of 7 days: Checks, 19, 19, 18, 19, 19. Checks, 20, 18, 19, 19, 17. Mercuric chloride (20 minutes), 20, 19, 19, 18, 20. Mercuric chloride (40 minutes), 16, 16, 18, 20, 18. Checks, 18, 20, 19, 19, 20. Checks, 20, 20, 18, 19, 19. Mercuric chloride (30 minutes), 19, 18, 20, 19, 19. Mercuric chloride (50 minutes), 19, 18, 15, 20, 18. Plain retardation in treated seeds of 8 and 9, expecially the longer exposures, but all growing well when removed. (10) Golden Bantam, examined at the end of 7 days: Checks, 15, 15, 13, 13, 16. Checks, 17, 16, 15, 8, 14. Mercuric chloride, (20 minutes) 12, 16, 16, 14, 16. Mercuric chloride (40 minutes), 18, 18, 16, 16, 16. Checks, 16, 13, 14, 17, 14. Checks, 14, 14, 12, 17, 17. Mercuric chloride (30 minutes), 19, 16, 16, 1.5, 18. Mercuric chloride (50 minutes), 10, 17, 14, 15, 17. Very little retardation of this variety in 20, 30, or 40 minutes' exposure. In all cases 100 cc. of the 1 :iooo mercuric chloride water was used for each 100 seeds. 198 BACTERIA IN RELATION TO PLANT DISEASES. Many experiments have been made with grains to free them from smut fungi and in this way considerable knowledge has been gained respecting the toleration of seed wheat, oats, etc., for hot water, copper salts and various other disinfectants. Some of the leading papers are mentioned under Literature. Only a few of the results will be cited here. Jensen who discovered the hot-water treatment for stinking smut (1888) advised temperatures between 1270 and 1330 F., and exposures of seed wheat and oats for not over 5 minutes. He did not, however, determine accurately the thing we are here specially inter- ested in bringing out, namely, the killing temperature for the grains. In 1890, Arthur determined the effect of hot water on the germination of wheat. Wheat seeds immersed 5 minutes in water at 1350 F. (570 C.) are not injured. Six hundred seeds exposed to 1300 F. (540 C.) for 10 minutes also gave excellent results on germination — 12.5 per cent in 24 hours and 93 per cent in 5 days. The injury to those treated 10 minutes at 1350 F., and 5 minutes at 1400 F. (6o° C.) equaled about 20 per cent. The limit of germ- ination is 1500 F. for 5 minutes (t,^ P^r cent). No germinations were obtained when wheat seeds were exposed to higher temperatures, e. g., 1550 F. for 5 minutes, or to 1500 F. for 10 minutes. In 1 89 1 Arthur tested the effect of hot water on oats, with especial reference to the prevention of loose smut. He states that the hot-water treatment — 10 minutes in water at 1350 F., or 5 minutes in water at 1350 F. to 1400 F. (570 C. to 6o° C.) entirely destroys the smut while at the same time it improves the growth and increases the yield of oats. The water may be even as hot as i45°F. when the oatsare first put into it without much injuring the germination. Arthur made the exposures in cheese-cloth bags. Latta found 5 minutes' exposure of oats in copper sulphate water (1 pound to 1 gallon) destroyed the smut but the germination was slower and the yield per acre was reduced. The comparative yields were: Hot water, t,^ bushels; untreated, 28 bushels; coppered, 24 bushels. Arthur, who reports this, tested the effect of copper sulphate on germination on lots of 200 seeds and obtained in the germinating chamber the following per cents : Hot water, 99; untreated, 98; copper sulphate, 67. Even in the soil where 98 per cent of the oats treated with copper sulphate finally germinated, they did so very slowly, the primary roots were often killed, and often they pushed out the plumule in advance of the roots. Kellerman & Swingle (1890) found that exposure of wheat at 1390 to 1400 F. for 15 minutes destroyed nearly all of the kernels, i.e., on a plot that should have yielded 3,000 or more heads there were only 9. Copper sulphate 8 per cent, 24 hours, limed orunlimed, reduced the germinations about one-fourth. Copper sulphate 5 per cent 24 hours, unlimed, reduced the yield nearly one-third. Bordeaux mixture reduced the yield over one-fourth. Eau celeste, 24 hours (on another page the time is said to have been 36 hours) destroyed all. Carbolic acid 5 and 10 per cent for 20 hours destroyed all. Mercuric chloride 1 per cent for 20 hours destroyed all. Potassium bichromate 5 per cent for 20 hours destroyed about half. According to Kellerman & Swingle (1891) oats which were treated at 141. 8° F. (6i° C.) for 5 minutes gave a good crop. The same result was obtained by exposing at 138.20 F. (59° C.) for 10 and for 15 minutes, i.e., there was no destruction of the seed. Potassium sulphide 0.75 per cent and 0.5 per cent for 24 hours reduced the number of stalks about one- fourth. Copper sulphate 0.1 per cent for 24 hours reduced the number of heads about one- fourth. Copper sulphate 0.5 per cent for 24 hours reduced the number of heads nearly half. Copper nitrate in 5 per cent solution for 24 hours, limed or unlimed, destroyed most of the seeds. Even 2.5 per cent or 1 per cent greatly reduced the crop. Corrosive sublimate 0.1 per cent for 24 hours reduced the yield three-fourths. Potassium bichromate 10 per cent for 23 hours killed all; same, 1 per cent for 9 hours, reduced the crop one-half or more. They recommend treating oats for smut by (1) hot water: temp. 132. 50 F., time 15 minutes; or, (2) potassium sulphide: 1 pound to 20 gallons of water, time 24 hours. INFLUENCE OF GERMICIDES ON GERMINATION. 199 In 1 89 1, Kellerman St., recommended hot water over all other fungicides for great efficiency without injury to seeds. His experiments were with wheat kernels for the pre- vention of stinking smut. As a result of many experiments (about 70) he recommends exposure for 15 minutes to a temperature of 1310 F. (550 C). Hot water for 5 minutes at 1370 F. and at 13S0 F. (seeds previously soaked 10 hours) destroyed all the grains. The following treatments greatly injured or nearly or quite destroyed the grains: Bordeaux, 24 hours; same, half strength; 1 per cent copper sulphate, 24 hours; 1 per cent copper acetate, 24 hours; 1 per cent copper chloride, 24 hours; 1 per cent potassium bichromate, 24 hours. The following treatments gave reasonably good yields, i.e., better than the checks, but not as good as the hot water: Copper sulphate 0.5 per cent, 24 hours, limed; copper sulphate 0.5 per cent, 12 hours, limed; copper acetate 0.5 per cent, 24 hours; copper nitrate 1.0 per cent, 24 hours; copper nitrate 0.5 per cent, 24 hours; mercuric chloride 0.05 per cent, 24 hours. The following gave a yield nearly equal to the checks: Eau celeste, 24 hours; mercuric chloride, 0.1 per cent, 24 hours. Ratio of grain to volume of fluid not given. Hot water at 1360 F. ( 5 7 . 7 ° C.) for 5 minutes, then quickly cooled, appears to be the severest exposure compatible with a good crop. In 1893, Hitchcock and Carleton published the results of their experiments with maize. They tested the effect on germination of 82 chemicals in various strengths, making a total of 400 experiments. They obtained in moist sand a germination of 80 to 100 per cent (retarded) after exposure to the following strengths of mercuric chloride water: 0.1 per cent for 1, 3, 5, 8, hours; 1.0 per cent for 1 hour. One per cent mercuric chloride for 24 hours or 3 per cent for 1 hour killed all. Chromic acid 1 per cent for 48 hours gave 75 per cent retarded germi- nations. Copper chloride 10 per cent for 24 hours gave 100 per cent retarded germinations. Copper nitrate 10 per cent for 24 hours gave about 80 per cent retarded germinations. Potassium permanganate 2.5 per cent for 24 hours gave 100 per cent germinations. Hy- posulphite of soda 10 per cent for 24 hours gave full germinations. From 80 to 100 per cent of retarded germinations were obtained after exposure to potassium cyanide as follows: 1 per cent, 1 hour; 5 per cent, 1 and 3 hours; 10 per cent, 1 hour. The same, 0.5 per cent for i hour scarcely affected germination. In 1897, Bolley published studies on the fungicidal treatment of wheat, oats and barley which he had carried on for a period of 5 years. The following are some of his con- clusions respecting resistance of the dry grain. He states that if the wheat grain is dried at once germination is not retarded by applying corrosive sublimate solutions in strengths up to 4 parts in 1,000 parts of water: Of selected seed of Scotch Fife wheat exposed 2 minutes, 95 per cent germinated; exposed 3 minutes, 82 per cent germinated; 4 minutes, 72 per cent; 5 minutes, 78 per cent; 6 minutes, 67 per cent; 7 minutes, 45 per cent; 20 minutes, 17 per cent; 25 minutes, o. As little as 0.1 per cent corrosive sublimate weakened the first growth in a rapidly increasing degree in exposures longer than 3 minutes, but even from too strong treatments the final after-growth is stronger than from untreated grain. Mixed samples of oats treated with 0.3 per cent corrosive sublimate water for 30 minutes gave good first growth (94 to 100 per cent germinations) and a good yield per acre. Barley after 15 minutes exposure to 0.3 per cent mercuric chloride gave 94 per cent germinations. Seed wheat treated 10 minutes or less with 1 to 2 per cent solution of formalin gave a normal number of germinations or better, but soaking over 10 minutes decreased slightly the per cent of germinating seeds. Exposure for 10 minutes to 10 per cent killed all, and merely dipping into 5 per cent reduced the germinations to 34 per cent. Subsequent experi- ments showed that wheat or oats would germinate perfectly after soaking in 0.4 per cent formalin 1 to 3 hours. Seed wheat will stand an exposure of 1 minute at 1500 F. (65. 50 C.) and give 80 to 90 per cent of germinations. Oats exposed to hot water at 1400 to 1430 F. for 5 minutes gave 200 BACTERIA IN RELATION TO PLANT DISEASES. 98 per cent germinations and exposure for 5 minutes in water at 1400 F. (6o° C.) or below may be recommended as not injurious to wheat. Barley dipped for 30 minutes in copper sulphate water (1 pound to 4 gallons) gave 86 per cent of weak germinations; and when exposed for 1 hour, 47 per cent. Wheat exposed to potassium sulphide (1 ounce to 1 gallon for 75 minutes gave 100 per cent germinations. Barley treated in the same way for 75 minutes gave 90 per cent germi- nations. Oats treated in same way gave 96 per cent germinations; but exposed for 19 hours gave 42 per cent weak germinations. According to Cranefield (1901), formalin used as weak as 2.5 : 1,000 (1 pound [pint] to 50 gallons of water) for 20 minutes may injure oats used for seed. The experiments cover 20 varieties of oats and the germination of over 25,000 seeds. The amount of injury varied greatly in different varieties, and was more noticeable in planted seeds than in those used in the germinating chamber. Longer exposures than 20 minutes at the standard strength (1 pint to 50 gallons) did not much increase the injury. The early growth from the treated seed was retarded and at no time did the treated quite equal the untreated in height. When more concentrated solutions of formalin were used the injury was progressively greater, c. g., 1 pint to 50 gallons of water, 91 per cent germination (check 94. 5) ; 1 pint to 25 gallons of water, 74 per cent; 1 pint to 20 gallons, 73 per cent; 1 pint to 10 gallons, 31 per cent; 1 pint to 5 gallons, 12 per cent. In 1 901, Windisch published many experiments on lupins, peas, horse beans, soy beans, corn, flax, rape, lucern, and clover, showing the effect of formaldehyde on germina- tion. Each of these was in duplicate. The following are some of his conclusions : Per cent of Germinations. *°d°f Distilled each water- White lupins 50 Victoria peas 50 Horse beans 50 Soy beans [ 00 Flax 200 Maize 100 Summer rape 200 Lucerne 200 Clover 200 100 84 ioo 99 97-75 IOO 98.25 90.50 05.00 Formaldehyd (Per cent in Water.) 80 97 98 94-25 IOO 80.25 88.75 89.50 99 56 100 97 11.75 IOO 4 27 34 89 19 98 92 995 2 7-75 7-5 4 6 94 40 0.40 4 12 26 6 94 7-5 4-5 No injurious action was observed on lupins, peas, horse beans, soy beans or maize, when the 0.02 per cent formaldehyde solution was used. Hiltner, in some root-nodule experiments, exposed soy bean seeds for 3 minutes and for 10 minutes to 1 :ioo mercuric chloride water, then carefully washed it away and planted. The plants came up badly, but this was not ascribed to the germicide. According to Dr. Windisch, winter wheat endured a soaking for 24 hours in 0.02 per cent formaldehyde, and in 0.04 per cent without lessening the power of germination. It also endured 0.08 per cent formaldehyde for 24 hours and gave a germination of 88.5 per cent at the end of 14 days. Exposure to 0.12, however, gave only 9.25 per cent germination at the end of 14 days, and exposure to 2 per cent gave o per cent germination at the end of 14 days. Even the diluted solutions delayed the germination somewhat. F. L. .Stevens (1909) reports that treatment of oats with a solution of 1 ounce formalin to 0.5 gallon of water reduced germination to 37 per cent, while a solution of 1 ounce to 1 gallon of water for 24 hours gave a germination of 73 per cent to 96 per cent, according to EFFECT OF GERMICIDES ON PLANTS. 201 the varieties used. He recommends for practical purposes a solution of i ounce to i gallon for 2 hours, followed by 10 hours treatment with lime. GERMICIDAL TREATMENT OF DORMANT PLANTS. Plants in the resting condition, especially roots and shoots protected by cork, will bear relatively strong doses of germicides. Bordeaux mixture (6:4:50), copper sulphate solution (3:50), mercuric chloride water (1:1000), soap and lye solutions, lime washes, cold boiled or hot boiled lime sulphur solution, lime-salt-sulphur wash may be applied rather freely. A few formulae are given at the end. A more difficult and important problem concerns the use of germicides and insecticides on growing plants. GERMICIDAL TREATMENT OF GROWING PLANTS. In the treatment of growing plants two things must be kept in mind constantly : (1) The foliage must not be injured; (2) the applications must be effective. A third very desirable quality in a germicide is adhesiveness, since if the substance is washed off by every rain the necessary reapplications will be expensive. Bordeaux mixture containing an excess of lime, e.g., formula 4:6:50, or 4:4:50 is borne very well by some plants. The foliage of others is liable to be burned, especially if the spray- ing is not repeated frequently either with Bordeaux or with milk of lime so as to keep an excess of lime present on the leaves. This mixture is an effective fungicide and also has some value as a germicide. Pierce used it on walnut blight in California with partial success. It might perhaps be used to protect from some of the leaf spots. Always, however, it is advisable to try the experiment on a small scale first, until the general effect of the copper on the foliage has been determined. The writer has seen a whole peach orchard defoliated in midsummer by the improper use of Bordeaux mixture. Copper absorbed in minute quantites, has a stimulating effect on growth. Chester observed this in 1890 while testing the effect of copper salts on Vitis. He says that Bordeaux mixture seems to stimulate the growth of the vines. In 1894, Frank & Kriiger stated that the assimilation of potato leaves is increased, the transpiration becomes greater, the leaves live longer, the harvest is increased, and the tubers contain more starch when the plant has been treated with copper salts, especially "the ordinary 2 per cent copper vitriol-lime mixture." In 1895 Galloway and Woods showed that Bordeaux mixture could be used with safety on growing grape-vines and potatoes, and observed that copper salts stimulated the growth of these plants. In 1898, Harrison stated that Bordeaux mixture has an invigorating effect on the foliage of plum, pear, peach, and quince. In 1898, Starnes in Georgia reported injury to peach foliage from copper salts sprayed thereon. In 1899, Duggar obtained shot-hole effects on peach foliage as the result of the use of copper fungicides. In 1900, Pierce published his observations on the physiological stimulation of Bordeaux mixture on peach leaves. As a result of his researches, published in 1902, Bain concludes that peach leaves are especially sensitive to poisons in general and to copper in particular. The self-boiled lime-sulphur mixture is less injurious to the leaves of peach and plum trees than Bordeaux mixture, and appears to be an equally good germicide. Scott has used it on peaches for the prevention of the leaf-spot due to Bad. pruni, and with brilliant suc- cess in the summers of 1909 and 19 10 for prevention of the brown rot due to Monilia. It should be tried also for the prevention of the walnut blight due to Bad. juglandis. 202 BACTERIA IN RELATION TO PLANT DISEASES. The stage at which cold water should be poured on to stop the cooking varies with different limes. Some limes are so sluggish in slaking that it is difficult to obtain enough heat from them to cook the mixture at all ; while other limes become intensely hot on slaking and care must be taken not to allow the boiling to proceed too far. If the mixture is allowed to remain hot fifteen or twenty minutes after the slaking is completed, the sulphur gradually goes into solution, combining with the lime to form sulphides which are injurious to peach foliage. It is therefore very impor- tant, especially with hot lime, to cool the mixture quickly by adding a few buckets of water as soon as the lumps of lime have slaked down. The intense heat, violent boiling and constant stirring result in a uniform mixture of finely divided sulphur and lime with only a very small per cent of the sulphur in solution. The mixture should be strained to take out the coarse particles of lime, but the sulphur should be carefully worked through the strainer. (From Scott and Ayres' Bulletin on The Control of Peach Brown-Rot and Scab.) Alsberg and Hasselbring in U. S. Department of Agriculture in the summer of 1909 subjected cabbage leaves to 1 :2oo mercuric chloride water for 30 minutes without entirely sterilizing their surfaces, i.e., a white schizomycete subsequently appeared in the flasks con- taining the leaves which were to have been examined chemically. In this instance, however, the leaves were rather large and were not previously soaked in alcohol. INSECTICIDES. Carbon bisulphide is an excellent insecticide for certain purposes. Its vapor is inflam- mable and care should be exercised in its use. It must not be used near an open flame. In 1897, Hicks and Dabney showed that there was no appreciable loss of germinating power in wheat, corn, barley, or rye, from treating the seed in bulk with carbon bisulphide for 24 hours at the rate of one pound of the chemical to 100 bushels of grain. In recent years it has come to be recognized that carbon bisulphide is better adapted to kill certain insects, e.g., weevils in grain, phylloxera in the soil, etc., than hydrocyanic acid gas, because it has greater penetrating power. One teaspoonful per cubic foot is the usual amount allowed in making small treatments (Jno. B. Smith). For aphides on plants in the open air, kerosene emulsion is useful. To make it one must have a force pump with good churning power. When properly made it may be kept for some weeks and diluted with water as needed for spraying. For aphides in houses tobacco smoke properly applied is very effective and not injurious to the plants. Improperly applied, it may burn the foliage seriously. The houses should be well wet down in advance, and then a prepared tobacco paper burned until there is a dense smudge, or else the house filled with the steam from strong tobacco water. This latter may be obtained by distributing shallow pans of the concentrated fluid at frequent intervals and dropping large red hot spikes into the liquid from a wire crate which has been heated in the engine-room furnace. It may also be evaporated from pans placed over oil stoves. This concentrated tobacco extract may be had on the market under the name of Nicofume. Aphides and most other pests in houses may be destroyed by hydrocyanic acid gas. This treatment is inexpensive and very effective. Only red spiders do not seem to be much harmed by it at least in such doses as can be used on plants. Plants are also sensitive to this poison but to a less degree than most animals. Different varieties of plants also vary considerably in sensitiveness. The tomato and olive are quite sensitive. The aim of the grower should be to generate just enough of the gas per cubic foot to destroy the insects without injuring the plants. If the grower has no knowledge of the amount of gas which his crop will tolerate, then he must determine this on a small scale before applying the remedy to a whole house, otherwise disastrous results are likely to follow. Eggs of insects, e. g., those of the white fly (.4 leyrod es) , are more resistant to this gas than the mature forms, or the plants infested, and therefore small doses of the gas at INSECTICIDES. 203 frequent intervals as the eggs hatch are necessary to control certain pests, ;'. c, as often as twice a week for a month, if the houses are badly infested. The gas is best generated in stone jars which should be distributed at equal distances through the house, and not set too close to the plants lest the near ones should receive an overdose of the gas and be scorched. The jars are dosed with a measured amount of crude sulphuric acid and water (1 to 2), and into these are dropped weighed amounts of cyanide of potash wrapped in thin brown paper so as to delay the evolution of the gas for a minute, and thus allow the operator to escape. To avoid the boiling over of the acid during the violent evolution of the gas, the jars should be deep rather than shallow. The house should be shut tight and arrangements made in advance to open it from the outside when the exposure is completed. The cyanide of potash may also be lowered into the jars from the outside by means of strings; this is a rather safer way since the generated gas diffuses through the air with great rapidity, i.e., nearly as fast as a man can run. Sunset of a still day is the best time for commencing the exposure. The house should be opened up after 1, 2, or 3 hours. The air space of each house must be calculated very carefully and for growing plants not more than 0.15 gram of the cyanide of potash should be used for each cubic foot and half this quantity for sensitive varieties. Any serious error in the calculation means, of course, either failure of the treatment or destruction of the crop. The gas is deadly to man and the higher animals, and exposures must not be made in hothouses connected with stables or living rooms; and if dwelling houses are near, the doors and windows on that side must be closed, or the rooms vacated. The potash salt is also very poisonous and must be kept out of the reach of children and animals and handled with rubber gloves. A good remedy for red spider is a desideratum. Repeated syringing with water is recommended. They are usually worse in dry seasons. For the destruction of larva, beetles, and bugs out of doors, a spray containing arsenate of lead is effective, and foliage usually bears this poison very well, i.e., much better than Paris green. Popenoe used 6 pounds to 50 gallons of water on potato foliage to destroy the Coloradopotatobeetle. All the larvaewere killed in 48 hours and the plantswere notinjured. Paris green is also an effective insecticide. It may be sprayed on the foliage, which is sometimes burned ; or may be dusted on mixed with flour, air slaked lime, or land plaster (1 part to 30 or 50). Both lead arsenate and Paris green may be combined with Bordeaux mixture, so as to avoid two separate sprayings. For suggestions respecting trap crops see pp. 282, 295, 296. Many of our experiment stations now publish annual spraying calendars and other literature, giving the principal formulae, and usually these publications may be had upon application. 204 BACTERIA IN RELATION TO PLANT DISEASES. FORMUL/E. Copper Sulphate. Dissolve in hot water, or by suspending the crystals in a sack in the top of the cold water. This is best done over night. It is convenient to make a strong solution (i or 2 pounds per gallon) and dilute as needed. Ammoniacal Solution of Copper Carbonate. Use 5 ounces of copper carbonate, 3 pints or less of strong ammonia water (260 B.), i. e., just enough to bring the copper carbonate into solution, and 50 gallons of water. The copper carbonate must be wet with water first and then stirred into the ammonia after the latter has been diluted with 5 or 6 volumes of water. Add always a slight excess of the copper carbonate and use only the supernatant clear liquid. Used on foliage and fruit when Bordeaux mixture would be unsightly. Apply frequently in case of rainy weather. Bordeaux Mixture. There are various formula1 in which the first figure represents pounds of copper sulphate, the second figure pounds of stone lime and the third gallons of water. The 6 : 4 : 50 is the usual combination and sticks better than the 4:6: 50. It also sprays easier. For sensitive plants 4 : 4 : 50 may be used or 4 : 5 : 50. The lime must be of good quality and fresh slaked. Air slaked lime must not be used, neither should concentrated solutions be mixed, nor hot solutions. Divide the 50 gallons of water into two equal parts. Dissolve the cop- per sulphate in one part. This is usually done over night. Slake the lime with a portion of the other part, adding the water slowly, then add the remainder of the water when the lime has ceased to be lumpy. When ready to spray, stir thoroughly to obtain an even mixture of the lime and water and pour the two fluids together through a strainer tied across the top of a clean barrel. The two streams should blend as they fall to insure a good product, the essential features of which are alkalinity and a fine grain insuring suspension in the fluid long enough to permit of the spraying, which should be undertaken at once. Concentrated solutions give a coarse precipitate which settles quickly. The mixture should be absolutely free from sawdust, sticks, straws, chaff, wool, fragments of leaves or any similar substances. ( (therwise, vexatious delays are likely to arise from clogging of the nozzle. In spraying use a good force pump. A Vermorel nozzle affords a well distributed fine spray. For dilute Bordeaux reduce the copper sulphate one-half, or double the volume of water. This may be sprayed upon soils to check the damping off of seedlings. Stock solutions of the two fluids may be prepared in advance and will keep indefinitely. They are conveniently kept in tubs or half barrels closely covered, the lime always under the required volume of water, and the copper sulphate in strong solution. When needed one then has only to measure out a portion of the copper sulphate water, dilute it to the required volume, stir up the settled lime very thoroughly , dip out the required volume of the milk of lime quickly, and pour the two fluids together, as already described. An acid Bordeaux should never be sprayed upon plants. The following are tests for acid Bordeaux: (1) A film of metalic copper deposited on polished iron or steel when plunged into the mixture; (2) a purp'ish red reaction on puttine into the Bordeaux a drop of a water solution of yellow prussiate of potash (1 to 10). If either of these reactions is obtained more lime must be added. It is best to avoid dry Bordeaux and similar commercial substitutes. Resin Bordeaux. This is made by adding to each 50 gallons of Bordeaux a clear liquid made by boiling for one hour 1 pound resin and 0.5 pounds crystals sal soda in 0.5 gallon water. Another way of making it is to melt 5 pounds of resin in 1 pint of fish oil, slowly add 1 pound of potash lye, stirring, and taking care that it does not become too hot and boil over. Then add 2 gallons of water and continue boiling for an hour. Finally add slowly with stirring an additional 3 gallons of water. The finished product should dissolve readily in cold water. Two gallons of this soap is added to each 50 gallons of the finished Bordeaux. Resin fish oil soap may also be bought and may be added to Bordeaux at the rate of 5 pounds per 50 gallons. Soda Bordeaux. Soda lye 1 pound; copper sulphate 3 pounds; lime 5 ounces; water 50 gallons. . I /senate of Lead. Three pounds to 50 gallons of water, mixed thoroughly with a little water first. FORMULAE. 205 Ar senile of Lime. Arsenic, 1 pound; stone lime, 4 pounds; water, 4 gallons. Boil half an hour then dilute to 200 gallons with water. Paris Green. Paris green, 0.5 pound; lime, 1.5 pounds; water, 50 gallons, or perhaps better: Paris green, 1 pound; quick lime, 3 pounds; water, 250 gallons. Combined Bordeaux Mixture and Paris Green or Arsenate of Lead. Add to the regular 50 gallon Bordeaux, 1 pound of Paris green stirred up thoroughly in a gallon of water, and stir thoroughly after- wards. If arsenate of lead is used, double the amount may be added in the same way. Boiled Lime-Sulphur. Lime, 25 pounds; sulphur, 17.5 pounds; water, 50 gallons. Boil 1 hour. Apply at once. Self-Boiled Lime-Sulphur {Scott's method). Sulphur, 10 pounds; stone lime, 15 pounds; water, 50 gallons. Put the lime into a barrel and pour over it 2 to 3 gallons of boiling hot water, add the sulphur at once, then 2 or 3 additional gallons of the hot water. Stir frequently. More water may be added if it becomes too thick, but add as little as possible. The cooking should take place in about 6 to 8 gallons of water. The mouth of the barrel should be covered to retain the heat. When slaked add the remainder of the water, i. e., cool quickly. Strain. Apply at once. Lime-Salt-Sulphur. Best stone lime, 30 pounds; sulphur, 15 pounds; salt, 10 pounds; water, 50 gallons. Slake the lime in hot water, then while hot add the sulphur and enough water to make a thin paste and boil for three-fourths hour, stirring thoroughly, adding more water as it evaporates. Then add the salt, boil an additional 15 minutes, dilute with hot water, filter and spray hot. Warren gives the following method of prepa- ration: Fresh lime, 15 pounds; flowers of sulphur, 15 pounds; salt, 15 pounds; water, 45 gallons. Bring 4 or 5 gallons of water to a boil in an iron kettle, mix the sulphur with hot water, crushing the lumps, then put into the boiler, add the lime in 4 separate parts, adding cold water gradually to subdue the violent boiling and prevent from overflowing. Finally add the salt, boil 1 hour or more, stirring frequently. Strain, dilute with the remainder of the 45 gallons (about two-thirds) and spray. There are other formula? in which the pro- portions vary somewhat. Potassium Sulphide. One pound to 50 gallons of water. To be used at once, because it soon loses strength. Carbon Bisulphide. Use 1 pound to each 100 bushels of grain, or 1 teaspoonful to each cubic foot of space. Mercuric Chloride. Solution of 1 part to 1000 parts of water. To be used in glass or wooden vessels, never in metal ones. For field use tablets may now be purchased so that it is only necessary to dissolve the requisite number in a given volume of water. Seed corn may be exposed 20 minutes with entire safety, wetting first in alcohol for a minute or two. Unsprouted potatoes 40 min- utes to 1 hour. Hydrocyanic Acid Gas. For treating dormant nursery stock, W. E. Britton, recoinmends 1 ounce cyanide of potash, 2 ounces sulphuric acid and 4 ounces of water for each 100 cubic feet of space. The acid is poured into the water, never the reverse, on account of over heating and danger of steam explosions; the cyanide is added, and the room shut up tight for half an hour. For greenhouse fumigation Woods and Dorsett recommended 20 minutes' exposure using 0.075 gram to 0.15 gram cyanide of potash per cubic foot, depend- ing on the kind of plants, ferns being very sensitive, and violets rather resistant. Symons has shown that dormant peach buds will endure 0.50 gram of potassium cyanide per cubic foot (2 ounces per 100 cubic feet) for 60 minutes. Apples will endure as much. The use of 0.30 gram per cubic foot for 30 minutes is scarcely sufficient to kill all of the San Jose scale, but 0.30 gram for 45 minutes would be. Formalin (40 per cent Formaldehyde). 1 pint to 50 gallons of water for smut of wheat and oats; 2 pints to 50 gallons for scab of potato; 4 pints to 50 gallons for disinfection of soils. The formalin should be taken from sealed (fresh) bottles, as it loses strength readily. Hydrogen Peroxide. Use 1 part to 200 of water. Must be fresh Whale Oil Soap. This may be used for plant lice at the rate of 2 pounds per 12 gallons of water. Dissolve in hot water. In greater concentration it should be tried in advance on a few plants. In proportion of 1 pound to 4 gallons of water it is said to injure tender plants (J. B. Smith). 206 BACTERIA IN RELATION TO PLANT DISEASES. Tobacco with Whale Oil Soap. Boil 3 pounds of dry tobacco stems or leaves in 10 gallons of water and add while hot 0.5 pound whale oil soap. Soap. Use 1 pound to 8 gallons of water. Kerosene Emulsion. Dissolve 1 pound of soap in 2 gallons of hot water, add 4 gallons of kerosene and churn for 15 minutes with a force pump. The thick creamv emulsion, which does not separate readily, is diluted for use with 9 parts of water and sprayed at once. Use rain water. If a good emulsion has not been obtained, do over. An imperfect emulsion must never be sprayed. Soluble Oil. This substance, sold under various names as Kill-o-Scale, Target Brand Scale Destroyer, is sprayed on the dormant plants after proper dilution (1 : 20). Formalin Vapor For freeing closed spaces from b Schering's formalin lamp may be used. For freeing closed spaces from bacteria LITERATURE. Germicides — Insecticides. 1890. Arthur, J. C. Treatment for smut in wheat. 1895. Purdue University Agr. Exp. Sta., Lafayette, Indiana. July, 1890, Bull. 32, vn, pp. 3 to 9. 1890. Kellerman,W. A. and Swingle, W. T. Pre- liminary experiments with fungicides for stinking smut of wheat. Bull. No. 12, Exp. 1896. Sta., Kansas State Agric. College, Aug., 1890, pp. 27-51. 1890. Chester, F. D. Diseases of the vine, controlled by several different salts of copper. Del. Agr. Exp. Sta., Oct. 1890. Bull. 10, 32 pp. 2 figs. 1 891. Kellerman, W. A. and Swingle, W. T. Additional experiments and observations on oat smut, made in 1890. Bull. No. 15, Exp. 1897. Sta., Kans. State Agric. College, Topeka, 1891, pp. 93-133. 1 89 1. Arthur, J. C. The loose smut of oats. Purdue University Agr. Exp. Sta., Lafayette, Indiana. Mar., 1891. Bull. 35, vol. II, pp. 81-97, 4 figs. 1897. 1 89 1. Kellerman, W. A. Second report on fungicides for stinking smut of wheat. Kans. Agr. Exp. Sta., Aug., 1891, Bull. 21, pp. 45-72, 1 plate. 1898. 1893. Hitchcock, A. S. and Carleton, Mark A. The effect of fungicides upon the germination of corn. Exp. Sta., Kans. State Agr. College. 1898. Manhattan, Kansas, 1893. Bull. 41, pp. 631- 679, Bibliog. of 26 titles. 1899. 1893. Kellerman, \V. A. Experiments in germina- tion of treated seed. Ohio Agr. Exp. Sta. Bull., April, 1893, vol. 1, No. 3, Tech. ser., pp. 201-205. 1899. 1 894. Frank and Kru'gER. Ueber den Reiz welchen die Behandlung mit Kupfer auf die Kartoffel- pflanze hervorbringt Ber. d. deutsch. bot. Ges., 1894, pp. 8-1 1. 1899. 1894. Galloway, B. T. The effect of spraying with fungicides on the growth of nursery stock. U. S. Dept. of Agr., Div. of Veg. Path., 1894, Bull. 7. Galloway, B. T. and Woods, A. F. Spraying with fungicides as a means of increasing the growth and productiveness of plants. Repr. from Proc. Soc. Prom. Agr. Sci., 1895. 16th Ann. Meeting, Springfield, Mass., pp. 42-53. Evans, W. H. Copper sulphate and germina- tion. Treatment of seed with copper sulphate to prevent the attacks of fungi. U S. Dept. of Agr., Div. of Veg. Path., 1896, Bull. 10, 24 pp. Effect on germination. Summary of other workers' experi- ments with oats, wheat, barley, etc. — Abstracts presenting many contradictory opinions relating to use of copper sulphate for prevention of smut. Effect on germination, root system, growth of aerial parts. Bolley, H. L. New studies upon the smut of wheat, oats, and barley, with a resume of treatment experiments for the last three years. Gov. Agr. Exp. Sta. for North Dakota. Fargo, Mar., 1897. Bull. 27, pp. 109-162, 13 figs. Hicks, G. H., and Dabney, J. C. Vitality of seed treated with carbon bisulphid. U. S. Dept. of Agr., Div. of Bot., Circular 11, 1897. Harrison, F. C. The effect of spraying Bor- deaux mixture on foliage. 23d Ann. Rep. Ont. Agr. College Exp. Farm, 1898, pp. 125-128. Starnes, H. N. Some peach notes. Ga. Agr. Exp. Sta., Nov., 1898, Bull. 42. Duggar, B. M. Peach leaf-curl and notes on the shot-hole effect on peaches and plums. Cornell Agr. Exp. Sta., Feb. 1899, Bull. 164, PP- 371-388, figs. 64-72. LinharT. I. Krankheiten des Riibensamens. II. Bekampfung der infektioscn Krankheiten des Riibensamens. Sep. Oester. Ung. Zeitschr. f. Zuckerindustrie, 1899, 1, 11, IV. Woods, Albert F., and Dorsett, P.H. The use of hydrocyanic acid gas for fumigating greenhouses and cold frames. Circular No. 37, Second series, Div. of Entomology, U. S. Dept. of Agriculture, Jan., 1899, 10 pp. 3 figs. GERMICIDES — INSECTICIDES. 207 1900. Kittlaufs, K. Ueber die Einwirkung der Kupfer-vitriol-Beize auf die Keimkraft des Saatgetreides bei verschiedener Zeitdauer und Starke der Losung. Fiihling's landwirt- schaftliehe Ztg., Stuttgart, Jahr. 1899, pp. 572- 586, 605-616. Auszug, Biedermann's Centr., Leipzig, 1900, Bd. xxix, p. 471. 1900. Pierce, N. B. Peach Leaf-Curl: Its nature and treatment. U. S. Dept. of Agr., Veg. Phys. and Path., 1900, Bull. 20, 204 pp., 30 pis., 10 figs., 24 tables. 1901. Arthur, J. C. and Stuart, W. Formalin and hot water as preventives of loose smut of wheat. 13th Ann. Rep. Ind. Agr. Exp. Sta., 1901, pp. 17-24- 1901. Cranefield.F. The influence of formalin on the germination of oats. 18th Ann. Rep. Wis. Exp. Sta., 1901, pp. 327-335. 1901. Demoussy, E. La germination des grains de ble traites au sulfate de cuivre. Annales Agronomiques, 1901, pp. 257-261. 1 901. Fantecchi, P. Influenza dei trattamenti con solfuro di carbonio sulla germinazione del grano. Bolletino di Entomologie agraria. Padua, 1901, vol. 8, pp. 38-39. 1901. Moore, R. A. Treatment of seed oats to prevent smut. 18th Ann. Rep. Agr. Exp. Sta., Univ. of Wis., Madison, 1901, pp. 255- 260. 1901. Shamel, A. D. Treatment of oats for smut. 111. Exp. Sta., 1901, Bull. 64, pp. 57-71, 6figs. 1901. Sturgis, W. C. Peach foliage and fungicides. Conn. Agr. Exp. Sta., Ann. Rep. for 1910, New Haven, 1901, pp. 219-254, plates 3-5, 6 tables. Spray injury. 1 901. TownsEnd, C. O. The effect of hydrocyanic acid gas upon grains and other seeds. Md. Exp. Sta., 1901, Bull. 75, pp. 183-198; Bot. Gaz., 1901, vol. 31, pp. 241-264. 6 figs. 1901. Tubeuf, Carl von. Anwendbarkeit von Kupfermitteln gegen Pflanzenkrankheiten. K. Gesundheitsamt, Biol. Abt. Arb., Berlin, 1901, Bd. 2, Heft. 2, pp. 367-368. Fungicidal action of copper. 1 90 1. Weits, J. Die Brandpilze und ihre erfolgreiche Bekampfungdurchzweckmassiges Beiszen des Saatgutes. Wochenblatt des landwirtsch. Vereins in Bayern, Miinchen, '91, Jahrg. 1901, pp. 733-734- 1901. Windisch, Richard. Ueber die Einwirkung des Formaldehyds auf die Keimung. Land- wirtschaft. Versuchs Stat. Berlin, 1901, Bd. lv. , pp. 241-252. 1902. Paddock, W. Plant diseases of 1901. Col. Agr. Exp., Sta., March, 1902, Bull. 69, 23 p., 9 plates. Spray injury — copper. 1902. Bain, S. M. The action of copper on leaves, with special reference to the injurious effects of fungicides on peach foliage; a physiological investigation. Tenn. Agr. Exp. Sta. Bull., Apr. 1902, vol. 15, No. 2, 108 pp. 8 plates. 1902. Saunders, D. A. Treatment of smuts and rusts S. Dak., Exp. Sta., 1902, Bull. 75, 7 pages. 1902. Stewart, F. C. and Eustace, H. J. Spotting and dropping of apple leaves caused by- spraying. N. Y. State Agr. Exp. Sta., Dec. 1902. Bull. 220, pp. 217-233, 5 plates. 1902. Cranefield.F. The influence of formaldehyde on the germination of oats. 19th Ann. Rep. Exp. Sta., Wis., 1902, pp. 268-272. 1903. Cobb, N. A. Effect of engine boiler steam on the vitality of seeds and spores. Agr. Gaz. of N. S. Wales, 1903, Bd. 14, pp. 26-29. • 903. Jacevskij. Die sterilisation der Samen un- serer Kulturpflanzen als Schutz gegen die Pilzkrankheiten. (Russ.) Zemled. Gazeta, St. Petersburg, 1903, pp. 870-872; 921-923; 969-970. 1903. REED, Z. Treatment of stinking smut in wheat. Bull. 79. Col. Exp. Sta. 1903. 1903. Strawson, G. F. Standard fungicides and insecticides in agriculture, with notes on charlock destruction. Part I. London (Spottiswoode) 1903, 76 pp. 1904. Ruhland, W. Zur Kenntnis der Wirkung des unloslichen basischen Kupfers auf Pflanzen mit Riicksicht auf die sogenannte Bordeaux- briihe. Arb. K. Gesundheitsamt, Berlin, 1904, Biol. Abt. Bd. 4, Heft. 2, pp. 157-200. 1904. Moore, R. A. Treatment of seed grain for the prevention of smut. Agr. Exp. Sta. Wis., Rep. for 1903, Madison, 1904, pp. 284- 292, incl. pi. 1904. Salmon, Ernest S. Cultural experiments with the barley mildew, Erysiphe graminis DC. Ann. Mycol., Berlin, Jan., 1904, vol. 2, No. 1, pp. 70-99. Tables. Effect of copper sulfate as a fungicide when absorbed by the roots of cereals. 1904. Schander, Richard. Ueber die physiologische Wirkung der Kupfervitriolkalkbruhe. Landw. Jahrb. Berlin, 1904. Bd. 33, pp. 517-584. 1904. Wheeler, W. A. Preliminary experiments with vapor treatments for the prevention of the stinking smut of wheat. Bull. 89. S. D. Exp. Sta., 1904, 19 pages. 1904. Aderhold, Rud. Der heutige Stand unserer Kenntnisse iiber die Wirkung und Verwertung der Bordeauxbriihe als Pflanzenschutzmittel. Jahresbericht der Vereinigung der Vertreter der angewandten Botanik., Erster Jahrgang 1903; Verlag von Gebriider Borntraeger, Berlin, 1904, pp. 12-36. 1905. Farrer, W. and Sutton, G. L. The effects of some solutions of formalin and bluestone which are in common use, on the germination of wheat seeds. Agr. Gaz. N. S. Wales, 1905, vol. 16, pp. 1248-1255. Effects of formalin and bluestone vary with different varieties. 1905. McAlpine, D. Treatment of seed for fungus diseases. Jour. Dept. Agr. of Victoria, Melbourne, 1905, vol. 3, pp. 187-188. 1905. McAlpine, D. Germination test of seed wheat treated with formalin. Jour. Agr. Dept. of Victoria, 1905, vol. 3, pp. 266-267. Five lots of 1 .000 grains each were soaked 15 minutes, each lot in a different strength of formalin. Check lot of 1. 000 seeds. Conclusion: Schering's formalin. I lb. in 40 gal. of water exercise? no injurious influence. 1905. SchrEnk, Hermann von. Intumescences formed as a result of chemical stimulation. 1 6th Ann. Rept. Mo. Bot. Gard., issued May 31, 1905, pp. 125-148, pis. 25-31. Effect on cauliflower of ammonium copper carbonate spray: The slight injury caused excessive cell multiplication in restricted areas, mostly on the under surface of the leaves, where hundreds of wart-like growths developed. 1906. KraemER, Henry. Dilute Sulphuric Acid as a Fungicide. Proc. Amer. Philos. Soc, 1906, vol. xlv, pp. 157-163. Also a separate. 1000 sulphuric acid water for mildews of Recommends roses, etc. 1906. Waite, M. B. Fungicides and their use in preventing diseases of fruits. Farmers' Bull. No. 243, U. S. Dept. of Agr., Feb. 1906, 32 pp., 17 figs. 208 BACTERIA IN RELATION TO PLANT DISEASES. 1906. Warren, G. F. Spraying. Bull. No. 194, N. J. Agric. Exp. Sta., March 1906, 60 pp. 1907. Amos, Arthur. The effect of fungicides upon the assimilation of carbon dioxide by green leaves. Journ. of Agr. Sci., Dec, 1907, vol. 11, part 3, pp. 257-266. Experimented with leaves of hop. grape-vine, and Jerusalem artichoke. Application of Bordeaux to the leaves diminishes the carbon dioxide assimilation for a time but after a time the effect passes off. 1908. Kirchner. Ueber die Beeinflussung der Assimilationstatigkeit von Kartoffelpflanzen durch Bespritzung rnit Kupfervitriolkalkbriihe. Zeitsch. f. Pflanzenkr., 1908, Bd. xvin, Heft. 2. 1908. Chick, Harrietts, and Martin, C. J. The Principles involved in the standardisation of disinfectants and the influence of organic matter upon germicidal value. Journal of Hygiene, Cambridge, Nov., 1908, vol. 8, No. 5, pp. 654-697. Bibliography of 37 titles given. 1908. Scott, W. M. Self-boiled lime-sulphur mixture as a promising fungicide. Circular No. 1, Bureau of Plant Industry, LT. S. Dept. of Agric, April, 1908, 18 pp., 2 figs. 1908. Smith, John B. Insecticide materials and their application: with suggestions for practice. Bull. 213, N. J. Agric. Exp. Sta., Sept. 1908, 46 pp. 1908. Symons, Thomas B. Miscellaneous treatment for San Jose scale. Bull. 131, Maryland Agric. Exp. Sta., Nov., 1908, pp. 129-149. 1908. Chick, HarriEtte, and Martin, C. J. A comparison of the power of a germicide emulsified or dissolved, with an interpretation of the superiority of the emulsified form. The Journal of Hygiene, Cambridge, Nov., 1908, vol. 8, No. 5, pp. 698-703. 1908. R ideal, S. On testing disinfectants. A lecture delivered at the London School of Tropical Medicine. The Journal of Tropical Medicine and Hygiene, May 1, 1908, vol. xi, No. 9, PP- 133-135 1908. Ehrlich, P. Ueber Moderne Chemotherapie. Verhandl. deutsch. dermat. Ges. x Congress, 1908, pp. 52-70. 1909. Ehrlich, P. Ueber den jetzigen Stand der Chemotherapie. Ber. d. deutsch. chem. Ges., Bd. xlvii. Rev. in Bull, de l'lnst. Pasteur, Tome vii, 1909, pp. 321-324. 1909. Roehl, W. Heilversuche mit Arsenophenyl- glycin bei Trypanosomiasis. Zeitschr. f. Immun. forsch. u. exp. Ther. 1909, Bd. 1, pp. 633-649. Rev. in Bull, d l'lnst. Pasteur, 1909, Tome vii, pp. 335-336. 1909. Stevens, F. L. Experiments upon the effect of formalin upon the germination of oats. Thirty-first Annual Report, North Carolina Agric. Exper. Sta., West Raleigh, N. C, June, 1908, pp. 30-36. 1909. Crandall, Charles S. Bordeaux mixture. Bull. No. 135, Agric Exp. Sta., Urbana, Illinois, May, 1909, pp. 201-296. 1909. PopEnoe, C. H. The Colorado potato beetle in Virginia in 1908. U. S. Dept. of Agric, Bureau of Entomology, Bull. No. 82, Part 1. Washington, 1909, pp. 1-8, 2 plates. 1910. The Duke of Bedford and Spencer U. Pickering. Eleventh Report of the Woburn Experimental Fruit Farm [A General Treatise on Copper Fungicides]. The Amalgamated Press, Ltd., London, 1910, vi, 190 pp., with an appendix. 1910. Scott, W. M. and Ayers, T. Willard. The Control of Peach Brown-Rot and Scab. Bull. No. 174, B. P. I., U. S. Dept of Agric, March 5, 1910, 31 pp., 4 pis. 191 1. Scott, W. M. and Quaintance, A. L. Spray- ing Peaches for the Control of Brown-Rot, Scab, and Curculio. Farmers' Bull. 440. U. S. Dept. of Agric, March 27, 191 1, 40 pp., 14 figs. VASCULAR DISEASES. WILT OF CUCURBITS. (Synonyms: Cucumber-wilt; Cantaloupe-wilt; Squash-blight; Pumpkin-blight). DEFINITION. This is a specific communicable disease of cucumbers, squashes, and some allied plants. It is characterized by the sudden wilting and shriveling of the foliage and by the presence in the vascular bundles of enormous numbers of a white sticky bacillus which is the cause of the disease. HOST-PLANTS. This disease has been observed in cucumbers {Cucumis sativus), muskmelons {Cucumis meld), pumpkins (Cucurbita pepo), and squashes (Cucurbita moschata and C. maxima). The disease has been successfully inoculated by the writer into all of the above mentioned plants many times over and into the following additional cucurbits : Cucumis odoratissimus, Benincasa cerijera, Cucumis angaria, Cucurbita foetidissima, C. californica, Sicyos angulatus and Echiuocystis lobata, the last four being wild plants of the United States. The disease is not known to the writer to occur in the watermelon but it has been reported by Selby. Inoculations into this plant, while sometimes producing a wilt of the pricked leaf generally failed to induce any secondary wilt. In one or two instances other leaves than the inocu- lated ones wilted and the vessels were found plugged by a bacillus. Often there was no wilt even in the punctured leaves. Inoculations from virulent cultures into the following cucurbits failed, or produced only local injuries from which the plants recovered: Melothria scabra, Cucumis erinaceus, Lujja acutangula, Momordica balsamina, Lagcnaria vulgaris, Trichosauthcs cucumeroidcs, Apodanthera undulata. The disease is not known to occur in any wild plant, but it is so very easily inoculated into Sicyos angulatus and Cucurbita foetidissima that it should be searched for on these plants and on other wild cucurbits. Inoculations into non-cucurbitaceous plants such as Solatium tuberosum, Lycopersicum esculentum, Datura stramonium, Passiflora incarnata, Yigna catjang, Nicotiana tabacum, Pyrus orientalis and Hyacinthus orientalis yielded only negative results. The disease is not known to occur outside the Cucurbitaceae, and probably many species of plants within the limits of this family are not subject to it. GEOGRAPHICAL DISTRIBUTION. The limits of this disease are not known. It occurs in Canada, Massachusetts, Ver- mont (?), Connecticut, New York, New Jersey, Delaware, Maryland, Virginia, West Vir- ginia, Pennsylvania, Kentucky, Ohio, Indiana, Illinois, Michigan, Wisconsin, Missouri, Iowa, Nebraska and Colorado. It is not known to occur south of latitude 350 and it is believed to be restricted in its southern distribution by the fact that the bacillus is very sensitive to heat. It should be searched for, however, in the Gulf States. The disease has been reported from Germany by Dr. Otto Appel; from Russia, near St. Petersburg, by Dr. Iwanoff ; and it is to be looked for in all north-temperate regions where cucurbitaceous plants are grown and where the temperature to which the vines are exposed does not exceed the thermal death-point, or the maximum temperature for the growth of this organism. 209 & 2IO BACTERIA IN RELATION TO PLANT DISEASES. Nothing is known to the writer concerning its occurrence in the far East or in the south- temperate zone except a statement to him in 1910 by I. B. Pole Evans that it occurs on pumpkins in the Transvaal. Contrary to certain statements the disease occurs in hothouses as well as in the fields, but it is more generally prevalent in open-air culture than under glass. It occurred naturally, however, two different years in a hothouse near Washington and in 1899 the writer iden- tified it from a hothouse at Morrison, Illinois, where it did considerable injury. The organism was cultivated out of plants from both houses and the disease was reproduced by inoculations from these cultures. It is perhaps worth while to discuss the occurrence of this disease in the United States at greater length. This disease was first observed by the writer (in 1893) near Washington, D. C. (plate 13), and has been observed in fields and gardens around Washington every year since 1893 (fig. 50). Fig. 50.* I saw it at Chuckatuck in Nansemond Co., Virginia, in 1898. This is the farthest south I have seen it. It was cultivated pure from the interior of cucumber-stems received from Bristow, Prince William Co., Virginia, July 19, 1897. Nothing is known of its occurrence south of Virginia. It might be looked for farther south in the mountains, but hardly on the flat lands of the far South owing to its low thermal death point. It is northern rather than southern in its distribution. It occurs in Pennsylvania and throughout New Jersey and Delaware, at least in places. In September 1901, I saw the disease in cucumbers at Woods Hole, Mass., and in October 1903, in muskmelons farther north on Buzzard's Bay. I received it from Connecticut in 1905 (melons), and from Rhode Island in 191 1 (cucumbers). It probably occurs all over New England. I saw it in a field of cucumbers on the western end of Long Island in July, 1902, at least one-third of the vines being affected. Gnawings due to the striped beetle (Diabrotica) were numerous and many of the wilted spots originated from the bitten places (see Etiology). *Fig. 50. — Patch of diseased Hubbard squashes on place of David Fairchild, at Chevy Chase, Md. Plants wilted by Bacillus trai heiphilus Summer of u>n-. Photographed by Mr. Fairchild. PLANT BACTERIA, VOL. 2. PLATE 13. Cucumbers showing various stages of bacterial wilt due to Bacillus tracheiphilus. Photographed by writer July 13, 1893, on a hill-side at Anacostia, D. C: Fig. 1 (top), healthy vines for comparison with diseased plants on same terrace. Fig. 2, primary and secondary wilt. Leaf-blades toward base of stem (left) have been destroyed by direct infection (beetles) and bacteria have penetrated into main axis. Leaves toward apex of stem have wilted as result of disturbed water-supply due principally to occlusion and destruction of vascular bundles in stem. Fig. 3. only two turgid leaves remain; petioles and stems were still green and normal in appearance. Bacteria filled the spiral vessels of main shoots with a white viscid slime. Fig. 4 (bottom), no sound leaves; stems beginning to shrivel. WILT OF CUCURBITS. 21 I Some years ago near Albany, New York, it did much damage to fields of cucumbers according to John E. W. Tracy of the U. S. Department of Agriculture, who also saw it near Rochester. This disease was common in squashes and cucumbers at Hubbardston, Michigan, in 1895 and subsequently. On August 20, 1897, I saw near Saginaw, Michigan, two fields of cucumbers in which the disease was present. There were vines in all stages of wilt from those just beginning to be affected to those which were dried up. When I made cross-sections of the stems toward the root and touched the cut end with my finger the bacilli strung out in the char- acteristic way in delicate sticky threads. On showing the plants to a cucumber-grower he said he was familiar with the disease but did not know its cause. During the same month I saw this wilt in cucumbers and squashes at Grand Rapids, Michigan. In 1898, Mr. S. S. Bailey, of Grand Rapids, Michigan, lost many of his squashes by this disease. It was seen by the writer in muskmelons at Racine, Wisconsin, in 1897. It has been reported by Prof. Pammel from Iowa. Mr. Ragan, the horticulturist tells me he has seen it in Indiana. Fig. 51.* Specimens of cucumber attacked by this disease were received by me from Morrison, 111., in 1899. The slime was so sticky that when touched with the finger, it strung out from the end of the cut stem over 76 cm. (30.5 inches). It also occurs in the vicinity of St. Joseph, Mo. Mr. Barlow has observed it at Guelph, Ontario, in a variety of cucurbits. SIGNS OF THE DISEASE. This disease is readily detected owing to the striking nature of the phenomena. The wilt is first local, affecting certain individual leaves (plate 1, fig. 1 and various text figures), but soon becomes general, involving the foliage of the entire plant (plates 13 and 14). Asso- ciated with the wilt we always find a white ooze exuding from the vascular bundles of leaves or stems on cross-section (fig. 51), and this exudate is usually viscid. The only other dis- eases of cucurbits liable to be confounded with this are: (1) The more or less sudden wilt due to the presence of the larvae of the squash-vine-borer {Aegeria cucurbitae) in the base *Fig. 51. — Cross-sections of cucumber-stems, showing bacterial ooze (Bacillus trachciphilus) from the bundles. Plants from New York. Photographed Aug. 1 1, 1904. Enlarged about ten times. 212 BACTERIA IN RELATION TO PLANT DISEASES. of the stem or main root; (2) A sudden wilt due to the filling of the vascular bundles with fungi of the form-genus Fusarium; and (3) A wilt due to the rotting off of the main stem at the surface of the earth. This disease may be distinguished readily enough by the facts that fungi are not present and that there is no stem-injury or root-injury of the kinds just described, and also by the further fact of the invariable presence of large numbers of white, sticky bacteria in the vascular system. These are so abundant and usually so viscid that if the tip of the finger be pressed against the cross-section of a diseased stem at once, or better, some minutes after cutting, and then gently removed, the bacteria will remain attached to the finger and string out in numerous delicate threads (fig. 52) resembling cobwebs. For the microscopic structure of these threads consult Vol. I (fig. 14). If a little time is allowed the bacteria also ooze from the cut surface (cross-section) of such stems in milk-white drops, especially if the stems are cut a second time and the basal end put into water or moist air. The wilting and shriveling of the leaf blades always precede the destruction of the leaf- stalks and of the stem by a considerable period, so that it is common to find plants which have lost all or nearly all of their foliage while still retaining a green and normal looking stem (plate 1, fig. 2), the vessels of which, however, for long distances will be found to be more or less fully occupied by the bacillus (fig. 6). In the end, petioles and stems shrivel and die, but the organism does not make its appearance on the surface of the plants and there is nothing resembling a soft wet-rot, not even in the fruits. In rather resistant plants, c. g., certain squashes, the foliage may wilt dur- ing warm, dry days and partially recover at night or during cool, moist days, to wilt again when the demands of transpiration are greater. In such plants there is dwarfing (fig. 53) accompanied, in some instances at least, by an excessive blossoming and branch- ing. In the cucumber and muskmelon the disease is, on the con- trary, quite speedily destructive, a few weeks after the close of the period of incubation being generally sufficient to destroy the plants. There is in these species much less tendency to recover from the wilt temporarily during cool or wet weather than there is in the squash, and the writer has not observed any prolifera- tion of shoots or flowers. When the disease is active there is seldom any yellowing of the foliage in advance of the wilt. The loss of turgor and change of color (from bright green to dull green) are sudden. The characteristic signs are well exhibited in the accompanying illustrations. The disease generally starts in the center of a hill, i.e., on the blades of the basal leaves which soon shrivel. In this stage of the disease in that part of the main axis near the pri- mary infections the vascular bundles, and especially the spiral vessels, are gorged with the bacteria and there are usually many bacterial cavities in the primary vessel-parenchyma (fig- 54)- ETIOLOGY. Fig. 52.' The cause of this disease is a white peritrichiate schizomycete named by the writer Bacillus tracheiphilus from its special fondness for the vessels of the plant. This organism was first isolated and described by the writer (1893-95), an^ most of the statements here given rest upon his own observations and experiments, which now cover a period of 18 years. The disease is very readily induced by needle-punctures without hypodermic injection. All that is necessary to produce the disease is to dip the end of a sterile steel needle into a recent *Fig. 52. — Viscid threads of B. tracheiphilus stringing from cut end of a cucumber-stem. of 1904. Threads slightly diagrammatic, i. e., not sufficiently cobwebby. Plant from field, autumn PLANT BACTERIA. VOL. 2. PLATE 14. Wilt of Cucurbits. Cucumber-plants inoculated with B. tracheiphilus in summer of 1903 (hothouse, U. S. Department of Agriculture, by James Birch Rorer). Infections were by means of needle-pricks, using pure cultures. In each case needle-pricks were eon fined to a single leaf-blade. This blade first wilted and then gradually all the other leaves. Inoculations were made on July 2J and photograph about lS days later. Much reduced. WILT OF CUCURBITS. 213 pure culture on steamed potato or nutrient agar, or in beef-bouillon, and make a few deli- cate punctures into a susceptible plant, e. g., into the blade of a cucumber-leaf or musk- melon-leaf. The period of incubation in the writer's experiments has varied from 3 to 31 days and must depend partly at least on the number of bacilli inserted. Ordinarily when 20 or 30 needle-pricks are made the first signs appear in from 5 to 9 days in the punctured part of the leaf (figs. 56, 60, 63, 74). When young cultures are used on very susceptible plants such as Cucumis sativus, Cucumis melo, or Cucurbita foetidissima, the disease appears with the certainty and regularity of clock-work. It is more difficult to inoculate squashes successfully, at least with some strains of the organism, and this corresponds to the observed fact that they are more resistant in the field. One winter, on several kinds of squashes the writer experienced repeated failures, using virulent cultures obtained from the cucum- ber. The pricked cucumber-plants and muskmelon-plants contracted the disease ; the squashes, both summer and winter varieties, inoculated at the same time, in the same way, and from the same cultures, resisted, or only showed traces of primary wilt. This Fig. 53.* resistance may be due to some extent to varying degrees of virulence on the part of par- ticular strains of the organism; or to varying degrees of resistance on the part of the host. Possibly the squash bacillus should be regarded as a variety. In the summer of 1905 the question of the identity of the squash-wilt and cucumber- wilt was gone over once more. Inoculations made into four varieties of squashes, using a strain isolated the previous year from a muskmelon, and proved by numerous control - experiments to be virulent to cucumbers, would not infect squashes. A little later in the season the same squashes were readily infected with a strain of the bacillus isolated from the vessels of a wilting squash-plant found in a garden in Washington, and the same cultures *Fig. 53. — A. Winter squash, variety Pikes Peak, No. 215, inoculated Oct. 5, 1895, by needle-pricks on two leaf blades, using viscid white slime from a cucumber-stem. Both leaves contracted the disease and shriveled slowly, one of them being shown at X- Although the plant was under observation for 66 days the only additional signs of disease were conspicuous dwarfing with yellowing of the foliage, especially the lower leaves. Photograph made Dec. 10; plant then cut and examined under miscroscope, bacteria being demonstrated in a few vessels of several (5) bundles. B. An uninoculated plant from the same lot of seedlings. About one-sixth natural size. 2I4 BACTERIA IN RELATION TO PLANT DISEASES. proved equally virulent to cucumbers, the disease occurring promptly and the signs being typical in all respects, including the presence of the sticky bacillus in the vascular system (plate 15, fig. 1). The inoculation-experiments were repeated a few weeks later with the same positive results; squashes and cucumbers being infected with uniform success. Addi- tional studies should be made. In the inoculated plant, the primary foliar signs (a dulled green with absence of turgor) always appear first in the punctured area and immediately around it, but never until after a definite period of incubation covering at least several days. The signs of disease gradually extend until the entire blade of the leaf is involved. The loss of turgor and change to dull green is soon followed by shriveling, after which the leaf-blade becomes brown. vSubsequently, and usually considerably prior to the collapse of the petiole of this leaf, the blades of other leaves up and down the stem suddenly wilt (plate 16 and text fig. 58). The first leaves to show this secondary wilting are ordinarily those which arise from parts of the stem nearest to the insertion of the inoculated leaf; exceptionally the first leaf to show secondary wilt is one standing over the inoculated leaf rather than one actually nearer but inserted on the opposite side of the stem. Gradually more and more remote leaves are destroyed until the whole plant is involved. Whenever this secondary stage of the dis- ease supervenes, the vessels in the stem (which still outwardly presents a green and normal appearance) will be found to be occupied more or less fully by the bacillus. Usually the organism is to be found in the vessels of such plants in extraordinarily large numbers. In the stem of the squash the writer traced the bacterial occupation microscopically in one plant to a distance of 210 cm. from the point of infection, and in another plant to a distance of 240 cm. Almost all of the writer's inocula- tions have been made by means of needle- punctures into the blade of the leaf, at first often directly from plant to plant, but in recent years generally from pure cultures (descendants of poured-plate colonies) on agar, carrot, potato or in beef-bouillon, and other fluids. This method closely resembles the natural manner of infection and has given very satisfactory results. These inocula- tions now number over 700. A great many check-plants were held for comparison and the number of accidental infections (when insect-carriers have been excluded) has been practically nil, whereas the number of successful inoculations in susceptible plants has frequently amounted to from 751085 per cent of the total number punctured. In certain experiments (pages 246, 276) every inoculated plant has contracted this disease, which is one of the most infectious known to the writer. So many experiments have been made, under such a variety of conditions, and with such good success (except in case of the squashes already mentioned) that not the least doubt remains, either as to the bacterial nature of this disease or as to the particular organism which causes it. Fig. 54/ *Fig. 54. — Cross-section of small portion of a cucumber-stem attacked by B. tracheiphilus showing condition of one of the outer bundles. The pitted vessels lie in the more heavily shaded lignified part of bundle and only a very few of them are occupied by bacteria. All the spiral vessels are filled and the bacteria have formed conspicuous cavities in the primary vessel-parenchyma which is a living non-lignified tissue. Other tissues are uninjured. Drawn from a photomicrograph. PLANT BACTERIA. VOL. 2. PLATE 15. — 0 p. a x — 3 S ,3 SU - E* £ 21 o i, _ id i3. X . ai .o 3 Ex.- S^ WILT OF CUCURBITS. 215 In the hothouse, the writer has succeeded in spreading this disease readily by means of leaf-eating beetles {Diabrotica vitatta). Moreover, numerous field observations seem to indicate quite clearly that this is a common method of dissemination. Leaf-eating insects, and especially Diabrotica vitatta (fig. 55), are, I believe, the chief agents in the spread of this disease. They feed readily, and sometimes the writer has thought preferably (fig. 7), on wilted leaves which are swarming with this organism. In this way their mouth-parts can not fail to become contaminated and to serve as carriers of the sticky infection. No other means of dissemination is known to the writer, and this is believed to be the common way in which the disease is distributed.* Seasonally the disease does not manifest itself until the leaf-eating beetles have put in their appearance, and this has led to the suspicion that the organism might pass the winter inside the bodies of these hibernating insects {Diabrotica vitatta). As to this nothing definite is known. The greater part of the bacilli as they occur in bouillon are easily killed by freezing, but it is likely that some winter over in the vegetative form or in some more resistant form, in suitable places in the soil. The writer has attempted to plate the organ- ism from the Diabroticas several times but always unsuccessfully, other organisms having speedily occupied the plates. Possibly the squash-bug {Coreus tristis) is also responsible for the distribution of this disease, but the evidence on which the writer formerly made this statement does not seem to him as conclusive as it did, i.e., the results obtained may be interpreted in another way, checks in sufficient number not having been made. The subject is open to further experi- ment, with the probabilities in favor of this bug being a carrier of the disease (see page 235). The large lady beetle, Epilachne borealis, is a greedy feeder on squash foliage but I have not seen it feeding on the bacterially wilted foliage. One experiment only was made with aphides, and this yielded negative results. Four cucumber-plants were sprayed thoroughly on both leaf-surfaces with one part of a potato- broth culture 2 days old diluted with three parts of water. On one of these plants 70 aphides {Aphis gossypii Glover) were colonized, and the other three were held as checks, all under bell-jars. The aphides (which were taken from watermelon-plants) crawled about on the wet surface and immediately began to puncture the plant in many places. In the end they injured it greatly, but no bacterial wilt appeared. This plant was under observation 32 days. Two of the check-plants remained free from the disease. The third was free from signs for the first 3 weeks, but lost all its leaves by wilt between the twenty-third and twenty-ninth day, and on examination of its stem at various levels the vessels were found *The above mentioned field observations were made by me several years ago. In the summer of 1905, accident enabled me to strengthen these conclusions. Several cucumber-leaves were inoculated late one afternoon in a hot- house containing about 20 well-grown plants. This house unknown to me contained a very fewspecimens of Diabrotica vitatta, and next morning it was observed that the punctured parts of the leaves (those parts wetted by the bouillon culture) had been gnawed out by this beetle, while the remainder of these particular leaves remained intact. It was also noted that leaves on various other plants had been eaten somewhat during the night, and as a result their infection was anticipated. This inference proved to be correct. About a week or 10 days later numerous cases developed. In every instance the wilt began in the leaves bitten by the Diabroticas immediately after the date of placing the infectious material on the leaves which were punctured. The latter also contracted the disease. One of the plants infected by gnawings of the Diabrotica is shown in vol. I, plate 23. The strain used for these inoculations was that mentioned as non-infectious for the squash. Two months later in a hothouse remote from the preceding (north part of the grounds U.S. Dept. Agric.) squashes were infected by needle-puncture with the strain already referred to as plated from a squash-stem and found infectious to cucumbers. Wilt of the leaves resulted. These wilted leaves were bitten by Diabrotica vitatta, of which the house contained a very few only. On a side bench about 15 or 20 feet away stood 16 fine young cucumber plants which I had propagated for a second set of experimentswith cucumbers. These were bitten at the same time as thewilting squash leaves, or soon after, and 15 of the 16 plants contracted the wilt within a few days, and in every case it began in the bitten leaves. The earliest (primary) stage of the disease in these insect-inoculated plants is shown in plate 15, fig. 2. In this case also there was no likely source of infection except the inoculated squash-leaves as there were no other cucurbits in the vicinity, and again the beetles were the carriers. Some muskmelons on the same bench were also bitten and inoculated by these beetles. Microscopic examinations were made, demonstrating the bacillus in the vessels, and typical poured-plate cultures were obtained. In the 3rd set of squashes inoculated in 1905, after wilted areas had developed on the leaves, it was twice observed that these areas were the only parts bitten by the Diabroticas on those particular leaves. See also pp. 281, 282, and 284. 2 Id BACTERIA IN RELATION TO PLANT DISEASES. to be full of bacilli. That this diluted culture was virulent is also shown by the fact that out of 10 large cucumber-plants inoculated from it the same day by needle-puncture, 8 promptly contracted the disease. Up to this time therefore, the weight of the evidence favors the view that aphides do not play any part in the dissemination of this disease. Further experiments should be made. The fact that one check-plant contracted the dis- ease in some unknown way shows that at least occasionally the disease may be induced by simple spraying in the absence of suctorial insects, and this is what invalidates the experi- ment with the squash-bugs. Some of the sprayed plants on which they were colonized contracted the disease, but the additional inference is of the post hoc sort. The disease seems to be worse in moist, warm weather than in dry cool weather, at the same time excessively hot weather seems to be unfavorable to its spread. A soft watery condition of the tissues is believed to be favorable to the spread of this disease. In a number of instances it has been observed to do most injury in wet seasons, but it is not restricted to such seasons. Possibly, the greater injury during rainy periods is attri- butable chiefly to the greater number of infections, favored by cloud-screens and the moisture of the air. In a dry air many infected wounds probably dry out before the bacillus has secured a foothold, or are rendered sterile by sunshine. The bacillus is so well distributed that if it were not for some such re- straining circumstances it is doubtful if ordinary cucurbitaceous plants could be grown at all in the Northeastern United States. Aside from suitable weather-conditions and the propagation of extra sensitive varieties, which should of course be avoided, the conditions most favorable to the spread of this disease, so far as yet known, are the multiplication of insect-depredators, particularly the leaf-eating beetles. Probably puncturing insects do less harm. Among growers of these plants there is, however, a widespread belief that Coreus tristis, the squash-bug, "poisons the plant, " and thispoisoning.as it is called, might well Further observations and experiments are necessary. The extent of the vascular infection soon after the first secondary wilt supervenes was studied in plant No. 18 a diagrammatic sketch of which is shown in fig. 59. This plant was inoculated by needle-pricks on the blade of one leaf. The second day after secondary wilt appeared, the entire plant was fixed in alcohol. Subsequently, portions of this plant were infiltrated with paraffin, cut, stained and studied for the presence of the bacteria at the different levels indicated in the figure. These sections show that in the course of the 15 days which intervened between the needle-pricks on the leaf-blade at X and the fixing of the tissues in alcohol, the bacteria had penetrated into some portion of the vascular system of nearly every organ of the plant, the only exceptions being one lower leaf, certain tendrils, and a few centimeters of the undeveloped stem at the extreme top of the plant. This plant was inoculated October 1, 1894; wilt first appeared October 9 (in the pricked leaf) ; a trace of secondary wilt appeared October 14 and was well developed on October 15 in the 1st leaf above and the 1st below the inoculated leaf; on October 16 the plant was put into alcohol. When pricked the inoculated leaf was large and was near the apex of the vine. The infec- tious material came from vine No. 2. The foregoing conclusions respecting the etiology of this disease are drawn largely from the following: . 55.— Stages in life history of Diabrolica vittata, the striped cucumber-beetle: a, mature insect; b, larva, c, pupa; culpture on egg. a, b, c, enlarged; d, more enlarged; e, highly magnified. After Chittenden. This beetle is the principal disseminator uf Bacillus tracheiphilus. f9« ■'/'J'i MS." Fig. 55/ be the transmission of this bacillus. WILT OF CUCURBITS. 21 7 FIELD. HOTHOUSE. AND LABORATORY NOTES. Early Studies. The work of the first few months was thrown away, principally because I did not know how to proceed, my technique being defective. Up to November 24, 1893, I had isolated five or six organisms as follows but had not obtained infections with any, and was very much at sea after a great deal of hard work: (1.) A slow growing white organism — on streaks and in poured plates. In the latter there were very numerous colonies but small even after months. (2.) A very rapidly growing white organism. It runs ail over the plate in two or three days and is more or less dendritic. Spore-bearing — one oblong spore, central or at one end. (3.) A greenish organism which colors the agar. Much faster growing than 1 but slower than 2. (4.) An orange colored organism with crenate edges and in old specimens with radial fissures. It grows faster than 1 . (5.) A faint pinkish growth. (6.) A rapidly growing, wrinkled organism, color dirty-Isabella. The form 1 was undoubtedly the right organism, and probably the only one obtained from the interior of the plant. The others were undoubtedly intruders dragged in from the surface of the plant when I made my cross-sections, but I did not know it at the time. Sub- sequently I learned how to exclude outside organisms by the use of hot instruments. After that, labor was lightened, and inoculations with the right organism soon threw a flood of light over what had hitherto been an obscure subject. The only genuine infections I had obtained up to this date were on several squash leaves by direct transfer as described below. A few earlier supposed infections on cucumber leaves obtained with the organism No. 6, did not progress beyond areas which had this organism plastered on them, and were undoubtedly not true infections but only suffocation spots due to the overwhelming mass of material used, i.e., to defective methods of pro- cedure with some soft-rot organism, or potato bacillus. Squash Blight. The following observations and experiments were made at Hubbardston, Mich., in 1S93. (1.) September 2: Several 9-foot vines became infected several weeks ago in the main axis, naturally, and have lost a dozen to 15 basal leaves by the blight. All the foliage is dwarfed and yellowish. The foliage shows wilt in the daytime with partial recovery at night, the squash plants being more resistant than cucumbers. One of the striking signs is the formation of a short branch in every leaf axis and the development of a flower cluster — often a dozen buds. One 9-foot vine bore 43 branches and several hundred flower buds. This strongly suggests what occurs in orange blight in Florida. The uninjured vines are very green and thrifty, 12 to 14 feet long. I know these vines have been infected several weeks, from the general appearance, which has changed only slowly in the last 8 days; from the old dried up appearance of the blighted leaves in the center of the hill; and from the ease with which I get the milk-white bacterial ooze on the cut surface of basal branches near the main axis and also a foot or more away from it. This sticky ooze, appears abundantly on the cut surface over the fibro-vascular bundles in as short a time as 2 to 4 hours when placed in moist air with the bottom of the inch-long segments in water. No such exudate appears on the cut stems of the healthy plants. September 1 1 : The vines are still living and look as if they would live 2 weeks longer, but many leaves have died and all are yellow or dwarfed. (2.) Four squash flowers were inoculated September 1, from the white ooze. The germs were thrust down upon the nectary and the mouth of the flower was tied up. August 30, two squash- flowers were so infected and 2 days later one was examined and the whole nectary disk found dead and one uniform colony of germs. September 7: Germs grew in the nectary; the stems were not infected, or at least no secondary signs appeared during my stay. (3.) Bacteria from the cut ends of a squash-stem were pricked into two leaves (two vines), on August 30, using a sewing needle. Up to September 2, no blight. When pricked the blades of these leaves were 4 inches in diameter. Later: Typical blight appeared in both leaves in about eight days. (See under No. 10.) 2l8 BACTERIA IN RELATION TO PLANT DISEASES. (4.) August 30, a young squash-fruit was punctured in 40 places and many germs thrust in from several of the white colony-like bacterial beads on cut stems. The fruit oozed juice profusely. September 6: Squash still sound externally. On making sections through the flesh it looks water- soaked around the stabs for a breadth of nearly a millimeter. It is not rotten. Placed in moist air under a dish there is a moist sticky ooze from these water-soaked parts; not white; no ooze from other parts. (5.) September 1 : A leaf 7 inches across was wet on the upper surface over a square inch and a large white bacterial bead from the cut stem stirred up and pricked in with a needle. The leaf was then doubled together and inserted in a glass fruit jar (moist inside) and left 24 hours, the mouth being plugged with damp cotton wool (a defective method). September 2: Removed. No result where the germs were pricked in, but a sun-burn has appeared on the other half of the leaf. Septem- ber 11, 10 a. m. : Two small wilted spots have appeared on the area which was pricked ten days ago; 6 p.m.: Spots have enlarged, each being about three-quarters of an inch long (each side of a vein) and not over one-fourth inch wide. The petiole of this leaf is 6 inches long and all leaves on the axis above and below are healthy. (6.) September 1 : A vigorous terminal shoot was enclosed in a large glass jar (wet inside and plugged with damp cotton wool), two leaves 3 or 4 inches wide, being first wet and several white masses of the bacterial ooze stirred into the water and then pricked in with a needle. September 2 : Both infected leaves have blighted one-third to two-thirds and three others which touched them also show it. The blight includes the veins. These leaves are much younger and tenderer than in Experiment 5 (all probably due to sun-burn). September 1 1 : Solely sun-burn. (7.) Two ends of a vigorous squash vine were put into a wet glass can, two leaves on each being infected with bacteria brought from Washington. This growth was wrinkled, dirty Isabella color [the wrong organism]. The bacteria were teased up in water on the leaf and pricked in with a needle ; two more were infected in the same way from another block of potato in the same Petri dish. Mouth of can was closed with cotton and wet rags. All 4 leaves blighted in 24 hours, but probably all was due to sun-burn. September 1 1 : Solely sun-burn. No colonies appeared on cut ends of stem or petioles which were yellowing. (8.) Some germs from the potato cultures [wrinkled dirty organism] were inserted into 2 green tomatoes and into the stem very thoroughly with needles. September 1 1 : Tomatoes rotted slowly. Stem turned dark around puncture 1 mm. or more on outside and germs evidently infiltrated some distance into the tissue. My father afterwards picked and threw away the tomatoes for rotted, not knowing that I had inoculated them. (9.) Germs from the very gelatinous Isabella-colored, wrinkled, colony [wrong organism] on potato were rubbed up in well-water and pricked into the parenchyma and veins of four large turgid squash leaves, both sides, pretty thoroughly, with a cambric needle on September 7, 11 a. m. Leaves several feet from ends, and marked with twine. No results. (10.) September 7: The two turgid leaves (two different vines) which had bacteria pricked into them August 30, from the white ooze on cut squash-stems (see No. 3) are now badly wilted, while all the leaves to either side are turgid. These leaves are about 18 inches from the growing ends of the vines. I first detected the wilt yesterday morning (September 6), i. e., about 8 days after the infection. They looked all right for 5 or 6 days and I had abandoned the experiments as hopeless and did not look at them for 2 days. The wilt to-day (September 7,11a. m.) is very decided and I can attribute it to nothing but the slow growth of the inserted germs. I now know all of the other supposed infections (i. e., those obtained in moist air in 24 hours inside of glass jars) to be due in great part at least, and probably altogether, to sun scald. This was determined by getting the same results without use of germs. The cans rested on hot sand and the air became very hot inside and was saturated or nearly so with vapor of water.* September 10: The two squash leaves wilted completely but slowly, and are now crisp dry. The petioles are still green and turgid, but one seems a trifle flaccid at the extreme tip although not yet shrunken or discolored. This one was cut away and divided into 0.75-inch segments, and put on end in moist air. Two hours later there were plain indications of bacterial ooze from the cut bundles on some of the segments, and at 6 p. m., i. e., in 7 hours, all of the segments had each several beautifully distinct milk white bacterial beads resembling colonies. This was true even of the segments cut 3 or 4 inches below the blade of the leaf. This sets at rest all doubt regarding the possibility of inducing the disease by pricking in the bacteria taken fresh from the cut stems. September 1 1, 6 p. m. : Leaves to each side of these two are still perfectly healthy and no similar case of natural wilt at end of vines has appeared on any of the vines during the time I have been here. There is no shadow of doubt now as to what caused these leaves to blight. The striking thing is [was to me at that time] that the blight should have taken 8 days to develop. Only 'Defective technique. WILT OF CUCURBITS. 219 one additional vine has sickened naturally since my visit beginning August 22. This was all right until recently having only a few blighted leaves and withering petioles in the center of the hill, but now the whole of the big vine has wilted. There was no decided change in No. 10 until .September 14, p. m. (after a rain which occurred September 13). Then a whole leaf was found wilted suddenly. This was not the nearest leaf to the one which I had infected and cut away, but the next nearest, i. e., the one on same side of the stem. This vine had been examined at 8.30 a. m. and found all right. At 6.30 p. m. this whole leaf had wilted. All the other leaves on the stem were upright. Turgid sections were cut from the petiole of this wilted leaf and put into a moist place over night and next morning they bore the milk-white drops on the cut ends. Other portions were put into alcohol. [These petioles were examined in February, 1909, in thin sections under the microscope and bacteria were found in the vessels]. September 15, 10.30 a. m.: Five additional leaves were found wilted on this stem near the original source of infection, two toward the center of the hill and three beyond the original source of infection 1.5 feet. Now the nearest leaf had wilted, i. e., the one on opposite side of stem. (11.) vSeptember 7: Many beads of the milk-white bacillus, which oozed from the cut end of squash-stems were stirred up in water and three large squash-leaves were pricked with a needle, not the one used for No. 9. These were tied with torn rags to identify them; No. 9, by white cotton twine. On September 14, this plant showed 3 wilted leaves. Possible Carriers of Infection of B. tracheiphilus. The following insects, identified for me by Mr. E. A. Schwartz, were found on diseased cucumber vines in 1893 and suspected by me at that time of being agents in the distribution of the wilt: Diabrctica vittata Fabr.; Diabrotica 12-punctata Oliv. ; Strigodcrma pygnicruui Fabr. ; Chauliognathns marginatus Fabr. ; Epilachne borealis Kirby (lady beetle) ; Halticus uhleri Giard = H. minutus Uhler (Hemipter) ; Coptocycla guttulata (not especially devoted to the cucumber) . Inoculations of September i, 1894.* One leaf on each of eight plants of Cucumis sativus, growing in the hothouse, was inoculated with bacteria taken directly from white beads oozing on the cut end of cucumber stems and squash-stems, the foliage of which was flabby or dying from the effects of the wilt-disease. A sterile steel needle was touched to the ooze from a cucumber-stem and twenty or thirty punctures were made in the center of the lamina of each of four healthy leaves. The needle was then flamed and an equal number of pricks was made on the blades of as many more leaves using bacterial slime from the ooze on a diseased squash- stem. Plants 1 to 4 were inoculated from the cucumber; plants 5 to 8 from the squash. The bacterial ooze from the cut cucumber-stems was so gummy and viscid that it could be drawn out on the end of the needle in a delicate thread over a foot long. The bacterial masses did not dissolve readily in water, not even after several hours, nor with vigorous crushing and teasing. The temperature of the hot-house during the early part of the experiment was high, as the following records show: Sept. 5, maximum ioo° F. ; Sept. 7, at ih 501" p. m., 990 F. ; Sept. 8, at noon, 980; Sept. 9, maximum 1090; Sept. 10, at 2h p. m. 1040; Sept. 11 at 9h a. m., 720, noon 900; Sept. 13, at gh a.m., 720, at noon 900. (1.) The first signs appeared the fourth day after inoculation and first in the pricked area. The sixth day the whole blade of this leaf was affected and drooping, and its apex beginning to dry out. All the other leaves remained healthy. The ninth day the blade of the leaf to each side of the pricked one was wilted. On September 1 1 the leaf-blade next above and the one below the pricked one were dry-shriveled, and the second leaf above showed change of color and wilt on one margin at the base of the blade. Twenty-seven hours later one-half of the blade drooped and had changed to that peculiar green characteristic of leaves wilted by the immediate presence of the bacteria. The other side of the leaf was expanded and turgid. The disease progressed more slowly after the plant was brought into the cooler laboratory (vSeptember 10). On September 13 the whole of the pricked blade was wilted. On September 15 all of the leaves were wilted, 5 above and 1 below the inoculated leaf. The sixteenth day after inoculation the whole stem was dry-shriveled except the hypocotyl which was still turgid. "Those who wish to have etiologic proof without following all of the inoculations are advised to read only those of July 16, 1896, beginning on page 276. 2 2G BACTERIA IN RELATION TO PLANT DISEASES. (2.) The first signs appeared the fourth day after inoculation, the pricked area being the first part to show the wilt. The fifth day the inoculated leaf was cut away at the extreme base of the petiole. The progress of the wilt had been that shown in the sketch (fig. 56). The wilt did not extend more than 0.25 inch below the pricks and the petiole was 2.5 inches long. It was hoped, therefore, that the bacteria were removed with the leaf and that the plant as a whole, would remain free from the disease. Such was not the case. On the twelfth day one or two leaves showed a slight tendency to wilt. The next day the wilt was more decided although the soil had been watered copiously. On the fourteenth day 6 leaf-blades were wilted and drooping, 1 below the removed leaf and 5 above. The sixteenth day the plant was dissected and its vessels found to be gorged with bacteria. All the leaves were shriveled, but the stem was still green and turgid. The bacterial ooze from the cut stem was viscid and strung out in delicate cobwebby threads when touched, the same as the slime from which the plant was inoculated. The organism was cultivated out of the interior of this plant and gave rise to a long series of cultures of Bacillus tracheiphilus. .Stem, dried for herbarium. The cultures referred to were direct ones (Beef-broth tubes Nos. 1 and 2,September 17). These were feebly clouded on September 19, and looked alike. Each contained an actively motile bacillus. Tube 1 was used for inoculating plants 12 to 15; the organism was also cultivated out of it on slant agar, yielding typical colonies. Xine of these colonies when transferred to as many steamed potato cylinders yielded in 6 days typical gray-white, wet-shining, thin, sticky growths, scarcely distinguishable in color from the potato itself; 6 other cultures were made on September 17 direct from the interior of this vine — 3 slant agar and 3 potato. Two agars yielded nothing, the other, a single thin-edged, smooth, wet-shining, slow-growing, white colony (diam- eter 2 mm. after 15 days). The 3 potato tubes yielded typical growths of B. tracheiphilus, one of which was transferred to a tube of slant agar on September 23. (3.) The pricked area was the first part to show signs of the disease, which it did on the fourth day after the inoculation (noon or earlier). On the fifth day at 3 p. m., the inocu- lated leaf was cut away at the ex- treme base of the petiole (the petiole was preserved in alcohol along with that of No. 2, and portions of the wilted parenchyma of each leaf). The inoculated leaf resembled that of plant 2 very closely, about one- third of the apical portion being wilted. The wilt did not extend more than one-fourth inch below the pricks and the petiole was nearly twice as long as in No. 2. The disease was not cut out, however, by removal of the affected leaf. Like plant 2 the vine showed some signs of disease on the twelfth day and unmistakable ones the thirteenth day. < >n the fourteenth day 3 leaf-blades, 1 below and 2 above the removed leaf, were wilted and drooping. On the nine- teenth day after inoculation the plant was about 80 cm. long. All of the leaves to the extreme tip had wilted. The stem was still normal in color and turgid except for a slight shrinking just under the insertion of the inoculated leaf. The vine was now cut and examined in many places. The bundles were gorged with bacteria, nearly every vessel being full of what seemed under the micro- scope to be one kind of organism. Part of these bacteria were motile. As in other vines of this series, the number of motile rods increased toward the tip of the plant, i. c, at more and more *Fig. 56 I.cal" of Cue 11 111 1 s sati *is (No 2) inoculated with B tracheiphilus by needle-pricks and shaded to show progress of wilt. First signs on fourth day (Sept. 5): 1, 6 a. m. ; 2, noon; 3, 4p. m.; f, Sepl 6,3 p.m. fin this date 1 In- leaf was cut awaj at its base, bul the bacteria had already passed down \ essels of th( leaf stalk and had entered m as shown by subsequent events (petiole 2.5 inches long I. run- cultures yielding a long series of successful inoculations were afterward obtained from interior of the stem of this plant. Drawn by tile writer. WILT OF CUCURBITS. 221 remote distances from the point of inoculation. Examined in drops of sterile water many of those rods taken from near the tip were actively motile, while those taken lower down, i. e., from vessels clogged solid with the bacteria, were not motile at all or only doubtfully so. The final condition of the bacteria in the vessels of the plant appeared to be a zoogloeae stage. The organism was cul- tivated from the interior of this plant at various levels and found to be Bacillus tracheipkUus. In making these transfers the stems were shortened with a hot knife and the end dug into with a flamed needle. On September 20, under these conditions, the following transfers into tubes of sterile media were made from the interior of this plant : (1) Four inches from the tip. Beef broth; (2) Do., Cucumber broth; (3) Six inches from the tip, Potato broth; (4) Do., Peptonized beef broth; (5) Eight inches from the tip, Beef broth; (6) Do., Potato broth; (7) Eleven inches from the tip, Beef broth; (8) Do., Cucumber broth; (9) Fifteen inches from the tip Potato broth ; (10) Do., Cucumber broth. One tube remained sterile (No. 2) ; 2 were contaminated (No. 8 with a pink organism and No. 10 with a white organism forming a pure white, wrinkled, fragile pellicle; 7 yielded moderately clouded cultures exactly alike and presumably all of them pure cultures of Bacillus tracheiphtlus. The behavior of the organism in 5 of these cultures (Nos. 1, 3, 5, 7, and 9 was tested further by transfers to cylinders of sterile cooked potato. On this medium each one developed a thin, smooth, wet-shining gray- white, sticky slime, scarcely distinguishable in color from the surface of the potato itself, and perfectly characteristic as was afterwards learned, of the behavior of this organism on steamed potato. (4.) The first sign appeared the fourth day after inoculation. The area in the vicinity of the needle-pricks was the first part to become flabby and discolored. Within a period of 4 hours, on the afternoon of September 5, the spot increased noticeably, being at least one-third larger at 4 o'clock than at noon. By the sixth day the wilt had extended on one side to the extreme base of the leaf- blade. The whole leaf drooped and two-thirds of it had changed color. The rest of the vine was normal except the tip of one leaf which was bruised in repotting. By the eighth day the whole leaf- blade had shriveled. On the eleventh day the blade of the leaf next above and of the leaf next below the inoculated one began to droop slightly and on the twelfth day they were entirely shriveled, while the next one above showed a tendency to droop. The thirteenth day the plant was photographed (see plate 16, fig. 1 ). There were then 3 shriveled leaf-blades and 3 freshly-wilted leaves further up the stem. At the top were three turgid leaves and at the bottom one. The fifteenth day the plant was wilted throughout except the stem and the base of some of the petioles which looked normal. There was no external indication of the cause of the disease. The stem was now cut open and examined under the microscope: 6 inches below the inoculated leaf every vessel of every bundle contained bacteria. Most of the vessels were gorged and large cavities had formed in the primary vessel- parenchyma of three bundles. The bacteria here were not clearly motile. Six inches farther down the vessels were still gorged but some of the bacteria were plainly motile. The tissue was less broken down. The bacteria also occurred an inch from the tip of the stem, but less abundantly. The vessels were full, however. The bacteria were also found in the petioles of the leaves where they were abundant. The disorganization of the bundles in this plant had proceeded further than in No. 8, examined the fourteenth day. Fourteen cultures were made from the interior of this plant as follows, the stem being cut with a hot knife, and its interior dug into deeply with a flamed, steel-needle and slime removed from the cavity : (1) 3 gelatin rolls; (2) 5 gelatin stabs; (3) 2 beef-broth tubes (1 peptonized); (4) 3 tubes of agar (stab cul- tures). Two of the gelatin rolls remained sterile, one contained a mixed growth consisting of two small, white colonies, two yellow ones, and a mold spore. Two of the gelatin stabs appeared to be pure cultures, the others were contami- nated. The two beef-broths clouded typically, but on microscopic examination they were found to contain round bodies as well as rods. The agar stabs all yielded typical, thin edged, gray-white, wet-shining surface growths. The round bodies in the bouillon cultures may not have been contaminations as they afterwards appeared in tube 1, September 17, made from vine 2. Also because a loop from one of these broths yielded a thin gray-white, wet- shining, typical culture when transferred to potato.* (5 to 7.) One leaf-blade on each plant was inoculated with bacteria from a diseased squash-stem. No result. The inoculated leaf did not wilt. Possibly the bacilli dried out in the pricks before they got a start. (8.) One leaf-blade was inoculated with bacteria from a diseased squash-stem. The first signs appeared the fourth day after inoculation. The area in the vicinity of the needle-punctures was at *In July, 1909, from a primary natural infection on a cucumber (petiole), a non-infectious white coccus was plated out, B. tracheipkUus being dead. This coccus form is viscid and closely resembles B. tracheiphtlus on potato, but its growth on agar, while smooth, is a denser chalkier white and it reddens litmus milk. It was stained by amyl Gram. It did not grow in Fermi. 222 BACTERIA IN RELATION TO PLANT DISEASES. that time flabby and somewhat discolored. Some hours later there was a distinct increase in the size of the wilted spot. The sixth day the whole leaf-blade was affected. The rest of the plant was normal. The eighth day the leaf-blade was shriveled. The ninth day the blade of i leaf below and of 2 above the inoculated leaf showed decided wilt. The tenth day the blade of the first leaf below had changed to a lighter green in places (same color as the primary wilt). The second below was also drooping, two bright green leaves above had collapsed beyond recovery, and two more further up were beginning to droop. The eleventh day the plant was badly affected. Two leaves below the infected one and 3 leaves above it had wilted. By the twelfth day the plant had developed an advanced and very typical case of the wilt. Two leaves below the pricked one (there were no more leaves on this part of the stem) and 4 above it were shriveled beyond recovery and becoming dry. The one next above drooped to a slight extent and the next one very slightly. The thirteenth day the plant was photographed, all the leaves being wilted at that time (see plate 16, fig. 2). The fourteenth day the plant was cut to pieces and examined. Many vessels were clogged full of bacteria, a portion of which when examined in water were seen to be distinctly motile. Some had a darting movement half across the vessel. Most of the vessels in all the bundles were gorged with the bacteria. The organism was very sticky to the touch and would string out when touched with the platinum wire. Fig. 57 was drawn from a smeared cover-glass preparation of this sticky ooze stained with carbol fuchsin. Of 6 gelatin tubes (roll-cultures) inoculated from this plant, 2 showed no growth, 2 were con- taminated by a greenish liquefying organism, 1 by a white liquefying organism dubbed the "angleworm," and 1 by a cadmium orange organism. Of 2 agar poured plates made from this stem, 1 contained nothing on September 18 and 1 about 50 colonies of a contamination — a gray-white, crenate-margined colony, wet-shining at the edge but the rest of the surface covered by a flour-like coating. This latter plate was made in an unusual way, i. e., by crushing a segment of the clean stem (not externally sterilized) in a tube of sterile water and transferring a loop of this fluid to the agar. Three hypotheses occur as an explanation of these failures: (1) The parasite was dead in the par- ticular part from wThich inoculations were made ; (2) the organism was so viscid that it did not wash off the needle or dissolve readily in the melted agar; (3) the agar was too hot when inoculated, i. c, exerted a killing influence [my technique was still imperfect]. Inoculations of September 13, 1S94. Plants 9 and 10 were inoculated September 13, 1894, from tube 1, September 8 (a potato culture from a cucumber fruit). The slime was flat, gray-white, wet-shining, in small patches on the center of the F.g. 57.* potato and growing slowly. It was very sticky and spun out in a fine thread when touched with a needle. No record of results was made or if made, it has been lost. Inoculations of September 19, 1894. These inoculations were made on healthy but small hothouse vines of Cucumis sativus by means of needle-pricks. The infectious material was a beef-broth culture 2 days old (No. 1, September 17) derived from vine 2 and containing actively motile rod-shaped bacteria. The needle-pricks were confined to a small part of one leaf-blade of each vine. (12.) The signs were so long delayed that no results were anticipated. Up to October 8 (p. m.) the plant was healthy in appearance, but by 9 a. m., October 9 (the twentieth day), the inoculated leaf-blade had changed color and begun to wilt. The blade and also the tip of the petiole were drooping at 2 p. m. The leaves at each side of the inoculated one were turgid. The twenty-first day the pricked leaf had begun to dry and hang down, its petiole being flabby. No other leaves showed any signs. The twenty-second day the next leaf below and the first one above the inoculated leaf were wilted and drooping— more at 1 p. m., than at 9 a. m. Three days later 7 additional leaves had wilted, all of them above the pricked one, making a total of 10 wilted leaves — i. e., the one pricked, 1 below and 8 above. The twenty-seventh day the vine was brought into the laboratory. Segments, including a small fruit, were preserved in alcohol. (13.) One leaf-blade was pricked. The plant subsequently dried up but not as a result of the inoculation. (14.) No record. *Fig. 57. — Cover-glass preparation of B. tracheiphilus stained with carbol fuchsin. Smear made Sept. 15, 1894, with white ooze from cut stem of vine No. 8, which was inoculated Sept 1 (plate 16, fig. 2). Organism motile in vessels. x 1000. PLANT BACTERIA, VOL. 2. PLATE 16. Wilt of Cucurbits. Hothouse-cucumbers inoculated with Bacillus tracheiphilm Sept. 1, 1894, by direct infection from cut surface of diseased stems brought in from a field, bacteria being introduced by needle-pricks on blade of one leaf. (i) Plant No. 4. inoculated with white sticky ooze from stem of cucumber. Photographed thirteenth day after inoculation — 3 blades shriveled, 3 wilting, and 4 normal. (2) Plant No. 8. inoculated with white sticky ooze from a squash-stem. Photographed on thirteenth day when all the leaves had shriveled. Photographs about ' natural size. WILT OF CUCURBITS. 223 (15.) The pricked leaf was the first to show signs of the disease. They were noted the eleventh day after inoculation, but as the droop did not seem to proceed from any particular spot I was in doubt as to its cause. The thirteenth day the leaf above and the one below the inoculated leaf showed wilt. The fourteenth day they began to shrivel, the whole blade of the inoculated leaf was dry- shriveled, three additional leaves farther up the stem also showed a decided droop, and a fourth one, still higher up, a slight flabbiness. The rest of the leaves above and below were turgid and showed no sign of the wilt. The disease was moving up faster than down, as in some cases previously recorded. The following day (October 4) three additional leaves nearer the tip were wilted and one more toward the base making eleven in all, to wit: eight above the inoculated leaf and two immediately below it. The remaining basal leaf and the four leaves at the tip of the vine were still turgid. Two of the four noted as having wilted the previous day (the two nearest the point of infection) were then shriveling. The plant was now brought into the laboratory. The seventeenth day the wilt showed on the lowest leaf. All the leaves farther up as well as the stem in places had begun to shrivel. When segments of the stem were examined microscopically, the vessels were found to be full of the bacteria, which varied in size noticeably and looked much larger than usual (involution forms?). In the primary vessel parenchyma were many destructive lesions. The bacillus was also found in a small, green fruit, hanging midway on the stem and looking sound externally. Here it was confined to the bundles in the outer ring from which it slowly oozed on cross-section. The next day (October 7) the cut surface over the affected bundles was covered with a milky and very viscid bacteria] slime which strung out on the tip of a needle a distance of 40 cm. (15.75 inches). The milky beads which oozed from the bundles of the cut fruit yielded a pure culture of B. tracheiphilus. On October 17 a potato cylinder inoculated October 7 was covered, except its edges, with a thin, gray- Fig. 58.* white, wet-shining layer which followed the irregularities of the surface of the steamed potato but was otherwise smooth, and was almost exactly the color of the potato but easily distinguished by its wet-shining surface. Thirteen days later there was no change in the appearance of this culture, except that it had spread over more of the potato. For the condition of this vine on October 4, see accompanying diagram (fig. 58). Inoculations of October i, 1894. Two sets of inoculations were made on Cucumis sativus in the hothouse. One plant (18) was infected with a gray-white, wet-shining organism from a potato culture of September 23 made from tube 1, September 17, which was inoculated from vine 2. The other two (16 and 17) were inoculated with a bacillus (examined in hanging drop), forming a cloudy growth in a fermentation tube of saccharose bouillon. This saccharose bouillon was inocu- *Fig. 58. — Diagram showing condition of cucumber-vine No. 15 on Oct. 4, 1894; a, leaf inoculated on blade Sept. 19 by needle-pricks. First signs of disease appeared on Sept. 30 in pricked leaf: (1) Position of leaves which showed secondary wilt on Oct. 2; (2) position of leaves which wilted on Oct. 3; (3) position of leaves which showed first signs of wilt on Oct. 4. Z and M, leaveswhich wereturgid on Oct. 4. Z, which was the last leaf to succumb, drooped on Oct. 6. F , a small fruit from the interior of which B. tracheiphilus was obtained in pure culture. Y, lower nodes from which leaves disappeared naturally owing to small size of pot. 224 BACTERIA IN RELATION TO PLANT DISEASES. lated September 16 from a smooth, wet-shining, gray- white culture on potato (No. i, Septem- ber 8). This potato culture was derived from the interior of a cucumber fruit. On Septem- ber 13 it was described as forming on the potato slow-growing, flat, gray-white wet-shining masses, which spun out in a fine thread when touched with a needle. The pricks were made with a steel needle on one leaf-blade of each plant, except No. 17, which was pricked on the center of the lamina of 2 leaves. (16.) The first signs appeared the fifth day after inoculation, at noon, and first in the infected leaf. By 4 o'clock of the same day the disease had spread considerably and occupied about half of the leaf, forming a wedge-shaped area from the pricked portion outward. The rest of the foliage was normal. The eighth day the blade of the infected leaf was wholly dried up except a small area(iX icm.) at the tip of the petiole which was yet green but not turgid. The internode below was 12 cm. long. The next node above had lost its leaf long since (killed by a tobacco-water spray). The distance from the infected leaf to the second node above, which bore a good leaf, was 18 cm. Both of these nearest leaves as well as more remote ones were perfectly normal on October 9 (the eight day) as was also the upright petiole of the pricked leaf. There was no further visible change until October 12. Then the first leaf below the pricked one drooped on one side. The first leaf above was still turgid. There was no change on October 13, 14, or 15. The last record of this plant was made on October 17. At that time the first leaf above the infected one was also drooping but all the others were normal. (17.) The pricked leaves were well toward the extremity of the vine and were separated only by one internode. The leaves were normal at noon on the sixth day, but the morning of the seventh day both blades were drooping slightly and showed change of color. At 4 p. m. these signs were more decided and involved the whole leaf-blade. The following morning (October 9) the blades of the infected leaves were collapsed and had begun to shrivel. At 2 p. m. of the same day the nearest leaf to either side of the infected ones (leaves which were perfectly turgid in the morning) began to show unmistakable signs of the disease, i. c, one blade was drooping decidedly at the tip and the other on one of the side lobes. The other leaves were perfectly healthy, and the petioles of the infected leaves were still turgid. The temperature on this day was 6o° to 700 F. On October 10, at 9 a. m., the leaf next above the upper pricked one was drooping on both sides as well as at the tip and the next two above were flabby. The leaf next below the lower pricked one was drooping at the tip, and on either side. The second leaf below was still turgid, but by noon of the same day it was drooping. The tenth day all the leaves above were wilted and also the four next below the lowest pricked leaf, two additional ones having drooped that morning. The eleventh day the fifth leaf below the inocu- lated ones was drooping. The three below this were still turgid and were the only sound leaves remaining on the vine. The following day the fifth leaf down was wholly collapsed. The rest of the wilted leaves had shriveled but were not yet dry-brittle. The three basal leaves were still normal. The vine was now brought into the laboratory. On October 14 the stem had begun to shrivel in places and the uppermost of the three basal leaves had drooped, leaving only two sound leaves on the vine. Two days later all the leaves were wilted and the upper part of the stem was shriveled. (18.) The blade of a large healthy leaf near the tip of the vine was pricked. On the afternoon of the seventh day the inoculated leaf was normal. The morning of the eight day a part of the pricked leaf-blade had changed color and over half of it was wilted. By 2 p. m. of the same day it was wholly wilted with exception of a square centimeter where the blade joined the petiole. The petiole was turgid as was also the leaf to either side. The ninth day the pricked leaf was wholly soft-flabby, drooping and beginning to dry. All the other leaves were turgid and remained so for some days. The thirteenth day the leaf next above and the one next below the inoculated leaf were slightly flabby. Bacteria were now present in the vessels of the fruit. On October 15 the leaf to each side of the inoculated one showed decided wilt. Two-thirds of each blade hung flaccid. The rest of the vine was normal, including all of the petioles. The diseased leaves were now removed and put into alcohol. On the sixteenth day none of the leaves remaining on the vine showed any signs of wilt. Sections from all parts were removed and put into alcohol (fig. 59). Inoculations of Oct. 25, 1894 (noon). Three young vines of Cucumis sativus, were sprayed thoroughly with a mixture which was three-fourths sterile water and one-fourth a potato-broth-culture of Bacillus trachei- philus 2 days old (tube 8, October 23). These plants were placed under bell-jars in order that they might be kept free from aphides which I suspected might be the means of intro- ducing the bacteria into the plant. Upon a fourth cucumber vine which was sprayed in like manner, numerous aphides (Aphis gossypii) were colonized, and the vine placed under a WILT OF CUCURBITS. 225 bell-jar. The first three vines were used as checks on the behavior of the fourth. The air under the bell-jars was quite moist and at 4 p.m. water stood in tiny beads on the margin of the leaves. At the end of 24 hours the under surface of the sprayed leaves was still wet in places especially that of the leaf on which the aphides were colonized. Some of the latter had migrated to other leaves. All were sucking the plant juices and for fear of mechanical injury I brushed off and destroyed most of them. None were observed on the check vines. The bell-jars were removed and the plants exposed to the air for half an hour to dry off a little and then the jars were put back. This was done frequently during the experiment. The fourth day the vines were still healthy and the checks were free from aphides. Tube S, October 23, was inoculated from a very sticky potato culture (tube 8, October 17), which was inoculated from a single , small , white colony on a slant agar culture streaked September 2 7 from tube 1 , September 1 7 , which was inoculated from the interior of plant No. 2. (20.) Cucumber (check). Plant about 5 inches high with two well-developed leaves and one more coming, also two green cotyledons. The bacterial fluid was sprayed on the under sur- face of the two largest leaves and the plant was then put back under the bell-jar. The eight day after spraying, this vine was still healthy. It had grown an inch or two since October 25. It was still free from ants and aphides. Two days later it was healthy and growing rapidly. The twenty- third day the bell-jar was removed and not re- placed as the plant was beginning to be spindling although no trace of the disease had appeared. The thirty-fourth day the vine was still free from the disease but had remained spindling since the removal of the bell-jar. By the fifty-first day the plant had lost all its leaves and the tip of the stem had wilted. It was not much over 1 foot high, having never recovered from the stunting due to keeping it under the bell-jar. Thin sections were cut and a microscopic examination made, but no bacteria were found in the vessels. (21.) Cucumber (check). This plant was the same size as No. 20. It had two green cotyledons, bore one well-developed leaf, which was sprayed on its under surface, one twisted deformed leaf, and two undeveloped leaves. After spraying it was placed at once under the bell-jar. The eighth day this vine resembled the preceding in all particulars. It remained healthy and grew rapidly for a time but on the final removal of the bell- jar (the twenty-third day) it was beginning to be spindling although free from the disease. The thirty- fourth day it was still free from wilt but had remained spindling. The forty-sixth day it was brought into the laboratory and examined for the presence of the bacillus in its tissues. It had never recovered from the stunting due to keeping it under the bell-jar. Since the removal of the latter it had also suffered to some extent from mildew, from aphides, and on two or three occasions from insufficient moisture. It was not over 12 inches high. For the 3 weeks preceding it had been losing its foliage Fig. 59.* *Fig. 59. — Cucumber No. 18, inoculated at x by needle-pricks with a pure culture of Bacillus tracheiphilus on Oct- 1, 1894. Numbered parts were removed and fixed in alcohol Oct. 16. They were subsequently embedded in paraffin, sectioned and stained for presence of bacteria, which were found in vessels at all points marked +, and not at those marked — . They occurred in greater or less numbers according to distance from inoculated leaf, or from main axis. Exclusive of wilt there were no surface indications of disease. About one-fourth natural size. 2 26 BACTERIA IN RELATION TO PLANT DISEASES. gradually and on this date (December 10) its last tiny leaf was found shriveled and also the upper 2 or 3 inches of the stem. Thin sections from the hypocotyl, and first, third, fourth, and fifth internodes, the last one of which was flabby, were examined under the microscope but there was not a trace of the bacillus. (22.) Cucumber (check). Plant about 6 inches high, with two big leaves, two small immature leaves, and two green cotyledons. The three largest leaves were sprayed on the under surface and the bell-jar at once placed over the plant. The eighth day the vine was healthy and free from ants and aphides. It had grown about 3 inches since inoculation. Two days later it was nearly as tall again as when put under the bell-jar. The temperature in the hothouse that day (November 4th) was 83° F. The twenty-third day the bell-jar was removed. The vine was somewhat spindling but free from the disease. It was very warm in the hothouse when the bell-jar was removed, and part of the leaves wilted, the rest soon following in spite of careful watering. As soon as the wilting became noticeable the vine was repotted (November 18th) in a 4-inch pot and this probably hastened its collapse. On the twenty-ninth day the leaves were all shriveled and the stem was beginning to shrivel also. The latter was now cut at various heights, where it was still partly turgid, and exam- ined microscopically. The vessels were found to be full of bacilli which varied greatly in size. (23.) Cucumber (colony of aphides). Plant about 6 inches high, bearing two good leaves (blades about 2x2 inches), and two undeveloped leaves. The under surface of one of the larger leaves on which 20 aphides had been colonized October 24 was sprayed with the bacterial fluid until it was wet with mist and tiny drops. Fifty aphides, from a neighboring watermelon plant, were then placed on the wet surface, along with those already there, and the bell-jar replaced. Four hours later many tiny drops remained on the sprayed surface, one-fifth to one-sixth of its surface was still covered by these drops. Two of the aphides had moved to the upper (dry) surface, and 5 had crawled to the under surface of another leaf. The eighth day this vine was less healthy than the others. It had made very little growth owing to the aphides which were clustered on the terminal portion, causing it to be twisted and stunted. After removing the jar, to clean off the aphides and give fresh air, the plant had a drooping aspect. The sprayed leaf especially and one other, lacked turgidity and on close inspection I found that large areas of the leaves were pale green. Two days later, however, there was no sign of the flabbiness. The top was badly bent, twisted and stunted by the punctures of the aphides and the plant had elongated not more than an inch. The leaves, however, were turgid. The lack of turgor the eighth day was due probably to leaving the plant uncovered too long, the atmosphere under the bell-jar being nearly or quite saturated, while that of the hothouse was com- paratively dry. The twenty-third day the bell-jar was removed permanently. The plant was spindling. Six days later (November 23), the vine was in a very bad condition although it was im- possible to tell by inspection whether or not this was due to the multiplication of the bacteria in its vascular system. Three days later all the leaves on the growing tip had dried up and nothing was living and turgid except the pale green stem. The final slow death of the leaves was probably attribu- table to the direct effect of the aphides as the plant was badly dwarfed by their presence and had made but little growth. No signs of the bacterial wilt appeared, and on examination of thin sections I could find no bacilli in the vessels. They were entirely free from obstructions. Remarks. — The manner of performing this experiment seemed to give every oppor- tunity for infection through the ordinary stomata, since the under surface of the sprayed leaves remained wet over night. The results, however, do not bear out this hypothesis. This experiment also lends no support to the hypothesis that the bacterial wilt of Cucurbits may be spread by aphides. The entire lower surface of one leaf was sprayed thoroughly (wetted) with an infectious culture and the aphides which were colonized on it soon made numerous punctures but the plant did not become affected. Only one of the four plants contracted the disease and that was a check. The bacteria probably entered the latter through some fissure or other injury rather than through the stomata. Otherwise it is difficult to explain the immunity of the other three vines which were equally exposed to stomatal infection. It is, however, not known positively that infection never takes place by way of the stomata. This remains to be determined. Further experiments should be made also with aphides. Inoculations of October 25, 1894. Ten old vines of ( 'ucitmis sativus, were inoculated in the hothouse from a potato-broth- culture (tube 8, October 23) of the gray-white, wet-shining, motile Bacillus tracheiphilus. The pricks were made with a sharp steel needle, in most instances on a single leaf-blade, but WILT OF CUCURBITS. 227 once on a fruit. The record shows that in each case numerous delicate punctures were made, 40 in one case and probably from 40 to 50 in the others. The infectious material consisted of one part of the feebly clouded broth mixed with three parts of sterile water, viz., a part of the same mixture that was sprayed on Nos. 20 to 23, just described. The day temperature during the first 10 days of this experiment generally ranged from 650 to 850 F. but once it was as high as 900 (November 4, noon) and once as low as 50° (early morning of November 5). (24.) The middle basal part of the blade of the fifth leaf from the tip was pricked. The vine showed no signs of the wilt until the morning of the eighth day. Then there was a wilted area about 1.5 cm. in diameter in the pricked portion. The rest of the leaf was turgid. By noon a vshaped area extending from the pricks to the end of the leaf had wilted and by night the whole tip and one side had changed color and hung flaccid. The following day the entire blade of the pricked leaf had collapsed although its petiole was turgid. No other leaves showed signs at that time. The tenth day after inoculation the petiole of the pricked leaf was still turgid, as were also the leaves to either side of the pricked one, and a pistillate blossom in the axil of the inoculated leaf. The morning of the 1 2th day the extreme tip of the petiole of the pricked leaf was slightly flabby. By late afternoon of the same day the blade of the first leaf above the pricked one had wholly collapsed, except about 1.5 cm. around the apex of the petiole. There was no further change until the late afternoon of the following day. Then the blade of the first leaf below the pricked one was flabby and hanging down. The morning of the fourteenth day the blade of the pricked leaf was brown, the petiole was still green but flabby at the extreme tip. The small fruit in the axil was green and looked healthy. The second leaf down had lost its turgidity on one side. The second leaf above, which by reason of longer internodes was three inches farther from the pricked leaf than the second below, was still turgid. The vine bent in such a way that the second leaf down was 3 inches higher than the pricked leaf or any above it. The stem was green and turgid. At 3 p.m. of the fifteenth day the third leaf down was found to be flabby. No further records were made until 4 p.m. the nineteenth day after inoculation. It was then found that the disease had proceeded gradually farther and farther down the stem, wilting leaf after leaf until the seventh down had become flabby. In the morning of the same day this leaf was turgid. The morning of the twenty-third day the vine was brought into the laboratory and examined microscopically. The stem contained enormous numbers of bacteria, which did not seem to be confined to the bundles but to be out in the parenchyma to some extent as well. Part of the stem was shriveled but where the examination was made (4 to 6 inches below the pricked leaf) it was still green and turgid. All the foliage had been shriveled for some days. The upper part of the stem had also shriveled. The organism was cultivated out on November 17 into 4 tubes of potato-broth. On November 21, these four tubes were all alike, each being faintly clouded with rolling clouds on shaking. Potato-cultures were made from these tubes. On November 26, three of these tubes contained pure cultures of B.tracheiphilus; the fourth was contaminated by a yeast. The pure cultures on potato were thin, gray-white, wet-shining, and smooth, except that the layer was not thick enough to hide the coarsest undulations on the surface of the potato. The cul- tures were viscid to varying degrees (1 mm. to 6 cm.). The most viscid one (No. 2) contained the largest number of actively motile bacilli (one-half motile). The color of the slime was almost exactly that of the steamed potato itself. (25.) The middle basal part of the blade of the fifth leaf from the tip was pricked many times. On the morning of the sixth day there were no signs. At 1 p.m. there was no change. At 2 p.m. there was a very faint change of color in the pricked area. This was scarcely noticeable. At 2 :30 p.m. a piece of the leaf extending from the pricks to the apex had changed to a light green and wilted. At 5 p.m. a sketch was made showing the wilted part shaded. This was now a decidedly lighter green than the rest of the leaf (the change of color being much more noticeable than at 2 p.m.) and there was a slight flabbiness at the tip. The leaf, with the exception of that part shaded in the draw- ing, was turgid. The seventh day at noon the change in color (to light green) and the wilt were very typical, extending from the middle pricked portion of the leaf to its apex. The apex of the leaf was flabby and hung down. By 3 p.m. of the following day the whole blade of the pricked leaf had wilted. The following noon (the ninth day after inoculation) the nearest leaf each way from the pricked one had lost its turgor and was drooping a little. The petiole of the pricked leaf was turgid. At noon of the tenth day after inoculation the leaves to either side of the pricked one had wholly collapsed and the second leaf up was losing turgor. The second below was fully turgid. The petiole of the pricked leaf was still green but slightly flabby at the tip. The blade was becoming brownish. The morning of the eleventh day the second leaf above the pricked one was more flabby than on the preceding day and the third above had lost its turgor. The second below was flabby and the fourth below was 228 BACTERIA IN RELATION TO PLANT DISEASES. turgid. (The third down was wanting except the base of the petiole.) The next morning all the tiny leaves at the tip (bevond the second leaf up) had collapsed. By 5 p.m. of the same day the wilted leaves had begun to shrivel. The fourth down was still turgid. The fourteenth day the fourth leaf had lost most of its turgor. The petiole of the pricked leaf was still green but it was flabby nearly to the base. The blades of the first and second leaves below had shriveled, also those of all the leaves above the pricked one. The upper part of the vine was now removed for examination and cultures. The vessels were found to be full of the bacillus which strung out in fine gummy threads from the cut surface of the stem when a needle-tip was touched to it and withdrawn. The bacteria were exceedinglv abundant and the inner tissues were considerably broken down. The organism was cul- tivated from this portion of the vine at different heights, inoculations being made from the stem into potato-broth from which a pure culture of Bacillus trackeiphilus was subsequently obtained on steamed potatoes. The fifteenth day the fifth leaf down was flabby. The twenty-ninth day this plant was removed together with the other old cucumber vines, to make room for squashes. Dry material was saved from this vine for the herbarium. A futile search was made in it for spores of the bacillus. (26.) The fifth leaf from the tip was pricked many times in one of the side lobes. At 10 a.m., October 31 there were no signs but by ih 30'" p.m. of the same day the pricked lobe had wilted. The first signs appeared, therefore, in this case at the end of the sixth day, the inoculations having been made in the afternoon. By noon of the seventh day the wilt and change of color had made marked progress in the pricked lobe and that portion of the latter in which the wilt first appeared had dried out. The temperature in the hothouse when this observation was made was 8o° F. The following day (3 to 4 p.m.) the whole leaf-blade was flabby. The pricked portion was dry-wrinkled, the petiole turgid. There was no further change until noon of the tenth day. Then the petiole of the pricked leaf was slightly flabby at the apex but still green. The blade was turning brownish on the pricked side. The nearest leaf to each side was turgid. At 4 p.m. the following day, the first leaf below had fully collapsed although turgid at 10 a.m. The second leaf above was still turgid. (The first leaf above was wanting, only the base of the petiole remaining.) The upper part of the vine hung down in such a way that the lower (collapsed) leaf was uppermost. The morning of the twelfth day the petiole of the first leaf below the pricked one was flabby. An observation made at 5 p.m. showed no further change. The morning of the fourteenth day after inoculation the blade of the pricked leaf was brown and dry-shriveled throughout. The petiole was still green and it was flabby only at the tip. The second, third, and fourth leaf down were flabby as was also the petiole of the second leaf. The blades of the fifth and sixth leaves below (all that remained) were flabby and shriveled including the petioles. This latter seemed anomalous. The petiole of the first leaf below was flabby and shriveled to its base although still green. It was in a much worse condition than the pricked leaf. Here we probably have to take into account the smallness of the pot and the age of the vine, the lower leaves being weaker than the others. The twenty-ninth clay the plant was pulled up to make room for squash vines. A portion of it was saved dry for the herbarium. Unavailing search was made in it for spores of the bacillus. (27.) The seventh leaf from the tip was pricked many times in the center of the lamina. There werenosignsuntilthemorniugof October3i (sfdays). Then the pricked leaf was wilted in a V-shaped area opening outward from the pricks to the apex of the leaf. In the afternoon the condition was that shown in fig. 60. The apex was drooping. The rest of the leaf was normal. The following noon, *Fig. 60. — Bacteria] wilt at 4 p. m., Oct. 31, on an inoculated leaf of cucumber plant No. 27. For condition 20 hours later, see fig. 61. Leaf pricked Oct. 25, 1904. First signs of wilt (darker shaded part) morning of Oct. 31. Drawn by Theodore Holm. WILT OF CUCURBITS. 229 the whole leaf-blade was affected with the exception of about 1 cm. at the apex of the petiole (fig. 61). Over two-thirds of the blade was dry-wrinkling. The leaf succumbed very quickly. The eighth day after inoculation the pricked leaf-blade was almost wholly dry-wrinkled but the petiole was still turgid. The next noon the nearest leaf each way which showed only the faintest trace of want of turgor at 9 a. m. was flabby and drooping, especially the one below. The petiole of the pricked leaf was still turgid. The tenth day the petiole of the pricked leaf was still green but it was flabby half- way to the base. The shriveled blade was becoming brownish. The first leaf each way was more collapsed than on the preceding day, expecially the lower one. The second leaves each way were still turgid. The eleventh day (10 a. m.) the second leaf below was fully flabby and drooping. The second above then showed only a very slight lack of turgor at the apex of the blade, but at 4 p. m. it was quite flabby and drooped. The twelfth day the petiole of the pricked leaf was green but flabby half-way to the base, the same as on the tenth day. The blades of the first, second and third leaf up were wholly collapsed and also the petioles of the first and second. The same was true of the blades of the first and second leaves down and the upper third of the petiole in the first leaf down. The third leaf below was still turgid. At 5 p. 111. the blades of the first and second leaves below and the first, second, and third above were shriveling. The fourteenth day the blade of the pricked leaf was brown-shriveled. The petiole was still green but now flabby nearly to the base and shriveled half- way down. The third leaf down had begun to lose turgor. The petiole of the first and second leaf up were more flabby than that of the pricked leaf. The twenty-ninth day this vine together with the other old cucumber-plants, was pulled out to make room for squashes. Dry material was saved for the herbarium. Search was made in it for spores of the bacillus, but none were found. (28.) The fourth leaf from the tip was selected for inoculation; the pricks were made near the apex of the leaf and were numerous. The first signs appeared on the morning of Octo- ber 31 (534 days) and first in the pricked area. The pricked portion (the leaf had been purposely inoculated near the apex) was wilted and had become a paler green. Upward the wilting had extended to the apex of the leaf, 1.5 cm. beyond the pricked area, while downward it had extended only 3 mm. beyond the pricked area. The rest of the leaf was sound and turgid. The middle portion of the pricked area (earliest wilt) was brown. At 5 p. m. the wilted area had widened 2 mm. or more on each side but had not extended any farther down. The following noon the tip of the leaf had dried out and was hanging down. Both of the apical side lobes were now drooping. The basal lobes and middle basal part were turgid. The eight day the pricked leaf-blade was wholly wilted and two-thirds dry-shriveled. The petiole was turgid. Drawings were made of this leaf in the different stages of wilt (see fig. 63). The ninth day (noon) the petiole of the pricked leaf was flabby at the tip and half-way to the stem, i. c, for a distance of 2 inches. No leaves above were drooping. The nearest below was beginning to be flabby. The following noon the petiole of the pricked leaf was still green but flabby nearly to the base. The shriveled blade was becoming brownish especially in the pricked area. No leaves above were wilted and the first leaf below had recovered its turgidity during the night. The eleventh day at 10 a. m. there was no change but at 4 p. m. the first leaf below was flabby. The morning of the twelfth day the petiole of the pricked leaf was flabby to the base and shriveled two-thirds down. No further changes were recorded until the morning of the fourteenth day after inoculation. Then the petiole of the pricked leaf was shriveled to the base and hanging down limp. It was green only in the basal portion. The second leaf up was gone (removed by some one) and the third was flabby. None of the leaves below the pricked one had sound blades but there was a branch 8 inches long some distance below which was green and sound. The petiole of the first leaf below was flabby and shriveled two-thirds of the way to the base. The twenty-ninth day the vine was uprooted and dry material saved from it for the herbarium. Search was also made in it for spores of the organism. Fig. 61. *Fig. 61. — Same as fig. 60, but 20 hours later, the only turgid part of the blade being at the extreme base. 230 BACTERIA IN RELATION TO PLANT DISEASES. (29.) The sixth leaf from the tip was pricked many times. The pricks were made along the midrib and in the mid-basal part of the blade. The first signs were noted at 10 a. m., October 31. Two-thirds of the pricked leaf hung down flaccid (that part beyond the pricked area) . The rest of the vine was normal. The following day the inoculated leaf -blade had entirely collapsed and was hanging down. It was dry-shriveling but still green. The petiole was turgid. The eighth day the pricked leaf-blade was wholly dry-shriveled. At noon of the ninth day the petiole of the pricked leaf was flabby and shriveled nearly to the base but was still green. The first leaf Fig. 62.* above had become flabby also. Twenty-four hours later the blade of the pricked leaf was wholly brown-shriveled, and its petiole much as before, i.e., shriveled nearly to the base. The second leaf above had become flabby and was drooping. The first leaf below, which was separated from the pricked one by a long internode, was still turgid. The eleventh day (a. m.) the second leaf above was wholly flabby and drooping. The tip of the third leaf above was also flabby and drooping. The *Fig. 62. — Segment of a cueumber-stem attacked by Bacillus tracheiphilus: Cross-section (inner phloem of an ■ >utcr bundle to surface of stem) after fixing in absolute alcohol. All the spiral vessels are occupied, also two pitted vessels. Bacterial cavities in primary vessel parenchyma. Bacterial masses and softer tissues contracted by alcohol. Anacostia, D. C, July 21, 1903. Drawn from stained section with aid of Abbe camera. Slide 178-5. WILT OF CUCURBITS. 231 first leaf below was now flabby on one side and the second below was drooping slightly and lacked its normal turgor on one side. In both cases the affected side of the leaf was anatomically nearest to that part of the stem which was in direct line with the insertion of the pricked leaf. By 4 p.m., the third, fourth and fifth leaves above were flabby. The first and second below were entirely limp and drooping. The third below was losing turgor at the apex. The morning of the twelfth day all the small leaves at the apex, above the fourth leaf up, were flabby but there were no further changes. At 5 p.m., the third leaf below was still turgid. The morning of the fourteenth day there was still a small portion of the base of the petiole of the pricked leaf which was turgid. The flabby portion just above this was yellowish. The blade of the third leaf below was now flabby. This was separated from the next leaf above by a long internode. Nothing now remained free from signs of the wilt except the stem. The upper portion of the vine was removed and taken into the laboratory for microscopic examination. The vessels were found to be full of the bacillus which strung out in fine gummy threads from the cut surface of the stems. A part of the vine was put into alcohol for parafnne sections and the rest was saved dry to search for spores. The bacteria were exceedingly abundant in the stem and the inner tissues were considerably broken down. The organism was cultivated out and found to be Bacillus trachciphilits. My method in this case was as follows: The stem was cut with a sizzling hot knife, a hole was then worked into the end 3 to 5 mm. deep, i.e., below the burned surface, using a stiff, sterile steel needle. The stem was then squeezed a little, and fluid was trans- ferred from the bottom of the moist cavity into sterile potato broth, using a freshly flamed small platinum oese. In this way four broth-cultures were made from the interior of this plant. Subse- *Fig. 63. — Inoculated leaf of cucumber plant No. 28, shaded to show gradual progress of wilt. Dots indicate needle-pricks. The first wilt appeared some time between fifth and sixth day after inoculation. At end of 5^ days (morning) 2 was wilted and 1 shriveling; on afternoon of same day 3 was wilted; the next morning 4 had wilted. By the eighth day the whole blade had wilted and two-thirds of it had shriveled. Plant inoculated Oct. 25, 1894. Drawn by Theodore Holm. 2 1,2 BACTERIA IN RELATION TO PLANT DISEASES. quently potato-culture No. 3, November 12, from one of these potato broths, yielded a thin, smooth, gray-white, growth, almost exactly the color of the surface of the steamed potato, but easily dis- tinguished from it by its wet-shining appearance. This bacterial slime was viscid, stringing up 2 to 6 em. when its surface was touched with a needle. Once I pulled the needle entirely out of the test tube before the gummy thread broke, and once from a cover-glass I stretched it up 20 cm. Examined in a hanging drop some of the rods were quiet, while others of the same form were actively motile. In other words, this potato-culture was an exact duplicate of No. 8, October 17, from which the broth was inoculated which served to infect this vine. The organism stained readily in carbol-fuchsin ( 1 to 3 min. exposure). The remaining part of the vine was left in the hothouse till the twenty-ninth day but no further developments were recorded. Dry material was saved for the herbarium. A futile search was made in it for spores of the bacillus. (30.) The sixth leaf from the tip was pricked many times in one of the side lobes. There was no result. The pricked leaf remained thrifty although, by actual count, it had received forty pricks in one lobe. The plant was under observation until November 8(14 days) and probably until Novem- ber 23. (31.) The twelfth leaf from the tip was pricked many times in the basal portion of the blade to one side of the center. By 10 a.m. October 31 the whole leaf had collapsed, changed color and was flaccid. It had already begun to dry out on the pricked side showing that the infection had come from that portion of the leaf. These were the first signs noted. The rest of the foliage was sound. The following day at noon the drooping leaf-blade was dry-shriveled except one basal lobe which was still flabby. The petiole was turgid. The next afternoon the blade of the pricked leaf was wholly dry-shriveled and the upper part of the petiole was flabby. The leaves to either side were turgid. The tenth day (noon) the petiole of the pricked leaf was still green, but flabby nearly to the base. It was not yet shriveling. The shriveled blade was becoming brownish. The rest of the foliage was still normal. There was no visible change until the twelfth morning. Then the petiole of the pricked leaf was flabby throughout and shriveled nearly to the base, but green. The leaves to each side, which were separated from the pricked one by long internodes, were still turgid. At 5 p.m. the first leaf below was flabby and drooping on one side (basal and middle lobes). The morning of the fourteenth day the blade of the pricked leaf was greenish brown (like No. 28). The petiole, shriveled nearly to the base, was still green but of a dull shade at the tip. The entire blade of the first leaf down had collapsed and had begun to shrivel at the edges. All the leaves above (7 in number) had collapsed and hung on limp petioles. They were still green. The twenty-ninth day the plant was pulled up and a part of it saved dry for the herbarium. (32.) The eleventh leaf from the tip, i.e., one well down on the old stem was pricked many times in the center of the blade. In a few days the leaves began to turn brown at the edges and by the sixth day the pricked leaf had browned and was almost entirely dried out but not from this disease. The plant had exhausted the soil in the 4-inch pot and done its life-work. Like many others of this planting it had probably ripened a small fruit. This vine was left to grow as long as the others but there was no result. The bacteria never reached the stem but were isolated from their necessary water-supply and destroyed in the browning leaf. (23-) Many deep punctures were made in a small green fruit. The sixth day there were no signs of the wilt. The little cucumber was still green, for the most part, but there was a slight yellow- ing on one side, due probably to ripeness. It was still turgid and healthy. The foliage was normal. The tenth day after inoculation the little pricked cucumber looked as healthy as ever save for a water-soaked appearance around some of the pricks. There was no suspicion of rot and the water-soaked appearance might have been due to handling rather than to the bacteria as I saw simi- lar appearances the preceding winter in Anacostia hot-houses on unpricked as well as pricked fruits. The blade of the small leaf from the axil of which the pricked fruit arose was normal the night before but was at this time flabby. Its petiole was turgid and also the leaves above and below. The next morn- ing there was no change but at 4 p.m. the little leaf subtending the fruit had wholly collapsed and was shriveling. The first, second, third, and fourth leaves above the inoculated one were now flabby. Those below were turgid. The vine had bent or trailed so that the former were below the fruit. The morning of the twelfth day there was no rot in the fruit but it was less turgid and yielded more under slight pressure than on the preceding day. The first and second leaves below (up in relation to the earth ) were still turgid. The fourteenth day the tip of the vine was shriveling. A little white fungus which had undoubtedly gained entrance through one of the needle-pricks, now caused a small sunken place. The tuft of white hyphae was first noticeable the preceding day. The leaves below were not yet flabby except the one subtending the pricked fruit. The leaves above were shriveled. The upper part of the vine was now bri lUght into the laboratory for microscopic examination. The vessels were found to he full of the bacillus which strung out in fine gummy threads from the cut surface of the stem and of the inoculated fruit. The bacillus was especially abundant in the latter. The inner WILT OF CUCURBITS. 233 tissues of the stem were considerably broken down. The organism was cultivated out using the same method as in No. 29 and found to be Bacillus tracheiphilus. Samples were also saved in alcohol for paraffin sections. The following day the second leaf below was flabby. The twenty-ninth day this vine, together with the others, was dug up to make room for new plants. A portion of it was saved dry. Search was made in it for spores. Potato-culture No. 9, November 12 (from No. 9, November 8, which was inoculated direct from the interior of this vine) when examined November 28 was covered with a smooth, wet-shining thin gray-white slime, exactly typical of B. tracheiphilus. This was the most viscid culture seen. It strung up readily 30 to 40 cm., once 50 and once 53 cm. Exam- ined in a hanging drop the rods were of variable length, but many short. All were about the same breadth, and none were motile. A cover-glass preparation was stained. Remarks. — Signs appeared on four of the plants in 55 days after inoculation. Two failed to take the disease, and signs appeared on the other four as follows : On two at the end of the sixth day, on one at the end of the eighth day or the beginning of the ninth and on one (that inoculated in the fruit) the tenth day. On October 31, the day on which six plants first showed the disease the temperature of the hothouse was 720 F. at 10 a. m. and 8o° a few hours later. The history of the culture used for making the foregoing inoculations is as follows: (1) Beef broth No. 1, Sept. 17, inoculated direct from vine No. 2, which was inoculated direct from a cucumber plant diseased naturally. (2) Streak from 1 on slant agar (Red x September 27). (3) Potato cylinder (No. 8, October 17) inoculated with one colony from margin of the preceding. (4) Potato broth (tube 8, October 23) made from 3. (5) Vine 25, etc., inoculated from tube 8, October 23. Inoculations of November 2, 1894. An attempt was made to discover whether squash-bugs and cucumber-beetles were the means of introducing this organism into the tissues of the plant. Cucumber plants (Cucumis sativus) were used for the experiment. Some of them were sprayed with a pure culture of Bacillus tracheiphilus and placed under bell-jars along with the insects. Other plants were not thus sprayed but were placed in insect cages into which the bugs or beetles were intro- duced after infecting them with the bacteria. Several attempts were made to infect some of the plants. The infectious material first used was obtained from a potato-broth- culture (tube 8, October 28) which was from slant potato-culture No. 8, October 17. The latter was the first sub-culture from a colony, the original source of which was the interior of vine No. 2. The potato-broth was faintly cloudy with rolling clouds when shaken and was full of an actively motile bacillus, as determined by examination of a hanging drop. The greater part of this tube was poured into about three times as much sterile water, [''or the cage experiments this diluted culture was sprayed on three pieces of cut squash-fruit each about 2X2 inches so that the whole cut surface was wet. The sprayed squash was in three covered dishes. Into one dish I put the spotted beetle (Diabrotica 12-piinctata). Into another I put the striped beetle {Diabrotica vittata) and into a third, a dozen squash-bugs (Coreus tristis). These insects were collected from a squash-field the preceding afternoon and were quite lively. After allowing them to feed upon and crawl over the infected squash- flesh for an hour, they were transferred to cucumber-vines inside of three insect-cages. This work was completed about 1 p. m. For the bell-jar experiments I used the same liquid, but the plants themselves were sprayed with it. These latter experiments (Nos. 35, 36, 39, 40) were begun about 3 p. m. (34 a to e.) Five well-grown vines in three pots were placed in an insect-cage. The vines were 6 to 10 inches high and had 18 good leaves in addition to the cotyledons which had not yet withered. About 10 squash-bugs were taken from the sprayed squash-fruit already described and were put into the cage. Inside of the first half hour they began to stick their beaks into the vines and they were very particular to put them into the veins. The next morning they were not active. Only 3 were on the vines and these were not sucking. At 1 1 a.m. these bugs were taken out and put on pieces of squash freshly sprayed with the same fluid as on the preceding day. This fluid had remained in the atomizer and a microscopic examination showed that the bacilli were still motile. At noon the bugs were once 234 BACTERIA IN RELATION TO PLANT DISEASES. more introduced into the insect-cage. They were taken out at 3'' 30'" p.m. but not until I had seen several of them insert their beaks — always into the veins. The second day the vines were beautifully thrifty. Small diaphanous spots 2 mm. in diameter had appeared on several leaves. November 28, there was no trace of wilt. The twenty-seventh day one of the vines looked suspicious but no un- mistakable signs of wilt appeared until the morning of the thirty-eighth day. Then one small leaf was wholly collapsed and part of its petiole hung flabby. The color and general appearance suggested the bacterial blight. Above and below this wilted leaf were turgid bright green healthy leaves. Three days later no additional leaves had wilted. If this plant was examined under the microscope records do not show. (35 a and b.) A pot containing 2 vines about 8 inches high was placed under a big bell-jar. At 3 p.m. the leaves were atomized on both sides with the bacterial fluid and half a dozen squash bugs were turned loose under the bell-jar. The next morning the bugs were all on the pot, none on the vine. The second day the bugs were rather sluggish and not feeding much. They were removed the morning of the third day. The twenty-fifth day one leaf on 35a was freshly wilted and another small one was found shriveled down into the petiole and so must have become flabby some days before. Possibly this is the leaf referred to on November 9, as showing a tiny trace of wilt on one side of the lamina and of which there is no intermediate record. By the next morning the whole tip had wilted and hung flabby. Below was one good turgid leaf. The vine was small and had no other leaves. The following afternoon the last turgid leaf had wilted. The vine was now removed and brought in for microscopic examination. The vessels in the middle part of the stem were found to be full of the bacillus. In the hypocotyl, below the last wilted leaf, part of the vessels were still free from infection and in the others a portion of the bacteria were plainly motile. On the twenty-sixth day vine 356 was still normal. Nine days later (thirty-fifth day), a small leaf (about the third or fourth from the tip), which had been collapsing for some days, was wholly shriveled. The leaves above it were slightly flabby but as the plant had suffered considerably from mildew, aphides, and lack of water, I was uncertain how to interpret the phenomenon. I watered the pot well and awaited develop- ments. The next morning 1 leaf below and 3 above had wilted, although the soil was moist enough. The plant was brought in and sections were made in various places. The vessels were found full of the bacillus and some of the rods were actively motile. In the vicinity of the first leaf to wilt the parenchyma of the stem also contained the bacillus, farther away the bacillus was sharply restricted to the vessels and in the hypocotyl even these tissues were free from it. Assuming the first signs to have appeared on December 3, the period of incubation was 3 1 days. The vine never recovered from the stunting due to keeping it under the bell-jar. (36 a and b.) A pot containing two vines, 4 and 6 inches high, was placed under a bell-jar. It was sprayed like the preceding and half a dozen squash-bugs were turned loose upon it. The following morning the bugs were on the stem and ground and were not feeding. This was also true the second day. The bugs appeared to be sluggish. The twenty-sixth day the vines showed no signs of the dis- ease. The thirty-fifth day, 36a which was very small and had been languishing for weeks, partly on account of the attacks of aphides, was quite dry with the exception of the base of the stem. The stem was now examined microscopically but no bacteria were found. On the thirty-fifth day 366 had a small leaf (about the third or fourth from the tip) which was wholly shriveled. It had been collapsing for some days. The leaves above it were slightly flabby, due perhaps to lack of water. Three days later vine 366 was badly shriveled with the exception of the hypocotyl and one or two of the first internodes. It was a small vine, never having recovered from its long sojourn under the bell-jar. It was brought in and examined microscopically. In the third internode above the hypo- cotyl the bacillus was abundant in a number of the bundles but motility was doubtful. In the first internode above the hypocotyl the vessels of one bundle contained plenty of the bacillus but they were not so closely packed as farther up and a part of the rods were plainly motile. No bacteria were observed in any part of the hypocotyl. (37 0,6, and f. ) Three pots containing 3 vines, 7 to 9 inches high, with 12 good leaves besides the cotyledons, were placed in an insect-cage into which 8 infected cucumber-beetles {Diabrotica 12- punctala) were introduced at 1 p.m. The latter were very active, flying away from the vines imme- diately to the top of the cage. The next morning the vines had sustained no injury. All the beetles had disappeared except three which lay dead on the ground and were being eaten by ants. Twenty striped cucumber-beetles {Diabrotica vittata) were now fed for half an hour on squash-fruit freshly sprayed with Bacillus tracheiphilus and put into this cage at 10 a. m. Half of them were removed at 3.30 p.m. and the rest were removed the next morning. Two of the 3 vines had been gnawed a little. All were beautifully thrifty. The twentieth day the inoculations were repeated using Diabrotica vittata which had been kept alive on squash-fruits for this purpose. These had been transferred to clean cut squash-fruits in clean covered dishes. (The old squash was moldy and rotten and mixed with dirt.) On them I sprayed at 11 a.m. the contents of a pure potato-broth-culture (tube 2, WILT OF CUCURBITS. 235 November 20) diluted with four times its bulk of sterile water. The insects were allowed to feed until 1 p.m. Then the squash was removed and the beetles allowed to crawl over the bottom, top and sides of the moist infected glass dishes. At 4'' 30'" p.m. they were turned loose in the insect- cage. The next morning most of the beetles were at the top of the cage and had eaten but little. At 1 1" 30"' I removed nearly all of the beetles and put them in dry, clean, glass dishes in order to starve them. They were left thus without food for 24 hours. I then sprayed the beetles and the top, bottom, and sides of the glass dish with the contents of a tube of potato-broth (culture of November 20) mixed with a tube of sterile potato-broth to which an equal amount of sterile water was then added. The potato-broth-culture was faintly clouded with rolling clouds when shaken. It was examined in a hanging drop and one in perhaps 20 to 50 of the rods on the edge of the drop were found to be motile. At 1'' 30'" p.m. the dish was uncovered in the insect-cage and the beetles turned loose again after wetting down the pots and sand on the bench, inside and around the cage. Four days later (the twenty-sixth day from the beginning of the experiment) there was no trace of the wilt. Some of the Diabrotica vittata were feeding very slowly but most of them not at all. On the forty-first day 4 out of 5 leaves on one of the vines were wilted, the lowest leaf and the upper leaves. These leaves had been all right in appearance the preceding day. Diabrotica vittata was still in the cage: 1 or 2 had begun to eat a little of late but most of them were hibernating. On another plant in this cage the third and fourth leaves up yellowed and shriveled in December from serious gnawings but without suspicious signs. Sometime between January 1 and 4 the second leaf up wilted with signs regarded as suspicious. On the morning of January 4, the fifth leaf up showed a decided droop, although the earth was moist enough. Twenty-four hours later 4 small leaves above the fifth leaf were wilted. The stem was green and turgid but there were no healthy leaves on the plant. The stem was now cut open and examined in several places, but no bacilli were found, and the cause of the wilt of the leaves remained uncertain. Possibly bacteria might have been discovered in the leaves. (38 a, b, c.) Three pots containing 3 vines, 6 to 8 inches high, with 12 good leaves besides the cotyledons, were placed in an insect-cage at 1 p.m. with 15 or 20 specimens of the infected cucumber- beetle (Diabrotica vittata). The beetles began to feed at once and all but one or two were taken out at 5 p.m. Holes had been gnawed in the leaves of each vine. The next morning the vines appeared normal, only a trifle gnawed. The second day the vines looked very thrifty. The twentieth day more specimens of Diabrotica vittata were sprayed with a pure potato-broth-culture (tube 2, Novem- ber 20) and introduced into the cage. The next morning the beetles were at the top of the cage for the most part and had eaten but little. At 11'' 30'" a.m. I removed nearly all the beetles, starved them for 24 hours, sprayed them again (with potato-broth-culture of November 20 — see 37a, b, c) and turned them loose in the cage as in the preceding experiment. The twenty-sixth day there was no trace of wilt. Most of the beetles were not feeding at all but some were eating slowly. The forty- first day there were still no signs. (39 a and b.) A pot containing two vines, 16 and 17 inches high with 7 good leaves besides the cotyledons, were placed under a bell-jar after spraying both surfaces of each leaf. About 4 p.m. 6 or 8 cucumber beetles (Diabrotica vittata) were turned loose on them. The next morning all the beetles but two were removed. The leaves were considerably gnawed. One of the remaining beetles was removed the afternoon of the second day and the other the third morning. The twentieth day there were no traces of wilt. (No further record.) (40.) One vine 4 inches high with 2 green cotyledons and 3 good leaves was placed under a bell- jar. About 20 specimens of Diabrotica vittata were introduced at 4 p.m. All went to the top of the bell-jar. The next morning all were removed. The leaves were riddled by bites. The seventh day the upper half of the upper leaf hung down flabby but without change of color. This leaf had been gnawed on both margins. Two days later (noon) 2 of the gnawed leaves showed very suspicious signs and the next morning the local wilt and change of color was unmistakable. Both leaves cer- tainly had the bacterial disease. This vine had been sprayed with a pure culture of Bacillus trachei- phihts. The beetles were placed on it November 2, consequently the first unmistakable signs appeared in 9 days. On the tenth day at noon the leaves were more wilted. All the others were turgid although one was as badly bitten. At 4 p.m. a third leaf had changed color and wilted. This had been only slightly bitten. Evidently there were three distinct infections and perhaps more. On the eleventh day the three infected leaves hung down limp. On November 16 the vine had partly damped off at the surface of the earth. This was due to being watered too heavily the preceding day. (It was still under the bell-jar. ) It had been going long enough to give striking results, however. The petioles were still turgid. On November 17, the vine was brought into the laboratory and examined micro- scopically. Bacteria were found in the vessels, and samples of the plant were saved in alcohol. Remarks. — Up to December 10 only four good cases appeared and all were upon sprayed plants standing under the bell-jars. Three were on plants punctured by the squash- 236 BACTERIA IN RELATION TO PLANT DISEASES. bug and one was on a plant bitten by the striped beetle. The latter was the only ease which came on with any promptness. In the others the signs appeared at such a late date that we must assume the introduction of an extremely small number of the bacteria, and possibly the same results would have occurred without the introduction of the squash-bugs (see notes on check plant No. 22). Records later than December 10 are wanting. Inoculations of November 17, 1894. One well-grown tomato-plant and 6 young squash-plants which were large for their age were pricked with a steel needle at 3 p. m., and inoculated with Bacillus tracheiphilus (cucumber-strain). Nothing could be more vigorous or apparently more disease-resisting than these young squash-vines. They were growing in 3-inch pots in the hothouse at the time of inoculation. (41.) Tomato (Lycopersicum esculentum). A plant about 12 inches high, healthy and growing rapidly, was pricked on several leaflets of 3 leaves. Into one set of pricks I put bacteria taken from a potato-broth-culture of November 15. This was derived from slant potato-culture No. 1, Novem- ber 12, which was made from slant potato-culture No. 8, October 17, already found by previous inoc- lations to be virulent (see Nos. 34, etc.). The second leaf received very sticky, actively motile bacteria direct from tube 1 November 12. The third leaf was inoculated with bacteria taken from a slant agar-culture (No. 12, October 28) which was from No. 1, October 23. This agar-culture was alive the preceding day as determined by its motility. The ninth day there was no wilt. The vine was growing rapidly and was exceedingly vigorous. It was now re-inoculated on 3 end leaflets of one of the leaves, from a colony (in the tube used for inoculations of November 26) in which more than half of the bacilli were actively motile, darting about in the water with great rapidity. The nineteenth day no signs had appeared. On this day (December 6) the vine was inoculated a third time, the part selected being a stout branch on the upper part of the stem, one which grew from the axil of the leaf pricked the ninth day. About 30 pricks were made, the needle being dipped into the culture each time. Sometimes the needle was thrust entirely through the stem. The culture used was a potato- broth (tube 2, December 3) in which the bacilli were motile. The twenty-third day this vine was making a magnificent growth. It had been inoculated three times but was pricked again on this date. The pricks, which were very numerous and which carried into the plant great numbers of sticky motile bacteria derived from potato-culture No. 5, December 6, were made on the main stem toward the top over a distance of 4 to 5 inches — both in the internode and the two nodes (which were less woody). A fifth inoculation was made January 3, using a great quantity of motile living bacteria taken prob- ably from the slant agar culture of December 28. The needle was touched to the culture each time before using and 50 punctures were made. On January 26, this tomato-plant into which I had pricked hundreds of thousands of living rods of B. tracheiphilus at various times during 2 months preceding, had made a strong growth and was entirely free from the wilt. It had outgrown the pot and the whole top was now cut away preparatory to repotting. The inoculation of January 3, like the others, caused no disease. The most that could be seen externally on January 26 was a slight shrinking and change of color to a duller green around the pricks, as if the bacteria had perhaps grown out into the tissue for a slight distance. The pricks themselves had enlarged with the increase in length and diameter of the stem and were bordered by a narrow ring of dry, dead tissue. Owing to growth of the plant the pricks were now about as long again as broad and, including the border of dead cells, most of them measured 1X0.5 mm. The leaves near these pricks had remained healthy and the upper one had sent out a shoot from its axil since the inoculation. This shoot was 12 cm. long, and healthy. Beginning with the node out of which this shoot grew, and extending down 2 cm. there were 1 7 needle-punctures, all on the same side as the shoot and the nearest one within 4 mm. of its base. There was no evidence of any rot or softening around the pricked place externally or internally and a microscopic examination of unstained free-hand sections showed no bacilli in the tissues around the punctures. The branch into which the bacteria were pricked had shown no sign of disease and there was no change of color in the living tissue around the punctures, which were narrowly bordered by a band of white dead cells. This branch was pre- served in alcohol. On March 12, there was still no trace of wilt. (42.) Hubbard Squash (Cucuibita sp.). This vine, which showed the second true leaf, was infected near the tip and in the middle of one of the green cotyledons. Many pricks were made using a white, stickv, wet-shining bacillus from slant potato-culture No. 1, November 12 (cucumber-strain). The eighth day the cotyledon was yellowing in the pricked middle portion but without wilt. The next morning the yellowed area had increased in size. The tip seemed a little flabby but the wilt WILT OF CUCURBITS. 237 was still doubtful. The eleventh day the pricked cotyledon was the only one out of more than a hundred which showed any yellowing. The following day the yellowing cotyledon exhibited a char- acteristic, wilted place in the yellow pricked area. This was about 1.5 cm. long and 8 mm. wide. The disease was progressing very slowly. The twenty-sixth day the pricked cotyledon had wholly shriveled. February 28 the vine was wholly dried out but no bacilli were found in the vessels or parenchyma of the stem. (43.) Hubbard Squash. The first true leaf was pricked many times and inoculated with a white, sticky, wet-shining bacillus from potato-culture No. 1, November 12. The eighth day an irregular area including about 2 sq.cm. at the tip of the blade of the pricked leaf had wilted and changed to a light dull-green color. Certain pricks were inside of the wilted area. The following morning the wilted area had enlarged only a very little. The eleventh day the wilt was progressing very slowly. There was scarcely a larger area invaded than 3 days before. The twenty-sixth day the small wilted place first discovered had been dried up and brown for a long time. The disease had not spread. Six days later the leaf was brought into the laboratory and the portion which had wilted was put into alcohol. February 28 the vine had all dried out except the stem which was yellowish green and flabby. No bacilli were found either in the vessels or in the parenchyma of the stem. (44.) Hubbard Squash. One of the green cotyledons was pricked and inoculated with bacteria from the potato-broth-culture of November 15. This was a pure culture and was actively motile on November 16 at noon. The eighth day no signs had appeared. The twenty-sixth day the pricked cotyledon was wholly shriveled. February 2S the vine was dried out completely. There were no bacilli in the vessels of the stem. The same day that this plant was inoculated, loops from the potato- broth of November 1 5 were transferred to other sterile potato-broths to see if the organism would grow. In 4 days there was typical clouding in these tubes. (45.) Hubbard Squash. The first true leaf was pricked and infected from the potato-broth- culture of November 15. The eighth day an irregular patch including about 1 sq.cm. had wilted and changed to a lighter, dull-green color. This was on the margin of the leaf about 3 cm. to one side of the midrib. It included some of the pricks. The eleventh day the wilt was progressing very slowly. The invaded area was scarcely larger than 3 days before. The twenty-sixth day the pricked leaf was yellow and flabby on one side. Six days later the wilted portion was brown and dry. The disease appeared and dried out without spreading. The leaf was brought into the laboratory and the diseased portion put into alcohol. February 28 (one-hundred and third day) the vine had dried out with the exception of the stem which was still yellowish green and flabby. There were no bacilli in the vessels of the stem. (46.) Hubbard Squash. One of the green cotyledons was pricked and inoculated from a thin, white, wet-shining growth on slant agar (tube 12, October 28 from 1, October 23). The eighth day there were no signs of the wilt. The twelfth day the pricked cotyledon was slightly yellow at the apex. The twenty-sixth day the pricked cotyledon was wholly shriveled. On January 26, (2 months after inoculation) the vine was stunted and had lost all of the lower leaves, probably because it was still in a 4-inch pot. The upper ones were green but small. The vine was brought into the laboratory and its stem was examined but no bacilli were found either in the vessels or parenchyma. (47.) Hubbard Squash. Bacilli from a thin, white, wet-shining growth on slant agar (No. 12, October 28 from 1, October 23) were pricked into the first true leaf. The afternoon of the eighth day a small area of the pricked part of the blade looked wilted but it was somewhat doubtful. The next morning it was turgid. The vine was growing rapidly and was four times as large as when the bacteria were pricked into it. The eleventh day the leaf showed a very slight change of color in the middle of the pricked area, but I was doubtful whether this was the true wilt. In a day or two there was a tiny place radiating from the pricks (3X8 mm.) which was unquestionably wilted. The twenty-sixth day the disease had not spread. The thirty-second day the wilted spot was brown and dry. The disease had appeared and then dried up. It had not spread at all. The small spot was put into alcohol for future examination. On January 26 the vine was stunted and had lost all of the lower leaves. It was still in a 4-inch pot. The upper leaves were green but small. The stem was now examined but no bacilli were found either in the vessels or the parenchyma. Remarks— Ihe evidence derived from the inoculated tomato tends to show that the bacterial wilt of cucurbits is distinct from that of tomatoes, especially since 8 muskmelon plants and 1 cucumber plant inoculated on January 3 from the same culture as that used for the fourth inoculation of the tomato promptly contracted the disease. The inoculated squash-leaves doubled in size between November 17 and November 22. On vines 43, 45, and 47 wilt spots appeared after a time in the inoculated part of the leaves, 238 BACTERIA IN RELATION TO PLANT DISEASES. but did not spread far and soon dried out. A wilt spot also appeared on at least one of the inoculated cotyledons (42). None of these infections led to any secondary signs. These results were all it will be recalled with the cucumber strain of the bacillus. Inoculations of November 22, 1894. Nine small vines of crookneck summer-squash (Cucurbita sp.), 8 to 12 inches in height, were inoculated by means of needle-pricks, using young pure-cultures of Bacillus trachei- philus derived from three different plants (cucumber-strain). Three vines were pricked on one of the cotyledons, three on the blade of the first leaf, and three on the petiole of the latter. All of these were growing rapidly at the time of the inoculation. None of the leaves had reached their full size when pricked. Nos. 48, 49, and 50 were inoculated from tube 9, November 12; Nos. 51, 52, and 53 were inoculated from tube 3, November 12; Nos. 54, 55, and 56 were inoculated from tube 8, November 12. Tubes 9, 3, and 8 of November 12 were white, sticky, wet-shining potato-cultures looking exactly alike. These three cultures were derived from as many tubes of potato-broth which had been inoculated directly from the interior of diseased vines. No. 9 was obtained from vine No. 33, No. 3 from vine No. 29, and No. 8 from vine No. 25. (48.) One of the cotyledons was pricked. The twenty-first day both cotyledons were shriveled but this came about naturally and was not the result of infection. Three months after inoculation the vine was wholly dried up but there were no bacteria in the vessels or parenchyma of the stem. Nos. 48, 49 and 50 were in the same pot. (49.) This vine was inoculated on the blade of the first leaf. The twenty-first day the cotyledons had shriveled but there was no result from the inoculation. February 28, 1895, the vine was wholly dried up. Examination showed the vessels and parenchyma of the stem to be free from bacteria. (50.) The petiole of the first leaf was inoculated. The twenty-first day the cotyledons had shriveled. There was no result from the infection. Three months after inoculation the vine was wholly dried up. It was examined under the microscope for bacteria but none were found, either in the vessels or parenchyma of the stem. (51.) One of the cotyledons was pricked. The twenty-first day the cotyledon had shriveled naturally, and not as a result of infection. February 6, the vine was drying up. It was now exam- ined microscopically but there was no trace of bacteria in the stem. (52.) The blade of the first leaf was pricked. The cotyledons shriveled after a time and the vine dried up but this happened naturally and not as a result of inoculation. The stem was examined for bacteria on February 6, but none were found. (53.) The petiole of the first leaf was inoculated. There was no result from the pricks. The cotyledons shriveled and the vine dried up after a time but the stem contained no trace of bacteria. (54.) One of the cotyledons was pricked. There was no result from the inoculation. Both cotyledons shriveled and by February 1, the vine had lost all its leaves and begun to shrivel. On microscopic examination there was no trace of bacteria in the vessels. (55.) This was inoculated in the blade of the first leaf. The cotyledons shriveled and the vine finally lost its leaves and began to shrivel but without any signs of the disease. There was no trace of the bacteria in any of the vessels of the stem where thin sections were cut and examined under the microscope. (56.) The petiole of the first leaf was inoculated. The behavior of this vine was like the pre- ceding. It finally lost its leaves and began to shrivel but no trace of the wilt appeared. No bacteria were found in the vessels. Remarks. — This result was unexpected and discouraging. The squashes grew from seeds planted October 31. The cultures used were ten days old. Inoculations of November 26, 1894. Four squash vines and one potato vine were inoculated in the hothouse with a 6-day old culture of Bacillus tracheiphilus. A drop of sterile water was put on the surface of a squash leaf and a little mass of the white, sticky, wet-shining, actively motile (just exam- ined) bacillus put into it, stirred around and pricked in with a steel needle. Each leaf received many punctures. The culture used (cucumber-strain) was slant potato No. 2, OC: WILT OF CUCURBITS. 239 November 19 (reinoculated November 21) from potato broth No. 2, November 17, which was inoculated from the interior of vine No. 24. The squashes were crowded, two together, in 4-inch pots. (57.) Summer Crookneck Squash (Cucurbita sp.) This vine which was growing rapidly was pricked on the first leaf. The seventeenth day the cotyledons were shriveled but there was no result from the inoculation. The thirty-sixth day the vine had lost all its leaves and begun to shrivel. Sec- tions were examined under the microscope but there was no trace of bacteria in the vessels of the stem. (58.) Summer Crookneck Squash. The first leaf of a rapidly growing vine was pricked. There was no result from the inoculation. The cotyledons shriveled and after a time (36 days) the vine lost all its leaves and began to shrivel. No trace of bacteria was found in the vessels of the stem on microscopic examination. (59.) Winter Squash var. Improved Hubbard {Cucurbita sp.) . The first leaf was pricked. The vine was growing rapidly. There was no result from the inoculation. The seventeenth day the cotyledons had shriveled naturally. Two months after infection the vine was brought into the laboratory and the stem examined in several places. It was green and long and had lost most of its leaves except those toward the apex, where they were normal but small. There was no trace of bacteria. The plant was kept in a small pot. (60.) Winter Squash var. Improved Hubbard. The first leaf was pricked. This vine behaved like No. 59. After a time it lost nearly all its leaves but no trace of the wilt appeared. The stem which was still green was examined in several places on January 28, but there was not a trace of bacteria. (61.) Potato (Solatium tuberosum). This was a small shoot about 3 inches above the ground. Two small leaves were pricked. There was one other leaf lower down. Another shoot in the same pot was held as a check. The inoculation was without result. In three days the inoculated shoot trebled its size. The pricks were now open places and on both surfaces of the leaf there was a curious, pale greenish, ringed elevation surrounding each prick and Fig- 64.* consisting of an outgrowth of cells from the edge of the wound (fig. 64). February 6, the vine was wholly shriveled. It was brought into the laboratory and examined for bacteria but none were found. Inoculations of December 6, 1894. Five potato, 6 tomato, 7 muskmelon, and 5 squash vines were inoculated in the hot- house with Bacillus trachciphihts (cucumber-strain) from a motile potato-broth culture (tube 2, December 3). A small steel needle was used to make the punctures. (62.) Potato (Solarium tuberosum). Two shoots of a plant 7 inches high were inoculated. On one the stem was pricked, on the other two leaves were pricked. The plant became diseased but not as a result of the inoculation. The foliage was stunted and finally became dry-shriveled. By the twenty-second day the vine was very sickly, both shoots equally so, and without apparent cause. The roots looked healthy and the disease did not seem to have spread from the pricks on stem or leaves. On microscopic examination no traces of bacteria were found in the tissues bordering the wounds. On this date the other inoculated potato shoots were twice as large and looked healthy. (63.) Potato. Two of the three shoots of a vine 9 inches high were inoculated. One was pricked on the stem, the other on two leaves. For a time the vine grew rapidly but 40 days after inoculation the tops of the three shoots were dead and shriveling and none of the foliage was vigorous. This weakening had been gradual and had proceeded from the top of the shoots down and not from the pricked parts. The shoot pricked on the stem was now carefully examined microscopically at top and bottom but no fungi or bacteria were present. The pricked portion was sound. That the dis- ease had not arisen from the inoculations was shown by several facts: (1) It did not begin in the pricked parts; (2) no bacteria were present in any part of the stem, at least not in that part which was badly diseased and wilting; (3) the shoot which was not pricked at all and the one which was pricked only on a leaf were as badly affected as the one pricked in the stem. Subsequently the plant was taken out of the pot, the dirt washed off and the root system examined. There was no sign of fungi or decay. The mother-tuber was not rotten or black and it had preserved its original form and appearance. Sections treated with chlorzinc iodide showed that most of the starch grains had *Fig. 64. — Diagrammatic section of a potato leaf, plant No. 61, showing new tissue formed aboutlipsof needle- wound. Third day after inoculation with Bacillus tracheiphilus. 240 BACTERIA IN RELATION TO PLANT DISEASES. disappeared from the cells, not from all, however, as I should have supposed. The roots bore live- small healthy tubers which seemed to be mature. The largest was 1 inch in diameter and the smallest 0.25 inch. In view of these facts the probabilities are that the plant had simply reached maturity, performed its function, and was dying naturally. Its early maturity was induced by the smallness of the pot. This explanation is the more probable from the fact that the roots had not made any attempt to occupy the new soil of the larger pot to which the vine was transplanted. (64.) Potato. Two shoots, 6 and 12 inches high, were pricked, one on the stem, the other on two leaves. There was no result from the inoculation. The vine grew rapidly for a time, then the leaves began to yellow and shrivel from the top down, and by the forty-first day the upper part of the stem had shriveled also. Sections of the stem of the pricked shoot showed no bacteria. The earth was washed from the roots which were then examined. There were three small, mature tubers. The old one had not rotted. (65.) Potato. Two shoots of this vine were pricked many times, one inoculation well-down on the stem, the other on 2 leaves. The shoots were 10 and 12 inches high. The plant grew rapidly for a time and then began to show signs of disease. The thirty-ninth day it had lost nearly all its leaves and the upper 6 inches of the stem had shriveled. Sections were made from the base of the stem and from the upper part close to the shriveled portion, which was far away from the point of inoculation, but no bacteria or fungi were found. (66.) Potato. The plant bore four shoots, three 8 inches, one 10 inches high. Two shoots were inoculated. The tallest shoot was pricked on 2 leaves, and the other on the stem. After a compara- tively short period of rapid growth, the leaves began to yellow and shrivel from the top down and finally the upper 8 inches of the stem shriveled (after 41 days). Sections of the pricked stem made below and at various places above and in the pricked portion were examined under the microscope critically. No fungi or bacteria were present or at least none were sufficiently abundant to be detected in unstained sections. The other stems in the same pot were equally affected. One of these was not inoculated at all. Clearly the change was not due to the insertion of Bacillus tracheiphilus. On washing away the earth from the roots the cause of the decline was evident. The plant had matured five small tubers and finished its life work. The old tuber was not rotten. (67 a and b, 68, 69 a, b, 70) Tomato {Lycopersicum esculenlum). These 6 plants were 6 to 8 inches high. Each was given many pricks, some in one leaf, others in the stem. There was no result from the inoculations. (71, 72.) Muskmelon var. Miller's Cream (Cucumis melo). These vines were 3 inches high. One leaf on each was pricked many times. No result. (73, 74) Muskmelon var. Extra Early Hackensack. Same size as preceding. One leaf of each vine was pricked many times. No result. (75.) Muskmelon var. Extra Early Hackensack. One leaf of this vine was pricked. The eighth day there were slight indications of wilt at the tip of the pricked leaf. Twenty-four hours later half of the pricked leaf was flabby and hanging down. The twelfth day the first and second leaves above the pricked one had collapsed. The plant was not over 3 to 4 inches high. The follow- ing morning the vine was wilted. It was brought into the laboratory and examined under the microscope. Bacilli were present in the interior of the plant in large numbers. Material was pre- served in alcohol. (76.) Muskmelon var. Extra Early Hackensack. One leaf was pricked. No result. (77.) Muskmelon var. Extra Early Hackensack. A small plant like 75. One of the leaves of this vine was pricked. The seventh day after inoculation an irregular wilt spot (about 0.5 sq.cm.) had appeared on the side of the leaf toward the apex and within the pricked area. Two hours later it had spread considerably and now involved about a dozen pricks. The ninth day the whole of the pricked leaf was flabby and hung down, also the next leaf above. (The vine was small and these two leaves were close together.) The twelfth day the pricked leaf, the first leaf above, and the tip of the vine were wholly shriveled. The following morning the whole vine was wilted. It was brought into the laboratory and examined microscopically. Large numbers of bacilli were found in the bundles, some of which were actively motile. Material was preserved in alcohol. (78.) Winter Squash var. Hubbard {Cucurbita sp.). A leaf 3 inches broad was pricked many times at the apex. Nearly 2 months after inoculation (January 25) when the last leaf had shriveled the vine was brought into the laboratory and examined. The plant was stunted by being in the same pot with a larger vine. Aphides and mildew had begun to attack it. There was no trace of bacteria in the vessels or parenchyma of the stem. (79.) Winter Squash var. Hubbard. The apex of a leaf 3 inches broad was pricked. This vine proved very resistant for a time but finally began to shrivel. On February 27 it was wholly dried out. It was not observed very carefully during the two or three weeks immediately preceding this date. 1'his vine was now brought into the laboratory and examined microscopically. The vessels of one WILT OF CUCURBITS. 241 bundle contained bacteria (small bacilli) in large numbers. Higher up one or two vessels contained bacilli of much larger size and in smaller numbers. Some of the vessels also contained a branching mycelium suggestive of Fusarium. The pot stood on a bench where watermelon wilt experiments were carried on previously. (80.) Winter Squash var. Hubbard. The blade of a leaf 1.25 inches broad was pricked many times at the apex. Two months after inoculation the vine was brought into the laboratory. All but a few upper leaves were shriveled. The stem- was still green and turgid. The latter was examined in three places for bacteria but none were found. (81.) Winter Squash var. Hubbard. A leaf 4 inches broad was pricked many times at the apex. No result. Owing to crowding in a small pot only the upper leaves were alive on January 28, and these were dwarfed. The green stem was examined for presence of bacteria, but none were found. (82.) Winter Squash var. Hubbard. A leaf 4 inches broad was pricked many times at the apex. February 28 the vine was wholly dried out. No bacteria were present in the vessels or parenchyma of the stem. Remarks. — None of the potatoes or tomatoes took the disease and only two of the muskmelons. The squashes proved very resistant. They were in 4-ineh pots and grew well for some time after inoculation. No. 79 was not affected at first but seemed to be affected after a long time. On December 20 the squashes were still in 4-inch pots and growing satisfactorily. The early maturity of the potatoes undoubtedly resulted from keeping the plants too long in small pots. All the potatoes in the hothouse behaved in the manner described irrespective of whether they were inoculated or not. The tubers were planted in November and the shriveling began in December and on January 14 was apparent on all but two plants. Several examinations showed no bacteria or fungi in the tissues and the plants had not suffered from aphides or red spiders, nor had they been neglected or frosted. They were in 4-inch pots for about a month when they were transferred to 6-inch pots (about January 4) to see if this would help them to recover. It did not, however. When repotted most of the vines were over 2 feet high. These facts favor the supposition that the yellowing and shriveling was a natural one, occurring after the plants had performed their life-work, which was hastened by the small size of the pots. From this experiment it was evident that the disease could be produced in the musk- melon with bacilli taken from the cucumber, but not with certainty in squashes. The temperature in the hothouse the first two weeks after inoculation (December 6 to 20) varied from 6o° to 900 F. Inoculations of December 10, 1894. Fifteen plants including 3 hyacinths (Hyacinthus oricntalis), 2 Hubbard squashes {Cucurbita sp.), 2 summer crookneck squashes (Cucurbita sp.), 1 potato, 2 cucumbers, 4 tomatoes, and 1 cantaloupe, were inoculated in the hothouse, with a white wet-shining, sticky, motile bacillus (cucumber-strain) growing on a potato-cylinder (pure culture No. 5, December 6, from potato-broth No. 2, December 3). The greatest pains was taken to do the work thoroughly. After each plant was thoroughly pricked I went back over the bench and pricked them again. Much material was used in pricking which was done in the afternoon. More than half the bacteria in this culture were motile. The temperature of the hothouse from December 6 to 20 varied from 6o° to 900 F. (83a, 84, 85.) Hyacinth. These plants had been potted 3 days at the time of inoculation and the green bud had pushed up 1 to 2 cm. In each 15 or 20 needle pricks were made into the bud, some of them deep. Up to the eighteenth day there was no sign of the blight. (86.) Hubbard Squash. At the time of inoculation this vine was about 2 feet high and had 6 good leaves. It was very thrifty. About 40 pricks were made on the apex of the blade of the third leaf which was about 5 inches broad. Up to February 27 (79 days) the vine showed no trace of the wilt. The stem was then examined in several places but the vessels and parenchyma were free from the bacteria. 242 BACTERIA IN RELATION TO PLANT DISEASES. (87.) Hubbard squash. This was a very thrifty vine, about 16 inches high, with 6 good leaves. About 50 pricks were made on the middle and apical part of the blade of the fourth leaf which was about 4 inches broad. The eighteenth day there was no sign of the blight. On February 27 the vine was entirely dried up but an examination of the stem showed no bacilli in the vessels or the parenchyma. The Hubbard squashes were planted October 3 1 and were in small pots. (88.) Summer crookneck squash. This vine was about 1 foot high and had 4 good leaves: 75 pricks were made in the blade of the second leaf which was about 2.75 inches broad. The wilt did not appear. February 27 the vine was wholly dried up but no bacilli were found in the vessels or in the parenchyma. (89.) Summer crookneck squash. This vine was about a foot high and had 4 good leaves: 60 pricks were made on the blade of the second leaf which was about 2 inches broad. February 6 the vine was nearly dried iip. There was not a trace of bacteria in the stem. Three shoots in a 3-inch pot sufficiently explains the early death of the plant. (90.) Potato {Solatium tuberosum). A shoot about 16 inches high was selected for inoculation. Many pricks were made in the middle and upper part of the stem over a distance of about 8 inches. Two leaflets were pricked may times also. No sign of disease appeared on shoot or leaves. On February 6 the vine was brought into the laboratory and examined. It was still green and there was no trace of bacteria in the stem. (91.) Cucumber (Cucumis sativus). This vine, which was about 16 inches high, was rather badly mildewed at the time of inoculation and some aphides were present. The first and second leaves had shriveled from the mildew. Six good leaves remained. The blade of the fourth leaf which was about 2.5 inches broad was pricked many times. At 11 a. m. the eighth day after the inoculation there were no signs of the disease, but at 9 a. m. the following day the wilt was of such an extent and character that it must have appeared soon after the previous day's observation. The affected leaf was cut off close to the stem to see if in this way the plant could be prevented from taking the disease. The length of the petiole was 4 cm. The blade was 6 cm. long by 7 cm. broad. The wilt was in an irregular wedge-shaped piece (the pricked area) broadening toward the apex (fig. 65). It extended in the vicinity of the midrib, to within 1 .5 cm. of the tip of the petiole. Fourteen days after inoculation the second leaf above the pricked one was wilted and the following day the first leaf above and the first below the pricked one were wilted and shrivel- ing. The first constitutional signs were 5 days after the removal of the pricked leaf. On the sixteenth day the last leaf collapsed. The day follow- ing the stem was still green and turgid. It was 43 cm. high and had iointernodes. The vine grew from a seed planted September 21. It had been kept in a small pot (4 inch) along with several other vines and was also dwarfed from the presence of mildew (Erysiphe dehor acearum) and aphides. There were blossoms from each node. The diameter of the first internode (hypocotyl) was 3.2 mm. That of the sixth (just below where the inoculated leaf was cut away) was 2 mm. All the leaves were now dry-shriveled. Portions of the stem were saved dry in an envelope for future examination and portions were put into five small vials of alcohol as follows: ( 1) Base of pricked leaf and portion of internode above and of one below: The vessels of each bundle were gorged with the bacillus and the tissues were somewhat broken down. There were scarcely any motile bacilli. ( >nly after a long search could I find any whatever and then only a very few. (2) Middle part of the fourth internode: The vessels were gorged with bacilli, although the tissue was apparently sound; there were not many motile rods, although a larger number than in the section in vial 1. (3) Third inter- node (not examined). (4) Second internode: The vessels of one bundle were gorged but those of the other bundles were nearly free, some entirely so (?) ; a good many rods were darting and tumbling about, more than in the fourth internode. (5) Middle of first internode (hypocotyl): None of the vessels were clogged ; some appeared to be free, others to have a few scattering motile rods. *Fic. 65. — Leaf of cucumber plan! No. 91 (see text). Drawn by Theodore Holm. WILT OF CUCURBITS. 243 (92. 93> 94. 95-) Tomato. Vines 8 to 10 inches high and pricked many times, 2 in stem and 2 in the blade of one leaf. No results. (96.) Cucumber (Cucumis sativus). This was a vine about 10 inches high with three good leaves. The uppermost one was given many pricks. By the seventh day (9 a.m.) this vine had contracted the wilt very decidedly on the terminal (pricked) portion of the leaf (apical two- thirds of the blade). The preceding day there were no signs of the disease. The wilted portion of the leaf hung down flabby while the rest of the plant was turgid and healthy except for a little superficial mildew. It had taken six days for the disease to develop. It was plain from the different aspect of various parts of the wilted portion that the wilt began in a v-shaped apical portion within the pricked area. Bv 2 p.m. the leaf blade was wholly flabby. The following morning the petiole was still rigid and normal. At 9 a.m. of the eleventh day the whole of the small top above the pricked leaf was wilted. This top was normal in appearance at 5 p.m. of the preceding day. The leaves below were still turgid as was also the case at ib30m p. m. of the same day. The twelfth day (9 a.m.) the first leaf below the pricked one was wilted. It was normal at 5 p.m. the preceding day. The following noon the first leaf below had begun to dry out and the second leaf down was flabby. The fourteenth day the vine was brought into the laboratory and thin sections from the lower part of the first internode below the pricked leaf were examined. The lower portion of the internode was turgid but the upper part had begun to shrivel. The tissues examined were full of bacilli. There- was considerable variation in the breadth of the rods and some were much longer than others (samples were put into alcohol). The bundles were much broken down so that a large cavity had formed. None of the rods were motile. Five inches farther down (under the second leaf below the pricked one, i.e., the leaf which had wilted the preceding day) the vessels of the bundles were gorged. The tissue here was but little broken down and the parenchyma was nearly free from bacilli. A small portion of the rods were actively motile. One and one-fourth inches farther down (in the hypocotyl) the bacteria were confined to the vessels and a portion of them were motile. The bacilli strung up from the cut stem 2 to 6 cm. when touched with a needle. Beef-broth-cultures were made December 24 from the interior of the stem where some of the bacilli were observed to be motile. Potato-cyl- inders inoculated from one of these broths (December 27) yielded in four days a rather scanty, wet- shining, white growth. (97.) Extra early Hackensack cantaloupe (Cucumis melo). This plant was 3 inches high and had two small leaves besides the cotyledons, which were still green. Many pricks were made on the first true leaf, the blade of which was about 1.25 by 0.75 inch. The eighth day (9 a.m.) the pricked leaf was almost wholly flabby. It was normal at 2 p.m. the preceding day. The ninth day the vine was wilted and was brought into the laboratory and examined microscopically. Many bacilli were present in the vessels, but there were not so many as in 75 and 77, inoculated 4 days earlier, and examined the same day. Remarks. — This series of inoculations, and that of December 6, settled the fact that the muskmelon disease is identical with that of the cucumber. I was in much doubt about the squashes. Only one plant (No. 79) had contracted the disease while all the squashes in this series and those in several other experiments (November 26 and December 6) refused to take the disease. I then interpreted these results as perhaps due to individual or varietal resistance on the part of the squash-plants experimented with since subsequent experiments performed in the same way gave positive results (see Nos. 215, 216, 217, 218, and 220). No. 91 was very instructive in that it confirmed two suspicions: (1) The number of motile bacilli increases as one gets farther and farther away from the point of infection (i. c, among younger rods) ; (2) The organisms pass down the vessels a long distance ahead of the signs. In this case the inoculated leaf was certainly removed within less than 22 hours and probably within 12 to 18 hours of the appearance of the first signs, while the greater part of the blade of the leaf (at least five-sixths of it) was still apparently sound and while the stem was still separated from the nearest wilted part by a distance of 5.5 cm. Nevertheless the bacilli had already passed down into the stem, so that the progress of the disease was but little if any slower than it would have been had the leaf not been cut away. (i) Sept. 1. (2) Sept. 1. (3) Sept. 17- (4) Sept. 27- (5) Oct. ■7- (6) Nov. 12. (7) Nov. 15- (8) Nov. 20. (9) Dec. 3- (10) Dec. 6. (11) Dec. 10. 244 BACTERIA IN RELATION TO PLANT DISEASES. The culture used in making these inoculations had the following history: Typical diseased cucumber vine brought in from Anacostia, D. C. Vine No. 2, inoculated with white sticky slime direct from above plant. First signsthe fourth day. Beef-broth-culture direct from the interior of vine 2. Slant meat extract peptone agar streak culture from 3. Potato-cylinder inoculated from one colony in No. 4. Potato-cylinder from 5. Potato cylinder from 6. Streak on a slant tube of unfiltered alkaline potato-agar. Potato-broth tube No. 2 from 8. Potato cylinder No. 5 made from 9. Plants inoculated from Tube 5, December 6, which now contained the same wet-shining, white, motile, and very sticky bacillus with which I started on September 1 . On December 20 the squashes were in 4-inch pots and growing satisfactorily. Inoculations of January 3, 1895. Potatoes, hyacinths, squashes, tomatoes, muskmelons, pear and cucumber were inoculated in the hothouse with a white, sticky schizomycete from a slant meat extract peptone agar culture of December 28 (cucumber-strain). The culture was examined the day the inoculations were made and found to consist of bacilli, a large proportion of which were motile. Great care was taken to avoid contamination, to use the bacteria as soon as taken from the surface of the agar, and to make the needle-punctures as small as possible. My method in this instance was to put a little of the white, sticky mass on the surface of the plant (leaf or stem) and then prick it in elsewhere, touching the needle tip to the slime each time before inserting it. Each plant received many punctures. In case of the squashes extra pains was taken to select full grown or nearly full grown leaves, and to make many very small needle-punctures, so as to prevent the bacteria from drying out and to secure their introduction into suitable tissues. Up to this time the inoculations into squashes had been unsuccessful. The temperature in the hothouse was 68° F. For the past month the day temperature had been about 700 F. (98.) Potato {Solatium tuberosum). A very thrifty plant, growing in a 4-ineh pot, was pricked on a terminal leaflet and in the middle part of the stem. No result. (99.) Potato. This plant was growing in a 4-inch pot, was 14 inches high and very thrifty. It was pricked on one end-leaflet and in the middle portion of the stem. The eighteenth day the vine was examined for bacteria. The top had been shriveled for a week, and all the leaves had fallen. The rest of the stem, including the pricked parts, was normal. No bacteria were found in the vessels of the stem between the pricked part and the shriveled tip. (100.) Hyacinth (Hyacinthus orientalis). The leaves of this plant were two inches long at the time of inoculation. The flower scape had not yet elongated but the buds were visible. It was a very healthy plant. Many pricks were made on the apical part of four different leaves. No result. (101.) Hubbard Squash {Cucurbita sp.). Many pricks were made on the apical portion of the blade of the sixth leaf of a small thrifty vine. The twenty-fifth day the plant was brought into the laboratory for examination. Most of the leaves had fallen except those toward the apex where they were normal but small. The stem was green and long. It was examined in several places but not a trace of bacteria was found. The plant was crowded in a 4-inch pot. (102.) Hubbard Squash. A comparatively large, thrifty plant was inoculated. Many pricks were made on the apical portion of the blade of the eleventh leaf. This leaf was about 6 inches from the apex of the vine. The twenty-sixth day the plant was brought into the laboratory and examined. It was at this time 123 cm. long. Only the upper 30 cm. were leafy. The stem was green and turgid throughout. The stem was examined microscopically in three places — toward the base, in the middle, and toward the top. There was not a trace of bacteria. The plant had been kept in too small a pot. (103.) Red Hyacinth (Hyacinthus orientalis). This plant bore leaves only about 2 inches long and the flower-stalk had not yet elongated. The plant was a very thrifty one. Many pricks were made on the apical portion of each of four leaves. No result. WILT OF CUCURBITS. 245 (104a.) Summer Squash (Cucurbila sp.). This vine was about 2 feet high at the time of inocu- lation. It was pricked on the apical part of the blade of a leaf toward the top of the vine. The seventh day the pricked leaf had turned yellow and had become flabby on the pricked side but it was the lowest leaf and I could not tell whether the wilt was due to bacteria or to other causes. One day later the pricked leaf had wholly collapsed and shriveled. The other leaves were normal. Up to February 25 there had been no further wilt and no bacteria were found in the vessels. There was a small pocket on one side of the stem between the epidermis and sclerenchymatic ring in one place where a section was cut. This was filled with small bodies resembling bacteria but none were found in the deeper tissues. This plant was crowded in a 4-inch pot. (1046.) Summer Squash. This vine was growing in the same pot as 10417. It was about 15 inches high and was pricked on the apical part of the blade of a leaf toward the tip of the vine. The sixth day one side of the pricked leaf showed a faint trace of the wilt. The following morn- ing half of the pricked leaf had wilted and had begun to shrivel and hang down. The blade of a small leaf 0.5 inch above had also collapsed. All the other leaves were normal. Twenty-five hours later both of the wilted leaves had shriveled. The rest were turgid. Six days later, the fourteenth day after inoculation, the rest of the leaves had wilted but the stem was still turgid. It was then brought into the laboratory and examined. Sections of the stem were cut in the vicinity of the pricked leaf (above and below) and also an inch farther up. There was not a trace of bacilli in the vessels or paren- chyma. Many vessels were full of tyloses. The plant was erect and only about 15 inches high. It was growing in a small pot and had been overtopped and partly crowded out by a larger plant growing in the same pot. Some mildew was growing on it. (105a.) Summer Squash. This vine, which was about 2 feet high and was in blossom, was pricked on a side lobe of a well-developed leaf. Many very small punctures were made. On February 25 there was no wilt and no bacteria were found in the vessels or parenchyma. (105/3.) Summer Squash. A vine about 15 inches high, growing in the same pot as the preced- ing, was pricked on the apical portion of a well-developed leaf. There was no wilt and an examination after 53 days showed no bacteria in the vessels or parenchyma. (106.) Tomato (Lycopersicum esculentum). A thrifty but watery vine about 2 feet high, growing in a 4-inch pot, was pricked on one leaf and also very thoroughly the whole length of a middle inter- node including its two nodes. Thousands of living bacteria were put in. No result. (106a.) Check plant in the same pot. (107a, b.) Tomato. Two very healthy but rather watery vines about 20 inches high were inocu- lated: one was pricked very thoroughly the whole length of a middle internode and its two nodes, the other was pricked on one leaf only. No result. (107c.) Check. (io8a.) Muskmelon (Cucumis melo). A small muskmelon vine was pricked on the blade of one leaf. The eighth day there was a small wilted place on the apex of the pricked leaf, arising from the needle-stabs. Twenty-four hours later one-third of the apical portion of the pricked leaf had wilted. The tenth day all of the pricked leaf had wilted. January 31 the last leaf shriveled. The next day the plant was examined. There were no bacilli in the vessels of the stem and the barest trace of them in the petiole of the pricked leaf. None of the vessels were stopped up. Toward the end this plant was badly attacked by mildew, and its death must be ascribed to this fact and to the small pot. although it was infected by the bacteria. (1086.) Muskmelon. This vine, which was small and was growing in the same pot as the pre- ceding, was pricked on the blade of one leaf. The morning of the sixth day the pricked leaf showed a wilt spot on the side of the blade near the margin. This was in the pricked area and included only about one-twentieth of the blade. Twenty-four hours later about one-third of the pricked leaf had wilted and the following morning three-fourths of the leaf -blade had become flabby and wilted. The petiole and the other leaves were turgid. The next morning the whole of the pricked blade had wilted and also that of the first leaf up. The petioles were still turgid. The next day (afternoon) the upper part of the stem had wilted. The vine was cut and put into alcohol for sections. Bacteria were demonstrated in the bundles of the stem by a microscopic examination. (108c.) A check melon in the same pot remained free from the disease. (109a.) Muskmelon. This was a small vine and was pricked on the blade of one of the leaves. The fifth day three-fourths of the pricked leaf had wilted. The following morning the whole of the blade of the pricked leaf had collapsed. The eighth day the pricked blade was dry shriveling. 246 BACTERIA IN RELATION TO PLANT DISEASES. The petiole was still turgid and none of the other leaves were affected. Twenty-four hours later the tip of the petiole of the pricked leaf had begun to shrivel. The blade of the first leaf down had shriveled but the first leaf above showed no sign of the wilt in spite of the fact that the first internode below was three times as long as the first one above. The tenth day the first leaf up had collapsed. (1096.) Muskmelon. A small vine in the same pot as the preceding was pricked on the blade of one leaf. The morning of the fifth day one-eighth of the inoculated leaf (the apical pricked portion) had become dull-green and wilted. Twenty-four hours later the whole of the blade of the pricked leaf had collapsed. Two days later the pricked blade was dry shriveling. The petiole was still turgid and none of the other leaves showed any trace of the wilt. The following day half of the petiole of the pricked leaf had shriveled and the blade of the first leaf up had collapsed. The next day (afternoon) the upper part of the stem had wilted. The vine was cut and put into alcohol for study of location of bacilli by means of paraffin sections. They were found abundant in the vessels of the stem on microscopic examination. (109c.) A small melon in the same pot as 109a and 1096 was held as a check. It remained healthy. (iioa.) Muskmelon. A small vine was pricked on the blade of one leaf. The fifth day seven-eighths of the pricked blade had wilted. Twenty-four hours later the whole of the leaf-blade had collapsed. Two days later the pricked blade was dry-shriveling, the petiole turgid. The blade of the first leaf down, which had begun to show signs of wilting the previous after- noon, had collapsed. The first leaf up was still normal. Twenty-four hours later the petiole of the pricked blade was flabby half-way down. The blade of the first leaf down had shriveled and that of the first one up showed very slight signs of loss of turgor. The first internode above was not half as long as the first one below. The next day the upper part of the stem was wilted and the vine was cut, brought into the laboratory and put into alcohol for sections : These when examined under the microscope showed the presence in the vessels of numerous bacteria. (nob.) Muskmelon. The blade of the first true leaf of a small vine was pricked and inoculated. The fifth day about one-fifteenth of the pricked leaf-blade had wilted in a small spot on one side near the apex and within the pricked area. Twenty-four hours later about three-fourths of the leaf-blade hung flabby. Two days later the pricked blade had begun to dry-shrivel. One cotyledon and the next leaf above the pricked one had begun to wilt. The following morning the petiole of the pricked leaf had begun to wilt at the apex. The blade of the first leaf up had wholly collapsed. January 13 the upper part of the stem was wilted and the vine was cut and put into alcohol. Sec- tions examined under the microscope showed the presence of bacteria in the bundles of the stem. (hoc.) A vine growing in the same pot as 110a and b was held as a check and remained free from the wilt. (1 1 1 a.) Muskmelon. The blade of one of the leaves of a small vine was pricked and inoculated. The fifth day over one-half of the pricked leaf blade (apical pricked part) had wilted. By the following morning the whole leaf had collapsed. Two days later the first leaf above the pricked one had begun to wilt. (.Since 3 p.m. the preceding day). The morning of the ninth day the first leaf above had shriveled and the wilt had invaded the petiole of the pricked leaf. The plant was now removed and examined microscopically. The vessels were found to contain bacilli a part of which were motile. (1 1 lb.) Muskmelon. A vine growing in the pot with 11 ia was pricked on one of its leaf-blades. The eight day there was no trace of the wilt, but 24 hours later about one-fourth of the apex and one side of the pricked blade had wilted and changed to a dull green. The wilt began in the pricked area. The time from the insertion of the bacteria to the appearance of the wilt was about 8f days, i. e., wilt appeared on the ninth day. The tenth day the stem was still turgid. Vine nib was the last of the eight melons to show the disease. A microscopic examination was made and bacteria were demonstrated in the bundles of the stem. (inc.) A vine growing in the pot with ma and b was held as a check and remained free from the wilt. (112.) Japanese Pear (Pyrus sp.). A small green shoot, 2 inches long, and a half-grown leaf of a Japanese pear seedling were pricked carefully. The ninth day there was no trace of the disease, nor did any signs appear later on. (113.) Cucumber (Cucumis salivus). An old cucumber vine growing in one of the insect cages (No. 38) in which infection had failed was pricked in the apical portion of one leaf-blade. The after- noon of the seventh day there were no signs of the wilt, but at 10 a.m. of the following day about one- sixth of the blade had wilted in the pricked area (fig. 66). The lowest sign of wilt was 1.5 cm. from the base of the blade, and the petiole was 5.5 cm. long. Thus there were 7 cm. of healthy looking tissue separating the diseased part from the stem. The leaf was now cut away close to the stem to WILT OF CUCURBITS. 247 see if the disease could be prevented from entering the stem.* Two days later the first leaf down had shriveled. The eleventh day the vine was brought in and examined in three places, i.e., (i) at the base of the pricked leaf which was removed, (2) at the lower end of the same internode, (3) at the upper end of the next lowerinternode, i.e., just below the last leaf to wilt. Not a trace of bacilli were found in the vessels or parenchyma, i. e., the disease was cut out. This leaf (the last one on the vine) probably shriveled from the attacks of mildew, aphides and old age. Remarks. — This virulent culture promptly attacked the eight muskmelons and the one cucumber, but had little or no effect on the squashes. It caused no disease in potatoes, tomatoes, hyacinths or pear. Theoretically it should have attacked the squashes readily and their behavior was a great puzzle. In five of the melons signs appeared on the same morning, i. e., 4*4 days from the insertion of the bacteria. Inoculations of January 12, 1895. Two squash-vines and one muskmelon-vine were inoculated with fluid taken directly from the interior of the inoculated wilted muskmelon vine No. ma in which a microscopical examination had shown bacteria to be present. Many pricks were made with a small steel needle. (1 14.) Summer Squash (Cucurbita sp.). Many tiny pricks were made on the blade of an upper leaf. On January 25, the pricked leaf-blade and two-thirds of the petiole were wilted (they were turgid the morning of January 24). This leaf was now removed and the petiole examined care- fully, but I did not find any evidence of bacteria in the vessels or the parenchyma. The leaf was badly attacked by mildew. February 25 there was no wilt, and on microscopic examination I could find no bacteria in the tissues of the stem. (115.) Hubbard Squash {Cucurbita sp.). Many tiny pricks were made on the blade of one of the upper leaves. February 27 there had been no wilt and three sets of sections from the stem at different heights showed no bacteria in the tissues. (116.) Muskmelon (Cucumis melo). This vine which was a small one was pricked rather carelessly many times on the blade of a small leaf and held as a check on the aquashes. The first wilt appeared the ninth day on the blade of the pricked leaf, involving about half of it. The following day the blade and petiole of the pricked leaf had begun to shrivel. The vine was brought in and its interior examined microscopically. The base of the petiole was full of the bacilli, a large proportion of which were very actively motile. The motions consisted of a straight ahead, sinuous, slow or rapid movement for long distances in all directions ; a somewhat slow, curved movement carrying the rods long distances; and a slow or rapid tumbling motion. Remarks. — Up to February 28, 1895, the experiments with squashes were very dis- couraging. The failure to induce the wilt with virulent bacteria, capable of wilting cucum- bers and muskmelons in from 6 to 10 days when introduced simply by needle-pricks (as shown repeatedly by control inoculations) remained unexplained. Squash after squash was examined microscopically with great labor, but with exception of No. 79 and 104 a, no bacilli were found in the tissues. The inoculated plants were inspected day after day for many weeks, but none of them exhibited any distinct signs of secondary wilt, with the possible exception of No. 79. In case of a few plants inoculated late in the autumn a small wilted area appeared in the vicinity of the pricks, but did not increase much after its first appearance. In most of the squashes (36 were inoculated) no primary wilt appeared. My conclusions at this time were that the squash-disease must be due to the organism causing the melon- wilt and cucumber-wilt. "The whole of the petiole (exclusive of a few sections from the base cut to examine microscopically) and the lower part of the midrib were put into alcohol along with the upper and partially wilted portion of the blade, for paraffin infiltration. It is important to determine how generally diffused in the parenchyma the bacteria are when the wilt and change of color (to pale green) occurs. Query : Does this depend on destruction of tissues or only on lack of water? (See fig. 68.) fFiG. 66. — Leaf of inoculated plant No. 1 13 (cucumber) showing pricked area and extent of wilt on eighth day. There was none on seventh day. From basal part of wilted area to stem was a distance of 7 cm. (see text). Fig. 66.f 248 BACTERIA IN RELATION TO PLANT DISEASES. Up to this time the evidence in favor of the oneness of the squash-disease and cucumber- disease rested on the following facts : (1) One case of the wilt disease in cucumber obtained by inoculating jour plants with the milky bacterial ooze taken directly from the interior of a diseased squash-stem (September 1, 1894). (2) Primary wilt on several squash-leaves inoculated November 17, 1894, with a pure culture obtained from a cucumber. (3) The rather inconclusive evidence afforded by the presence of bacilli in some of the vessels of plant No. 79, examined 2§ months after inoculation, fungi being present also in some parts of the stem. Opposed to this were many failures to convey the disease by needle-pricks using pure and virulent cultures of the bacillus. In my experiments on these squashes some unknown conditions necessary to infection were not fulfilled. In nature the squash takes the disease readily enough although it succumbs to it much less easily than the cucumber. That the disease is inoculable from squash to squash was also established by my experiments in Michigan in September, 1893. Whether the squash would contract the disease in the field as readily from melons and cucumbers as from squashes, remained to be determined. Inoculations of March 18, 1895. In the hothouse 24 tomato-vines and 6 Japanese pear-seedlings were inoculated with bacteria from a white, wet-shining, thin, sticky, motile growth on slant agar (tube No. 3, March 13, from stab No. 4, January 9. The culture in great part was much younger than March 13, the uninoculated surface of the slant having been spread over with bacteria from the other parts on March 16). Each plant was given a dozen or more pricks. The bacteria were lifted on a sterilized cooled needle. The inoculations were begun about 10 a. m. and finished at 4.30 p. m. It was a windy drying day with a hot sun under the glass. The tube was carefully shielded from all but very short exposures to direct sunlight. Thousands of living bacilli were pricked into each plant. The needle and loop were flamed before each set of inoculations and not used till cool. As an additional precaution the needle was always thrust first once or twice into the cool stem and then used. A loop of the white slime was put on the surface of the stem and stabbed in repeatedly. An examination of each one of the inoculated plants was made on March 27, April 15, May 9, and later dates. (117.) Tomato (Lycopersicum esculentum) . A thrifty vine about 30 inches high was pricked at the base of one of the upper branches (two nodes and the internode). The tissue was rather firm. There was no result (May 9). (118.) Tomato (same plant). Two internodes and one node near the tender apex of a basal shoot. The tissues were soft and the needle entered very easily. There was no result. By May 9 the shoot had grown 2 feet since it was pricked. (119.) Tomato. A thrifty vine about 25 inches high was pricked in a node and two internodes in the apical part of a branch. The tissues were immature and tender. Between the time of inoculation and May 9 the pricked branch grew over 18 inches. There was no result from the introduction of the bacteria. (120 to 140.) Tomatoes of the same size and inoculated in the same way as the preceding. No disease resulted. The 2 1 plants were under observation for 52 days. (141.) Japan Pear Seedling (Pirus sp.)- Pricks were made in the growing tip of a shoot. There was no result. ( 142 to 146.) Five Japan Pear Seedlings. Like No. 141. No result. Remarks. — None of the tomato-shoots or pear-shoots showed anything but slight local disturbances as a result of the inoculations. The pear-shoots blackened around the needle- punctures almost immediately (host reaction) and after that showed no change. The toma- toes swelled slightly around the pricks in some cases, and in others there was a slight falling in and discoloration of tissue immediately surrounding the needle-pricks, nothing more. WILT OF CUCURBITS. 249 Clearly, therefore, Dr. Halsted's inoculations were not made with this organism. On April 5, five of the tomato-plants were watered as follows: Two with 500 cc. 0.1 per cent watery solution of KNO3, one with 500 cc. of 0.1 per cent watery solution of kainit, one with 500 cc. 0.1 per cent watery solution of muriate of potash, and one with 500 cc. of water allowed to stand, with shaking on 2 grams of fine, dissolved bone ash. Five days later each plant was given another 500 cc. of the solution. There was no visible effect from the previous waterings. Only the nitrate of potash seemed to show a doubtful trace on one or two plants. There was nothing distinct and there was no change up to June 2 2 . For the check-plants see the next series. Inoculations of March 19, 1895. Three cucumber-vines (Cucumis sativus) bearing 2 and 3 leaves were inoculated as checks on the tomatoes and pears pricked on March 18. The remnant of the slant agar- culture No. 3, March 13, was used for making the inoculations. A small, sharp, steel needle was used to introduce the bacteria. (147.) Many pricks were made on one leaf. The third day, the vine rotted off at the base: It had been kept too wet and was attacked by nematodes. (148.) Many pricks were made on one of the leaves. The sixth day at 9 a.m. there was no wilt but at 10:30 a.m. there was the slightest trace of wilt at the apex of the inoculated leaf. There was no change of color. The following morning half of the leaf-blade had wilted with a characteristic change of color. Twenty-four hours later (eighth day) the whole of the pricked blade had wilted except a small area near its base. The petiole was rigid. The tenth day there was a slight wilt of the first leaf up and the first one down. The following morning the two leaves last mentioned showed a decided wilt, the first one down being the more badly affected. The thirteenth day all the leaves were wilted and the vine, which was a small one, was cut for examination. The vessels of the stem were plugged by bacilli which were present in enormous numbers. They were also abundant in the midrib and smaller veins of the partially developed second leaf above the pricked one. No motile rods were observed in any of the sections, not even in those near the tip of the vine (specimens saved in alcohol). (149.) Many pricks were made on one leaf. The fifth day (3 p.m.) there was no trace of the dis- ease but the next morning the pricked leaf showed wilt at the apex. About one-fifth of the leaf was affected. There was yet no change of color. Twenty-four hours later over half of the leaf-blade was wilted with the characteristic change of color. The eighth day the whole of the pricked blade had wilted with the exception of a small area near its base. The petiole was turgid. The following day the first leaf on either side of the pricked one was slightly wilted. The eleventh day the blades of these two leaves showed decided wilt, that of the lower one being the most wilted. The thirteenth day all the leaves were wilted and the vine was cut for examination under the microscope. The vessels of the stem were plugged by enormous numbers of bacilli even to the extreme tip (fig. 67). Remarks. — This sufficiently settled the virulent nature of the bacteria pricked into the pear-stems and tomato shoots out of this same pure culture the day before. We may, therefore, conclude that pears and tomatoes are not affected by this organism. The micro- scopic examination shows that bacteria are sometimes present in the veins of the leaf very soon after it wilts, whether they are always present in advance of the wilt or whether the secondary wilt is sometimes due entirely to occlusions of vessels in the stem or petiole remains to be determined (see fig. 68). Inoculations of April 15, 1895. Muskmelons, cucumbers, squashes, pumpkins, watermelons and tomatoes were inoculated in the hothouse at midday with bacilli (cucumber-strain) from a pure, motile, slant agar-culture 48 hours old (tube 1, April 13, from agar stab No. 2, April 8). All the inoculations were made by needle-pricks. The loop and needle were flamed and cooled each time before using. A fresh mass of the slime was fished out of the tube for the inoculation of each plant and the bacteria were pricked in immediately to avoid danger from drying. 250 BACTERIA IN RELATION TO PLANT DISEASES. The inoculations were made under a newspaper-shade to keep off the direct rays of the sun. The temperature was about 8o° F. The vines which were planted March 12 were in good condition and had several leaves besides the cotyledons. (150.) Muskmelon var. Shumway's giant (Cucumis melo). Many pricks were made in the middle and upper part of a leaf 2.25 inches broad. The morning of the fourth day there was no trace of the disease, but at 2 p.m. there was a distinct wilt covering about 1 sq. cm. near the apex of the pricked leaf. It was then 4 days and 3 hours since the leaf was pricked. Twenty-four hours later the Fig. 67.* apical one-third of the leaf-blade had wilted and changed to a dull green verging on slightly yellowish it 9 a.m. the wilted area was not much greater than on the preceding day). The following day (3 p.m.) two thirds of the pricked leaf had wilted. The blade of the first leaf below was also flabby. The internode between these two leaves was very short, i.e., not over 2 mm. The temperature in the hothouse was over 90° F. The seventh day the blade of the first leaf above the pricked one had wilted. It was separated from the latter by an internode only 2 cm. long. The second leaf below, a *Fig. 67. — Cross-section of extreme top of cucumber-vine No. 149 some days after infection with B. tracheiphilus. The plant was inoculated on the blade of a leaf by needle-pricks. The bacteria are a long distance from point of inocu- lation and confined strictly to the bundles, all of which are invaded. Drawn from slide No. 205-2. WILT OF CUCURBITS. 251 \ vUf ¥ I green cotyledon separated from the pricked leaf by a distance of 2 cm., was also wilted. At 4 p.m. of the same day the vine was brought into the laboratory and photographed along with a healthy vine of the same age (Vol. I, fig. 8). The first leaf below the pricked one was examined for bacilli in the veins of the blade. They were present but not yet numerous. The leaf-blade had been wilted rather more than 25 hours (samples were put into alcohol for the microtome). The blade of the first leaf up which was normal at 3 p. m. the previous day but had wilted either the morning of the seventh day or the previous night, was also examined : The spiral vessels at the apex of the petiole contained numerous bacilli. The veins of the blade were not examined but samples were put into alcohol for sections. Bacteria were also found in the vessels of the uppermost small leaf at the junction of blade and petiole. The petioles of these leaves (pricked one included) were rigid and neither these nor the stem had changed color. For the appearance of a single infected bundle in cross-section see fig. 69. For appearance of a whole petiole in cross- section with low magnification consult Vol. I, pi. 3. (151.) Cucumber var. White wonder (Cucumis sativus). Many pricks were made in a basal lobe of a leaf about 2.25 inches across. The morning of the fifth day the first slight trace of the wilt appeared in the pricked part of the leaf. At 2 p. m. the wilted part was duller green and covered an area of nearly 1 sq. cm. in the middle of the pricked part. The following day (3 p. m.) there was only a small increase of the wilt: Not one-twentieth of the leaf -blade was involved. The eighth day about one-fourth of the pricked leaf was flabby and the middle of the diseased area was now brown and dry. The following day the pricked side of the leaf was dry-shriveled and the rest of the blade had wilted. The petiole and the other leaves were turgid. The ninth day the first leaf down (3 cm. below) wilted and the following day the first leaf up (2 cm.) drooped its blade. The second leaf up was then turgid (9 a. m.), but at 1 p. m. its blade was drooping. The petioles were still rigid except that of the first leaf down which was a very slender one. The fifteenth day after inoculation all the leaf-blades were wilted but the stem and all the petioles were green and turgid. Thin sec- tions of the blade of the second leaf up were now examined under the micro- scope and the bacilli were found to be plentiful in the spirals of the leaf- blade and forming cavities around them. They were apparently not in the green parenchyma-cells. Portions of the petiole of the pricked leaf were put for 2 minutes into boiling absolute alcohol containing 1 per cent picric acid and were then transferred to 75 per cent alcohol which was repeatedly renewed, i. e., until all the picric acid was removed. Other portions were fixed directly in 75 per cent alcohol. The first mentioned fixative gave the best results. The petiole of the pricked leaf was also examined microscopi- cally. The vessels were gorged with bacilli and the primary vessel-paren- chyma was broken down. (152.) Winter Squash var. Sibley's or Pikes Peak {Cucurbita sp.). Many pricks were made on a leaf 3 inches broad, on one side about midway from the base to the apex of the blade. The ninth day there was no trace of the wilt, nor did it appear later. (153.) Winter Squash var. Sibley's or Pikes Peak. Many pricks were made on the apical part of a blade about 2 inches broad. There was no result from the inoculation. (154.) Winter Squash (same variety). Many pricks were made on one side of the blade of a leaf about 2.5 inches broad. No result. (155.) Winter Squash (same variety). This was growing in the same pot as 154. were made on the apex of the blade of a leaf about 2.25 inches broad. There was no result from the inoculation. k ?fo l&. m& Many pricks *Fig. 68. — B, cross-section of a squash-leaf wilted by Bacillus tracheiphilus , showing that wilt of parenchyma is due to cutting off water-supply rather than to actual occupation of parenchymatic tissues by the bacteria. At base is a bundle destroyed by the bacteria. Beyond this is a long wilted area in which no bacteria occur. Only a portion of this wilted area could be shown in the picture, the whole length being shown in fig. C; the portion represented in the drawing corresponds to the black part of C. A, neighboring uncollapsed portion of the same leaf, the bacteria in this being confined to a portion only of the vessels of the bundle. Slide 362-1, lower row, last section but one at the right. Drawn with a Zeiss 8 mm. apochromatie objective. No. 1 2 eye-piece, and Abbe camera. 252 BACTERIA IN RELATION TO PLANT DISEASES. (156.) Pumpkin var. Nantucket Sugar (Cucurbita sp.)- Many pricks were made on various parts of a leaf-blade about 2.75 inches broad. There was no result from the inoculation. (157.) Pumpkin var. Nantucket Sugar. This vine which was growing in the same pot as the preceding was pricked many times on the middle and apical portion of a leaf-blade about 3.5 inches broad. There was no result from the inoculation. (158.) Pumpkin var. Nantucket Sugar. This vine was in the pot with 156 and 157. Many pricks were made on the middle and apical part of a blade about 2.5 inches broad. There was no result from the inoculation. (159.) Pumpkin var. Nantucket Sugar. Many pricks were made on the apical part of a leaf- blade about 2 inches broad. There was no result from the inoculation. (160.) Pumpkin var. Nantucket Sugar. This was growing in the pot with the preceding. Many pricks were made on the apical portion of a leaf- blade about 2 inches broad. No disease resulted. (161.) Pumpkin var. Nantucket Sugar. Many pricks were made on the apical portion of a leaf-blade about 3 inches broad. No result. (162.) Pumpkin var. Nantucket Sugar. This vine was growing in the pot with 161. Many pricks were made on one side of a leaf-blade about 4 inches broad. No result. (163.) Watermelon (Citrullus vul- garis). This vine, which was planted March 12, was pricked many times on a side lobe of a leaf -blade about 1.75 inches broad. There was no result from the inocula- tion. (164.) Watermelon. This vine was about a month old. Many pricks were made on the apical part of a leaf-blade about 1 inch broad. No result. (165.) Watermelon. This vine was Many pricks were made on the middle lobe on one side of a leaf- Fig. 69.! the same age as the preceding, blade about 1.25 inches in diameter. No result. (166.) Watermelon. This was growing in the pot with the preceding. Many pricks were made on the middle and basal lobes on one side of a leaf-blade about 2 inches broad. No result. (167.) Muskmelon var. Shumway's Giant. About 20 pricks were made in the center of a leaf -blade over 2 inches broad. The pricks were in a space not over 5 mm. in diameter, and to each side of the midrib. The fifth day (9 a. m.) the first trace of wilt appeared. It extended in a narrow line along the midrib from the pricked area to the tip. It was most noticeable at the extreme tip. At 2 p. m. the apical one-sixth of the leaf-blade had wilted in a V-shaped area from the pricked part outward. The wilt did not yet extend downward beyond the pricked area more than 1.5 mm. The next afternoon about one-third of the pricked leaf-blade (apical part) had wilted and changed color. The seventh day there was no change. The next morning the blade of the first leaf down and of the first leaf up had wilted. The petioles were rigid. About two-thirds of the blade of the pricked leaf was flabby. The ninth day the cotyledon under the pricked leaf had wilted. The opposite one was green and turgid. The second leaf up was flabby. The leaves of the bud were still normal although 'FlG. 69. — Cross-section of middle of a muskmelon petiole.showing a bundle disorganized by Bacillus tracheiphihts with the formation of a large cavity. From inoculated plant No. 150. Drawn from a paraffin infiltrated stained section. Slide No. 208 A 9. WILT OF CUCURBITS. 253 they were lower down than the blade of the second leaf which was wilted. The tenth day the petioles were still rigid but all the blades were drooping except of one cotyledon and on the tiny leaves of the bud which were not transpiring much. The fourteenth day all the leaves were wilted. The petioles and stem were still bright green and turgid. The plant was now examined microscopically. The vessels were gorged with the bacillus and there was an extensive degeneration of the spiral vessels and of the primary vessel-parenchyma (slide 210). Portions of this vine were preserved in three different ways to determine the best way to fix the bacilli and slime in the vessels without shrinkage of the tissues of the host: (1) The first lot was put directly into absolute alcohol which began to remove the chlorophyll inside of two hours so that the lower half of the fluid was decidedly green; (2) 75 per cent alcohol which fixed the slime and in two hours had not withdrawn any chlorophyll; (3) 50 per cent alcohol which did not fix and was worthless. In the 50 per cent alcohol the bacilli oozed out within 2 hours in quantity and formed a milky slime in the bottom of the bottle and all over the ends of the segments. There was not much choice between the effect of absolute alcohol and 75 per cent alcohol. (168.) Cucumber var. White Wonder. About 20 pricks were made in the center of a leaf-blade about 2.5 inches broad. The pricks were all in an area not over 5 mm. in diameter and each side of the midrib. The eighth day there was no trace of the disease but the following day (11 a.m.) the wilt appeared over an area 5X15 mm. extending from the pricked part outward toward the tip along a side vein. The tenth day there was still no wilt except on the pricked leaf. The narrow oblong area which appeared wilted the day before had now dried out and much fresh tissue was involved in the wilt, about one-sixth of the whole leaf being affected, i. e. , each side of the midrib to the apex and also downward over half-way to the base of the blade. The seventeenth day the vine was brought into the laboratory and dissected. The leaf-blades were all wilted some days before. With the exception of the lowest petiole which was wilted and somewhat yellow at the tip, the petioles were all green and turgid. The stem was turgid and normal in external appearance. There was no rot at the base of the plant. The interior of the vine was full of bacilli. No. 168 was the fourth and last check against the squash and pumpkin inoculations, none of which had given any positive results. The internodes of this vine were cut into short lengths and put alternately into two bottles. Those in one were covered for 22 hours with 1 per cent tri-nitro- phenol dissolved in absolute alcohol; the others were put for the same time into absolute alcohol saturated with mercuric chloride. The former proved the best fixative. (169.) Tomato (Lycopersicum esculentum). About 20 pricks were made in a green fruit about one inch in diameter. Some of the pricks were shallow and some were deep. Many thousands of the bacteria were put in. The fourth day the fruit was one-third larger. There was a very narrow rim of dead tissue about the pricks and beyond this a narrow ring of tissue which was darker green than natural. This second ring was not over 0.4 mm. broad. On May 9, 24 days after inoculation, the tissue around the pricks was slightly sunken. The immediate border, 0.2 mm. in width, was dead; beyond this for a short distance (0.5 to 1 mm.) the tissue was a little darker green than natural. There were no other signs. The fruit had become three times as big as when pricked and the other fruit on the cluster (earlier set) was ripe. On May 14 the inoculated fruit was ripe. The second ring referred to above, ripened more slowly than the rest and was still greenish. The fruit was sound, normal and well flavored. (1 70.) Tomato. A green fruit about an inch in diameter was pricked 12 times and a great many bacteria were inserted. There was no result from the inoculation. The fourth day it resembled the preceding. On May 9 the four fruits in the cluster were all growing finely. The pricked one was exactly like 169 at this date. On May 20 it was fully ripe and was picked and eaten. It was entirely sound. The only result from the pricks was death of ruptured cells and retarded ripening just around the track of the needle. Remarks. — The following note of June 10, 1895, may be of interest in connection with the observations on Nos. 169 and 170. Two small green tomato fruits pricked with a pin some weeks ago have become darker green around the pricks just as did those previously inoculated with Bacillus tracheiphilus. Apparently the phenomenon is the reaction of the plant against the puncture and not against the bacteria. This experiment confirmed the earlier ones. Cucumber and muskmelon were found susceptible to the culture used while squash, pumpkin and watermelon were resistant. All of the former and none of the latter contracted the disease. 254 BACTERIA IN RELATION TO PLANT DISEASES. Inoculations of May 13, 1895. Nineteen potato-vines were inoculated in the hothouse with B. tracheiphilus (cucumber- strain) the virulence of which was checked by inoculation into five cucumbers. The pota- toes were planted in 7-inch pots April 23. Large tubers were halved and all but two eyes cut out. At the time of inoculation each pot bore five or six shoots 8 to 10 inches high, growing rapidly and all very thrifty. The cucumbers {Cucumis saiivus) were several months old, 2 to 3 feet high and in bloom. The bacteria used for inoculation were from a pure slant agar-culture (tube 1, May 11) made for this purpose from a glycerin-agar-culture (No. 8, May 1, reinoculated May 8). The growth was a characteristic, smooth, wet-shining, milk- white streak, much more sticky than the glycerine-agar-culture from which it was made. The weather was cool. The temperature in the hothouse at the time of inoculation was 8o° F. All the pricks made were deep. The needle and loop were flamed and cooled each time and thousands of living bacteria were thrust in. The potatoes received many pricks into young leaves and tender shoots. (171 to 189.) Potato (Solatium tuberosum). No result. (190.) Cucumber. The inoculation was made in a leaf borne on the tenth node (there were many nodes above). Many pricks were made near the midrib about half-way from the base of the blade to the tip. The blade was about 5 inches broad. Up to 1 p.m. May 27 (end of the fourteenth day) there was no trace of the disease, but at 5 p.m. of that day a very small area (less than one square centimeter) was wilted. The eighteenth day no constitutional signs had appeared — only wilt and shriveling of the pricked leaf-blade. Half of the latter was dry-shriveled and the rest hung flabby. There was no wilt above or below this leaf. Twenty-one days after the inoculation the pricked blade was wholly dry-shriveled and of a brownish color. The petiole was green and turgid except the upper inch which had become slightly yellowish and a little flabby. The ieaves above and below were normal. Three days later there was still no wilt of the leaves above or below. The tip of the petiole of the pricked leaf was flabby. On the beginning of the twenty-fifth day the first secondary wilt appeared. This was in the first two leaves above the pricked one. The rest of the leaves were turgid. The general infection of the plant was very slow. June 8th (26 days after inoculation) the blade of the first leaf below the pricked one showed wilt (9 a.m.). At noon the blade of the third leaf up and of the second leaf down were wilted and that of the first leaf up had dry-shriveled the same as the pricked leaf. June 15 the vine had lost all its leaves by the wilt but the stem was yet green and turgid. A petiole was now cut across and the sticky bacterial ooze was pricked into the leaves and stems of pumpkin and squash. (No. 198 and others.) The same day the vine was cut and put into 75 per cent alcohol for microtome sections. It was not then examined microscopically but has been since — enormous numbers of bacteria being found in the vascular bundles of the stem. (191.) Cucumber. Many pricks were made in the middle of a leaf-blade (5 inches broad) to one side of the midrib. The pricked leaf was on the tenth node. The sixth day (2.30 p.m.) the leaf had wilted over an area of 1X3 cm. from the pricks outward, along both sides of a main vein nearly to the margin of the leaf. There was no wilt the preceding day at 4 p.m. The seventh day there was little change. Twenty-five hours later the bulk of the leaf was still turgid. The ninth day the whole leaf-blade drooped and the pricked side was drying out. Two days later the whole blade of the pricked leaf had shriveled. The petiole was still green and rigid. In the afternoon of the eleventh day the blade of the first leaf up began to droop decidedly on one side. The following morning it had partly recovered its turgor. At 2 p.m. the leaf-blade hung down flabby. The fourteenth day the blades of the first, second and third leaves up had collapsed and also those of the first and second down. The petiole of the pricked leaf was beginning to shrivel in the upper two inches. Four days later the leaf-blades were all down. The petiole of the pricked leaf had shriveled nearly to the base. The petioles below were turgid but those above were beginning to be flabby. (192.) Cucumber. Many pricks were made in the middle apical part of a leaf-blade about 4.5 inches broad. The leaf was on the eleventh node. The sixth day (2b 30"' p.m.) the leaf had wilted from the pricked part to the apex, a length of 4 cm. and a breadth of about 1 cm. The following morning there was little change. The eighth day the bulk of the leaf was still turgid. The day was cool, cloudy, and rainy. The following day about half of the blade of the pricked leaf had wilted. The petiole was rigid. The eleventh day the whole blade of the pricked leaf had shriveled. The petiole was still green and rigid. Twenty-four hours later the petiole of the leaf was still normal externally as was also the first leaf to either side of the pricked one. Four hours later the blade of the first leaf up had wilted. The fourteenth day the blades of the first and second leaves down collapsed. WILT OF CUCURBITS. 255 The petiole of the first leaf down was rigid and green. That of the second leaf down was flabby and drooping. It was a smaller petiole. The blades of the first, second, third and fourth leaves up had collapsed. The petiole of the pricked leaf was still normal. The eighteenth day all the leaf-blades had collapsed except the two lowest. The petiole of the pricked leaf was still green and turgid as was the case with all those below it. All the petioles above the pricked leaf were flabby, especially toward their tips. (193.) Cucumber. Many pricks were made on the middle apical portion of a leaf-blade 5 inches broad. The leaf was on the ninth node. The 6th day (2 p.m.) the leaf was wilted from the pricked place to the apex. The wilted area widened outward, being about 1 cm. broad at the pricks and 2 to 3 cm. wide farther up. Two days later the bulk of the leaf was still normal but the wilt was spread- ing slowly. May 22, 9 days after inoculation, the whole blade of the pricked leaf had wilted. It was still green and the petiole was turgid. Two days later the whole blade of the pricked leaf had shriveled. The petiole was still green and rigid. The twelfth day (10 a.m.) the petiole of the pricked leaf was normal as was also the first leaf up and the first leaf down. Four hours later two-thirds of the blade of the first leaf up had wilted. The fourteenth day the blades of the first and second leaf down were flabby. The petiole of the first leaf down was rigid. The smaller petiole of the second down was flabby and drooping. The blades of the first and second leaf up were wholly flabby and the third up was beginning to show signs of the wilt. The petiole of the pricked leaf was still normal. Four days later the blades of the first three leaves below were dry-shriveled and the petiole of the pricked leaf was flabby at its tip. The petioles of the first and third leaves down were turgid and green. That of the second leaf down was shriveling at its apex. The blades of the first four leaves up were shriveled and the petioles flabby. The blades of the next two above were flabby but the rest of the leaves (half a dozen still farther up) were normal. (194.) Cucumber. Many pricks were made in the middle of the blade of a small leaf. The ninth day there were no signs but 2 days later the pricked leaf-blade had a narrow, shriveled, dry strip extending from the pricks to the tip (0.8X3 cm-) showing that the wilt had appeared the preceding day. The twelfth day the whole blade of the pricked leaf had shriveled. The petiole was turgid as were also the leaves above and below. The fourteenth day the first leaf down shriveled. Four days later all the leaves were down and the stem was bowed over in the middle. This was a small vine. Remarks. — Three of the cucumbers came down on May 19 (sixth day), and Nos. 191 to 194 were very badly diseased before No. 190 showed any constitutional signs. Signs outside of the inoculated leaf did not appear on the latter until after 24 days. From that time on the progress of the disease was as usual. May 29, and 30 were very hot. On May 21, all of the potatoes were healthy. There was some tearing of the pricked tissues due to rapid growth and on some of the stems there was a superficial blackening of the pricks but no disease resulted. On June 30, all of the potato plants were still free from the disease. These numerous inoculations on potato and tomato were made owing to statements by Dr. Halsted (in Bull. 19, Mississippi Agric. Exp. Station, and elsewhere) connecting causally the southern bacterial tomato blight with a bacterial rot of melons observed by him in Mississippi, New Jersey, and elsewhere, and confused at that time, at least in my own mind, with this disease. Inoculations of May 25, 1895. A series of inoculations was made at 3 p. m., on cucumbers {Cucumis sativus) by spray- ing the striped cucumber- beetle, Diabrotica vittata, with dilute broth containing Bacillus Iracheiphilus and placing them on the plants in an insect cage. The culture used was one in slightly acid potato-broth (tube 3, May 22), containing rolling clouds when shaken. It was examined in a hanging drop and found to contain comparatively few rods, some of which were feebly motile. It was diluted with three times as much distilled water before using for the cage-experiments. (195 a to/.) Cucumbers. Six small, old, rather stunted cucumber-vines, from which most of the aphides had been removed, but which would never amount to much without repotting (in 4 pots) were placed in an insect cage. The soil outside and in was thoroughly wet down and a dozen or two 256 BACTERIA IN RELATION TO PLANT DISEASES. specimens of Diabrotica vitlata were turned loose on them after the insects had been thoroughly sprayed with the dilute broth and left to crawl about in the infected liquid half an hour. Many of the leaves were already whitish on the margins and there were brown spots on the others and some mildew (Erysiphe chicoracearum) . July 16. The experiment failed. (196 a to g.) Cucumbers. Three pots containing seven plants much like those just described were placed in an insect-cage. The lower leaves were injured more than the preceding and there were large dead patches on the margins of some of the leaves and others were becoming whitish probably from malnutrition. The soil outside and in was thoroughly wet down and the wire of the cage was also wet so as to keep the air inside moist for the next 24 hours. The surface soil of the pots was then slightly sprayed with the dilute broth and also, thoroughly, every part of each plant — stems, young fruits, open flowers, old and young leaves (both sides), and the buds and leaf axils, so to as cause the disease if possible. The infectious material was from the same tube as that used for 195. Almost all the aphides were removed but not quite all. The broth used for infection was not very satisfactory. It was not swarming with rods, i. e., there was only here and there one in the hanging drop although of course where so much fluid was sprayed on, the aggregate number of rods was large. The inocu- lations were made on a cloudy, cool, rather damp afternoon. The plants in this cage were to be held as a check on the preceding. July 16. There was no result from the inoculation. Inoculations of June 15, 1895. Squash-vines {Cucurbita sp.) and pumpkin-vines (Cucurbita pcpo) were pricked and inoculated at noon with sticky bacterial ooze directly from the interior of a petiole of the inoculated cucumber-vine 190. Two cucumbers (Cucumis C} s~\ r\ ^g> sativus) were inoculated as checks. The day was sunny r\ c^> ^ /•) and hot. There is no statement as to the number of needle- U <=> ^ ^> (J punctures, but only that the checks were inoculated in the 'ltomm _' SI same way as the others. Cucumber. This vine was inoculated as a check. No record of where inoculated but undoubtedly on some leaf-blade. The fourteenth day (noon) half of the foliage had collapsed. Two r-y/ fl a y\J days later the vine was brought into the laboratory and examined. ^yl 0 0 (-S From the cut end of the stem there oozed a sticky white bacterial C/ slime, drawing out in slender threads, and a microscopic exami- F's- 70.* nation showed that the vessels contained great numbers of a bacillus morphologically like B. trachciphilus. (198.) Squash. Two terminal small leaves were pricked. No result. (199.) Cucumber. This vine was also inoculated as a check. On July 9 some of the leaves were wilted. It was brought in and examined microscopically. The vessels were found to be full of bacilli of variable size (fig. 70.) (200.) Squash. The third leaf from the tip was pricked and inoculated. No result. (201.) Pumpkin. The stem was pricked. No result. (202.) Pumpkin. The stem was pricked. No result. (203.) Squash. The stem was pricked. No result. (204.) Pumpkin. A leaf was inoculated. No result. Remarks. — -The squash-plants and pumpkin-plants were kept under observation for 38 days. The squashes were grown from seeds planted March 12. Here again cucumbers contracted the disease while squashes and pumpkins resisted it. *Fig. 70. — Bacteria, especially aberrant forms, from interior of cucumber-vine No. 199, inoculated with B. trachciphilus. The common forms are 1.8 to 2.5 by 0.6 to 0.7 m- About i : loooor 1 : 2000 is much larger, but transition forms were observed. Rarely one with a distinct capsule was seen. Cover-glass preparation stained by van Ermengem's nitrate of silver method. July, 1895. OS? 'ltomm ' /~) same way WILT OF CUCURBITS. 257 Inoculations of October 5, 1895. A new set of inoculations was made in the hothouse, at 3 p. m., on young cucumbers (Cucumis sativus) , gherkins (Cucumis anguria) , young muskmelons (Cucumis melo) , and young squashes (Cucurbita sp.), using sticky bacterial ooze from the interior of a cucumber-stem obtained from a field of late cucumbers a few miles northwest of Washington. The bac- teria were very sticky and strung out a long distance from the cut end of the stem. The stem was examined microscopically and found to contain many vessels gorged with a bacillus. I washed the surface of the stem, then shortened it several times with a razor, and finally with a flamed needle pricked the oozing bacteria into the healthy plants. The wet sticky surface of the stem was also pressed down on the surface of the leaves and many delicate needle-pricks were made within the wetted area. Especial pains was taken in each case to make the infection thorough. The plants were 6 to 8 inches high except the gherkins which were smaller. All the inoculations were made on the leaf-blades. (205.) Cucumber. The ninth day the pricked leaf had changed color and was drooping but the petiole was rigid. The blade of the next leaf up also drooped some but was a healthy green. Twenty- four hours later the blades of two more leaves were wilted but the petioles were still rigid. The cotyledons were not yet wilted although on nearly the same level as the base of the petiole of the pricked leaf. The eleventh day the cotyledons were drooping but all the petioles were rigid as also 24 hours later. The fourteenth day the vine was brought into the laboratory and examined. All the foliage had wilted and the stem near the earth had bowed over. Otherwise it was normal in external appearance. The upper part of the stem was cut and examined microscopically. The juice was sticky and stringy and the vessels were full of a bacillus which had also flooded out into the surround- ing parenchyma and was motile. The rods were all nearly the same size. The stem was cut cross-wise with a hot knife and dug into with a hot needle and from the cavity thus made bacteria were removed for eight cultures: No. 1, old alkaline potato broth; No. 2, streak on alkaline agar; Nos. 3 to 8 potato cylinders. The agar failed ; all the potato-tubes developed typical cultures of B. tracheiphilus of which four were exceedingly sticky, one moderately sticky, and one only slightly sticky (obser- vations after 5 days). (206.) Cucumber. The ninth day half of the pricked leaf-blade was drooping. The rest of the plant was normal but 24 hours later the blade of the first leaf up had wilted. The cotyledons which arose from nearly the same level as the pricked leaf showed no sign of the wrilt. The petioles were still rigid. The following day the blade of the next leaf up had wilted. The cotyledons and all the petioles were rigid. Twenty-four hours later one of the cotyledons was drooping. The blades of the wilted leaves were badly collapsed but the petioles were still turgid. Four days later (the six- teenth day after inoculation) all the leaf-blades had shriveled and also the apex of the petioles. The twenty-third day the whole plant had shriveled. (207.) Cucumber. By the end of the ninth day the entire blade of the pricked leaf had changed color and wilted. The blade of the first leaf up was also drooping but was of a good green color. Twenty-four hours later one of the cotyledons hung limp. The petioles were rigid. The following day the second leaf up began to roll at the edges. The twelfth day after inoculation the second cotyledon was drooping. The petioles of the wilted leaves were still rigid but the blades were badly collapsed. Four days later the blades and tip of the petiole were shriveled. The twenty-third day the whole plant was shriveled. (20S. ) Muskmelon. The ninth day after inoculation the pricked blade changed color and showed a trace of wilt centering in a group of pricks. The total affected area was only a few square milli- meters. The following day there was very little increase of the wilt, scarcely 2 sq. mm. Twenty-four hours later about one-third of the blade of the pricked leaf was plainly wilted. The following day more than two-thirds of the pricked blade had changed color and was drooping. The petiole was rigid. Four days later the blade of the pricked leaf had shriveled but the petiole was turgid. The next two leaves above now had wilted blades. A week later the whole plant had shriveled. (209.) Muskmelon. The ninth day the tissue in one small group of pricks was dead. Three days later there was a small amount of wilt near the margin. The twenty-third day the pricked parts were dead and the tissue around them was yellow. The tip of the leaf had wilted slightly. The twenty-sixth day the entire pricked leaf-blade had wilted and that of the next leaf up. (210.) Muskmelon. By the end of the ninth day the entire blade of the pricked leaf had wilted and changed color. The petiole of the pricked leaf was rigid. In 24 hours the wilt had increased somewhat but the petiole was still rigid. The following day the pricked leaf-blade and one coty- 258 BACTERIA IN RELATION TO PLANT DISEASES. ledon had begun to shrivel. The first leaf up showed a trace of flabbyness. The next afternoon the other cotyledon was drooping. The other leaves showed no sign of the wilt. Two days later the vine was bowed over at the root and all the foliage had wilted. When the stem was cut a stringy bacterial slime oozed out. The vessels were gorged with a schizomycete which had flooded out into the paren- chyma. Some of the rods were distinctly larger (longer and broader) than the rest. The motility was not made out satisfactorily. Cultures were started from the interior of this vine and a section was saved in alcohol for study. (The cultures made were No. 9, a streak on agar, stock 82c, and Nos. 10 to 14 on potato-cylinders.) The agar culture failed. With possibly one exception (tube 11), the inoculated tubes of potato developed pure cultures of B. tracheiphihis. At the end of 5 days the surface of the potato was covered by a very thin, exceedingly sticky layer, so exactly the color of the potato that it was to be distinguished from it only by its wet-shining appearance. In tube 1 1 most of the culture was of this character, but in the center there was a raised, slightly yellowish portion, believed to be a contamination. (211.) Gherkin. By the ninth day the entire blade of the pricked leaf had wilted and changed color. The petiole was rigid. The blade of the first leaf up was wilted and the edges were rolled inward. The second leaf up was still green and turgid. Twenty-four hours later the cotyledons were hanging down and the second leaf above the pricked one was drooping. The following day there was little change, but 24 hours later the blades of all the leaves (it was a small plant) were badly wilted. The petioles were still rigid. Four days later the leaves had shriveled and also the base of the stem. The plant was brought in and dissected. Under the microscope, the vessels were found to be gorged with a sticky bacillus which strung out when touched. The tissues around the vessels were disorgan- ized. The bacillus was motile and some of the rods were much larger than others — longer and especially broader. (212 a and b.) Gherkin. There were two vines in the same pot. ( )ne contracted the wilt and was removed on the seventh day when half of the pricked leaflet was wilted (2 p. m.). At that time the second vine was healthy but 2 days later (3 p. m.) it had two wilted leaves, the pricked one and the next one above it. (213 and 214.) Gherkins. These damped off . (215.) Winter Squash var. Pikes Peak. Two leaves were inoculated. The ninth day one of the pricked leaves had wilted outward from one of the groups of pricks and this area had changed to a light dull green. There were about twenty pricks in this group and the wilt did not include all of Fig- 71.* them (fig. 71). During the next three days there was little if any change. The sixteenth day about one-third of the pricked blade showed wilt and was drooping. There was also a small wilt-spot on the other leaf, including and surrounding the group of pricks. A week later the greater part of the blade of the lowest pricked leaf was flabby and yellow. The other inoculated leaf was dead in the pricked area, yellow around the pricks and slightly flabby on the whole of that side. November 1 1 (37 days after inoculation) the lowest leaf was dead and brown-shriveled down to the stem. The blade of the upper pricked leaf was shriveled and the petiole flabby at the apex and yellow its whole length. The blades of the next two leaves above were beginning to be a paler green. December 3 the vine was 13 inches long and had twelve leaves, those which had grown since the inoculation being dwarfish and not bright green. The next leaf up as well as the stem was very yellow. The plant was blossoming freely. December 10 the plant was still alive but stunted and yellowish (see photograph, fig. 53, made of this vine and a check). Thin sections were made of the stem and examined microscopically, the bacteria being detected in the bundles. Very few vessels were found to contain many bacteria. There was only one densely plugged vessel, and that was on the outer margin of the xylem near the cambium. The tissue was broken down on the outer margin of the xylem in several bundles. The cross-sections were not sticky. Stem, saved in alcohol. Sections cut from this stem show numerous bacteria in one or more vessels of five bundles but most of the vessels are free from them. (216.) Winter Squash var. Pikes Peak. The ninth day after inoculation about 1 sq. cm. of the blade changed color and wilted, beginning in one of the three groups of pricks (fig. 72). The leaf was also slightly yellow around the other two groups of pricks. Twenty-four hours later the V-shaped small marginal piece of sound tissue was wilted, but otherwise there was little change. The sixteenth day about one-fourth of the pricked leaf was yellowish green and drooped slightly. Seven days later the portion above the line of October 2 1 (see drawing) was yellow and slightly flabby. The rest of the leaf was normal and there were no constitutional signs. November 1 1 (thirty-seven days after inocu- M'ig. 71. — Leaf of winter squash (plant No. 215) inoculated with B. tracheiphihis, Oct. 5, 189.5. On Oct. 14 the shaded area was freshly wilted. Up to Oct. 17 there was little change, but on Oct. 21 about one-third of the leaf- blade was wilted and drooping. For further changes in this plant see fig. 53 A . WILT OF CUCURBITS. 259 lation) the pricked leaf was holding up remarkably well. The petiole was yellowish but turgid, and not more than one-third of the blade was dead although it had lost nearly all of its green color and was yellow. The next leaf above and the next below were beginning to be yellow but showed no wilt. November 29 the vine was still alive but stunted and yellowish and losing leaves toward the base. It was blossoming freely. December 3 this vine resembled 215 except that the three basal leaves had shriveled and the others were not so yellow. The dwarfing of the foliage was distinct. March 4 (5 months after inoculation) the vine was still living but was stunted and branched and bore more yellow leaves than green ones. It and all the other squash vines had resisted remarkably. (217.) Winter Squash var. Pikes Peak. There were no signs of the wilt until the twelfth day after inoculation. Then the tip of the pricked leaf was drooping (without change of color) and an area of about 2.5 sq. cm. had changed color and wilted, where pricked. Four days later there were very decided signs. The entire blade of the pricked leaf was drooping and over half of it had changed color, the change varying from a dull green to a yellowish. The blade of the next leaf below also drooped badly. The insertion of this leaf was half an inch below that of the pricked leaf. This was the only one of the squash vines which showed constitutional signs at this date (sixteenth day). The twenty-third day nearly all parts of the blade of the pricked leaf and of the one below it were shriveled. The blade of the next leaf up (insertion 1 inch above) was a fine green but was beginning to droop on one side — in the lower lobe. November 1 1 the blade of the pricked leaf was brown and dry but the petiole was green and turgid. No additional leaves were affected. November 29 the vine was still alive but stunted and yellowish and losing leaves at the base. It was blossoming freely. By Decem- ber 3, the pricked leaf (with the exception of the base of the petiole) and the cotyledons had shriveled. The plant was 13 inches long and had 12 leaves. The stem was yellowish and the vine was dwarfed but was blos- soming freely. December 31 (nearly 2 months after inoculation) the vine was still alive and, on the whole, looked better than 6 weeks earlier. It had made a new terminal growth, 12 to 15 cm. long, which was in every way more vigorous than the stunted, terminal growth which developed in the weeks immediately following the inoculation. The leaves were larger and of a good green and this part of the plant was blossoming. March 4 the vine was stunted and dead from the tip down a distance of about a foot. It had put out small stunted side shoots and there were some green leaves yet. (218.) Winter Squash var. Pikes Peak. The ninth day after inoculation there was a wilted, dull- green area starting from two of the pricked spots. In one, the wilted part began at one side of the pricks, not including all of them, and covered an area of only 2 to 3 sq. mm. In the other the wilt now covered about 2 sq. cm. and involved all of the pricked spot although the latter was excentric. Twenty- four hours later the previously wilted tissue had lost water and begun to be traversed by many fine wrinkles. The diseased area had increased but little. The following day there was no noticeable change. The progress of the disease was very slow and on the sixteenth day not over one-eighth of the pricked blade had wilted and the blade as a whole had not collapsed. Even 7 days later (October 28) one set of pricks (No. 3) had failed to infect and that part of the leaf was still green. The tissue in the immediate vicinity of the other two sets of pricks was dry and brown. Most of the leaf was yellow or yellowish green and flabby but not shriveled. The petiole was normal. The first leaf up, the insertion of which was 2 inches above that of the pricked leaf, was still green but one lower lobe was drooping. November 1 1 half of the pricked blade was brown and shriveled, the rest was yellow and green mixed. All of it was soft, flabby. The petiole was still turgid and not very yellow. The blade of the next leaf up had a yellow spot on the upper part but was not flabby in any part. No other leaves were affected. November 29 the vine was stunted and yellowish and losing leaves toward the base but blossoming freely. December 3 the stem was 22 inches long and bore 16 leaves, the terminal 7 of which were badly dwarfed. The basal 2 leaves and cotyledons had shriveled to the stem. The foliage was lighter green than that of the uninoculated plants. A month later (December 31) the vine had made a new terminal growth 12 to 15 cm. long. This was of a much more vigorous character than *Fig. 72. — Leaf of squash No. 216, inoculated with Bacillus tracheiphilus, Oct. 5, 1895, 3 p. m., in three places. On Oct. 14, 3 p. m.. the shaded area had changed to a dull green and wilted. Up to Oct. 17. 3 p. m., there was no distinct increase of wilt, but on Oct. 21, that part beyond the dotted line was yellowish green and drooped slightly. For further changes see text. 260 BACTERIA IN RELATION TO PLANT DISEASES. the stunted terminal growth which developed in the weeks immediately following the inoculations. The leaves were larger and of a good green and this portion of the plant was blossoming. March 4 the terminal iS inches was dead. There were some good green leaves and the vine was still blossom- ing. It grew about 3 feet after inoculation. (219.) Winter Squash var. Pikes Peak. This vine resisted admirably although the leaf received nearly 100 pricks. Twenty-three days after inoculation the leaf was yellow-green around the pricks and slightly flabby but there was no well-defined wilt. November 1 1 (37 days after inoculation) the blade of the pricked leaf and of the one next below were somewhat yellow-green but neither was flabby and there were no other signs. November 29 the vine was blossoming freely but was stunted and yellowish and losing leaves toward the base. December 3 the stem was 1 2 inches long and had 1 1 leaves. The basal leaf had shriveled nearly to the stem and the blade of the next up was yellow and shriveling. The stem and the foliage were light green. The foliage was dwarfed but the vine was blossoming profusely. This small plant had already borne 5 big blossoms (staminate). March 4 the vine was still in blossom. It had about 50 leaves, two-thirds of which were pale yellowish but not wilted. The rest were green. (220.) Winter Squash var. Pike's Peak. The ninth day the pricked leaf-blade had changed color and wilted around two of the largest groups of pricks, while 24 hours later at least one-third of the pricked leaf drooped and nearly all of that part had become a lighter green. The wilted area on the opposite side of the leaf-blade (spot No. 2) had not sensibly increased. The following day wilt spot No. 2 had extended a little. There was no other change. Twenty-four hours later this side of the pricked blade drooped also and one of the cotyledons which had its origin at almost the same level as this leaf was beginning to droop. The petiole of the pricked leaf was still rigid. On October 21, the part which was wilted 4 days before was yellow and shriveling and hung down. The petiole and basal part of the blade were still green and turgid. A week later the pricked leaf-blade was brown and shriveled at the edges but green in the middle part and erect. There were no constitutional signs. November 1 1 the pricked leaf was yellowish-green in the middle of the blade. The petiole was still turgid. The blade of the next leaf up was yellow and shriveling at the tip. November 29 this vine resembled the preceding. December 3 it had 14 leaves and the stem was about 16 inches long. All except the basal leaves were dwarfed. Two of the latter had shriveled with the exception of the base of the petioles. It was blooming freely. The lower leaves were pale green. December 31 this vine resembled 219. March 4 the vine was still alive but stunted. It was branched and had some good leaves. Remarks. — There was no sudden wilt of any of these squashes. A memorandum of October 14, states that "most of the squashes are taking the disease " but the final outcome shows that they proved very resistant in comparison with the cucumbers and muskmelons. Up to October 28, only two showed constitutional signs and in these the signs progressed slowly. Indeed on October 24, five of the six squashes appeared as if they were going to overcome the disease. This experiment indicates that the gherkin is also subject to the disease. In the inoculated squashes there was a tendency to profuse branching and blossom- ing. The same thing has been observed in the field (see p. 217). In general, infections are more certain when made from young pure cultures than when made with slime taken directly from the plant. On No. 220, the infections first appeared in the parts most severely injured, i. e., where many needle pricks were made close together. Possibly to start the disease in squashes a larger initial injury is necessary than in case of cucumbers or musk- melons. Probably also, to be very sensitive, squashes must be growing rapidly and these were growing slowly. Query : Are the tissues more acid in slow-growing than in rapid-growing leaves? or more acid in squash than in cucumber? These hypotheses were formulated at the time as a partial explanation of the failures previously recorded. The squashes were planted in 6-inch pots in good soil. Query: Is the squash-disease due to an organism slightly different from the cucumber-organism? Inoculations of November 16. 1895. Three vines, a muskmelon, a cucumber and a squash, were inoculated at 4 p. m., with a white bacillus taken directly from the interior of a diseased cucumber-fruit grown in Barnabas Bryan's hothouse at Anacostia, D. C, and brought in to me for determination on November 14. The bundles of this fruit contained a very sticky stringy bacillus, great WILT OF CUCURBITS. 261 quantities of which were pricked into the three plants. The cucumber-fruit was still green and firm but somewhat shrunken on the upper two-thirds and gummy drops had exuded and dried down. The flesh, except in the center, where the seeds were, had, in most places, a very decided water-soaked appearance but there was no soft-rot. The inoculations were made as follows: The freshly cut surfaces of the gummy fruit were pressed down gently on the leaf from opposite sides several times and then in each case about twenty needle- pricks were made through this moist surface. (221.) Muskmelon (Cucumis mclo). This was a small vine bearing six leaves. The pricked one was the third leaf up and its blade was about 2 inches broad. It was inoculated in two places, i. c, there were two groups of pricks. The ninth day there were wilted areas centering in the pricked spots and covering in each case an area of about 1 sq. cm. The change of color (to a dull faded green) was decided. Four days later the petiole of the pricked leaf was still turgid. The blade of the first leaf up was slightly flabby. The next below was turgid but the blade of the second below had collapsed. This second below was the first leaf above the cotyledons and came out on the same side of the stem as the pricked leaf. Seventeen days after inoculation (December 3) the vine was badly wilted. The blades of all of the leaves had collapsed and the stem was beginning to shrivel near the base of the pricked leaf. It was now brought into the laboratory for minute examination. The stem was first cut near its juncture with the pricked leaf. The vessels here were full of bacilli some of which had flooded out into the parenchyma, but the cross-section was not sticky. The rods were not motile. An inch or two lower down the cut surface of the stem was sticky. An inch or two above the first cut, the vessels were crowded full. Some of the bacteria were motile. There also the cut surface was sticky. Cultures were made December 3 into tubes 5, 6, 7 and 8 of November 20 (which had failed). (222.) Cucumber (Cucumis sativus). This vine was inoculated in the same way as the preceding. The pricked leaf was the third from the cotyle- dons and its blade was 4.25 inches broad. It received four sets of pricks, two on each side of the blade near its margin and about an inch apart. The ninth day on one side there was a wilt spot, about 2 sq. cm. in size, centering in one of the groups of pricks. On the other side there was a wilted area covering about 5X2 cm., including both sets of pricks. The change of color was decided. It is probable that signs appeared on this and the preceding vine the previous afternoon. This would make the time of incubation 8 days. Two days later about half of the pricked blade had wilted. The thirteenth day nearly the whole of the pricked blade had shriveled and changed color. The petiole was turgid and all the rest of the plant was normal in appearance. The seventeenth day the vine was 15 inches long. The first four leaves above the pricked one were now flabby. The petiole of the pricked leaf and of the first two above it were still rigid. Those of the next two, which were younger and softer, were flabby. The uppermost wilted leaf was 8 inches above the node of the pricked leaf. The first leaf beiow (2 inches down) was turgid at 9 a. m. but at noon was slightly flabby on one side. On December 7, bacterial slime from this plant was used for further inocula- tions (see 226, etc.). Bacteria taken from the vessels were not clearly motile. (223.) Winter Squash (Cucurbila sp.). One leaf was inoculated. Five sets of pricks were made, the method of inoculation being the same as that in the two preceding cases. The ninth day there were no signs. The eleventh dav the tip of the pricked leaf had wilted over a space of about a square centimeter. The thirteenth day the tip of the pricked blade had recovered its turgor but was yellowish. The seventeenth day the vine was about two feet long and vigorous. It looked much as if the bacteria would not get out of the pricked leaf into the stem. The leaf still preserved its color and turgor except a few square centimeters at the tip of the blade which was yellowish and alternately turgid and flabby (see fig. 73). The dots indicate the pricked parts and the shaded apical portion the only part which was wilted at the time. December 13 (27 days after inoculation) the pricked areas were dead and the leaf was yellow around them. There was no wilt and most of the blade was green. December 31 three basal leaves including the pricked one. which had become gradually yellow, were shriveled but though I had watched this plant carefully for many weeks the disease gave no indication of spread- ing from the pricked leaf to other parts of the plant. Remarks. — The bacillus used for these inoculations was taken, it will be remembered, from the interior of a green cucumber-fruit. In spite of its appearance I was in some doubt *Fig. 73. — Leaf of squash No. 223, inoculated with Bacillus Iracheiphilus (cucumber strain) Nov. 16, 1895. Plant very resistant. The shaded area was wilted on Dec. 3. Less than natural size. 262 BACTERIA IN RELATION TO PLANT DISEASES. at first as to the nature of the organism in the bundles of this fruit because I had not hitherto known of the occurrence of the cucumber- wilt in hothouses, except as the result of my inoculations. It proved, however, to be infectious and yielded a long series of cultures and successful inoculations. This experiment also shows that the squash is much more resistant than muskmelon or cucumber, at least to what I have come to call the cucumber strain. Inoculations of November 29, 1895. A muskmelon and a cucumber- vine were inoculated with Bacillus tracheiphilus from potato (?) tube 2, October 26 (from agar-stab of May 7). In each case many pricks were made on one leaf-blade. (224.) Muskmelon {Cucumis mrfo). The eleventh day the pricked leaf-blade was partially wilted but the same was also true of two below and the cause was doubtful. Perhaps the wilt was due to the sulfuring done in the house some days before as other vines showed similar results. The seven- teenth day it was still doubtful as to what was the cause of the wilt. December 31 the leaves were spotted and brown but it was doubtful if the bacteria were alive. January 7 the vine had shriveled down to the long hypocotyl which was still normal. It was now cut and examined in two places for bacteria but none were found. (225.) Cucumber {Cucumis salivas). March 4 this vine showed no result from the inoculation. Remarks. — The culture used was 34 days old, and probably dead. Inoculations of December 3, 1895. Bacilli were squeezed out of the cut stem of a wilted muskmelon-vine (No. 221) and direct inoculations were made, at 3 p.m., into the following cucurbitaceous plants : common gourd (Lagenaria vulgaris); Balsam apple (Momordica balsamina); Gherkin {Cucumis angaria); vegetable sponge {Luff a acutangula); and wild gourd {Cucurbita foetidissima) , a big rooted species native west of the Mississippi River. All of these were small plants, i. e., only 3 to 8 inches high, but not stunted except the Luffa which was an older vine growing slowly and blossoming. The inoculations were made by means of groups of needle-pricks on the blade of one leaf. The bacteria used were sticky. Four days later these plants, all of which were growing nicely, were re-inoculated, each on another leaf-blade, using sticky bacterial slime out of the vessels of cucumber- vine No. 222. Great quantities of the bacteria were used. All of the leaf-blades of vine No. 222, from which these inoculations were made, had wilted and the interior of the stem was gorged with the bacteria. The vines were re-inoculated to make infection doubly certain. A limited area of the surface was first thoroughly wetted with the sticky slime and then many needle-pricks were made into this area. The inoculations were made in the hothouse. (226.) Common Gourd. Twenty-eight days after inoculation (December 31), the plant had grown much and was blossoming freely, having shown no signs of the wilt. Three months after inocu- lation there was no general wilt. (227.) Common Gourd. The eighth day at 1 p.m., the pricked leaf had changed color in the terminal part of the blade and two-thirds of this blade hung limp (it was normal at 10 a.m.). The period of incubation was a few hours less than 8 days. Five days later the blade of the pricked leaf had wholly collapsed and part of it was yellowish. The next leaf below (the one which was re-inocu- lated December 7), was still normal. About 3 weeks after inoculation the wilted leaf had shriveled to the stem but none of the others showed any indications of wilt although the first internode up was only half an inch long and that next below was only 0.75 inch. January 10 (38 days after inoculation) the next four leaves above the pricked one and also one below had collapsed. Previous to this there had been no indication of the disease (except wilt of these four leaves for a few hours only on December 16, ascribed to lack of water), and I thought the plant had overcome the bacteria. Careful examina- tion of the stem in two places, i. c, just above and below the pricked leaf, showed no bacteria and the wilt was not accounted for. (228.) Common Gourd. The plant grew and blossomed freely. The inoculation produced no disease. WILT OF CUCURBITS. 263 (229.) Common Gourd. This plant was like the preceding. With the exception of vine 227, all the inoculations into Lagenaria failed to induce anything more than trivial local injuries. They were inoculated for the third time on January 7 and numbered 260-262 (q.v.). (230.) Balsam apple. The tenth day there were no signs of the wilt, but 3 days later both pricked leaves showed very small wilted areas. December 3 1 the pricked portions were dead and had narrow yellow borders, but the rest of the leaf was normal, green, and turgid and there were no indications of any secondary wilt. The plant was growing and had five times as much leaf-surface as when pricked. The inoculation failed to induce anything more than local injuries and the vine was re- inoculated January 7 and numbered 263 (q.v.). (231.) Balsam apple. Like the preceding. Re-inoculated and renumbered (264.) on January 7. (232.) Balsam apple. The seventh day there seemed to be a little wilt around some of the pricks first made. The following day there was a plain case of wilt with change of color to a dull green on the leaf-blade pricked December 3. About one-fourth of the outer part of the leaf had drooped. The thirteenth day the wilt was increasing slowly on the first pricked leaf and a wilt-spot of a few square millimeters had developed on the other pricked leaf (pricks of December 7). The twenty- eighth day the inoculated leaf which had shown distinct wilt where pricked had dried out. Half of this blade (the pricked part) was dead. The rest was a good green. The plant had grown well and had five times as much leaf-surface as when pricked. It was re-inoculated January 7, and numbered 265 (q.v.). (233.) Cucumis angaria. The sixth day there were no signs but the following morning both the pricked leaves and the one next above were badly wilted. The eighth day this vine had developed a bad case of the bacterial wilt. The thirteenth day the plant, which was a small one, had collapsed. (234.) Cucumis angaria. The sixth day there was distinct wilt on the blade of the leaf first pricked, also on the first leaf up and the first down (pricked December 7). It was 5! days since the first inoculation. Two days later this plant was badly wilted and the thirteenth day it had collapsed. (235.) Luff a acutangula. The eighth day the plant was normal except for a little yellowing around the groups of pricks made on December 3. The tenth day there were no signs of wilt. Three days later both leaves were yellowish around the pricks but there was still no wilt. On December 31 (28 days after inoculation) the pricked areas were dead and had yellow borders but there had been no wilt even of the pricked leaves. The vine had doubled its size since it was pricked. On January 7 it was re-inoculated and numbered 266 (q.v.). (236.) Cucurbita foetidissima. The sixth day the leaf inoculated on December 3 had wilted and changed to a dull green where pricked so that the terminal third of the leaf-blade hung down flabby. Twenty-five hours later the next leaf down (pricked December 7), was beginning to show a trace of wilt. The following day the blade of the first leaf up and of the first below the pricked leaves drooped. Thirteen days after inoculation (December 16) all the foliage had shriveled. The plant was a small one. Two days later it was brought in and examined microscopically. Cross-sections of the stem were not noticeably sticky but the interior of the vessels and surrounding parenchyma were gorged with bacilli. At least one-tenth of the rods were noticeably larger than the rest. The largest were estimated to be ten times larger than the smallest — longer and broader. None were clearly motile. Material saved in alcohol. (237.) Cucurbita foetidissima. The sixth day there were no signs of the disease but 2 days later the blade of the leaf inoculated on December 3 snowed a trace of wilt at the apex where it was pricked and also a slight change of color. The period of incubation was nearly 8 days. The following after- noon the blade of the leaf first pricked was badly wilted. The next morning the blade of the second pricked leaf was half wilted. Three days later (13 days after inoculation) all the foliage of this vine which was a small one, was shriveled. The fifteenth day it was brought in and examined microscopi- cally. The vessels and surrounding parenchyma were crowded with bacilli which were like those found in the preceding. The cross-sections were not noticeably sticky. Material saved in alcohol. (238.) Cucurbita foetidissima. The sixth day the blade of the leaf which was first pricked had changed to a light green and wilted in tiny spots around three of the four groups of pricks. Twenty- five hours later the terminal half of the leaf first pricked had changed color and was drooping. The following day the first leaf above and the first below were wilted, the petioles of the lower two leaves being involved in the flaccidity. The thirteenth day the foliage had shriveled. The vine was a small one. Remarks. — Cucumis angnria and Cucurbita foetidissima contracted the disease promptly and proved as sensitive as cucumber or muskmelon. Luffa, Momordica, and Lagenaria were resistant. 264 BACTERIA IN RELATION TO PLANT DISEASES. Inoculations of December 7, 1895. Inoculations were made at 2 p. m., into gherkin (Cucumis anguria), eow-pea {Vigna), muskmelon {Cucumis melo), and cucumber {Cucumis saiivus), using sticky bacterial slime out of the vessels of cucumber-vine No. 222. Great quantities of the bacteria were used. A limited area of the surface was first thoroughly wetted with the sticky slime and then many needle-pricks were made into this portion. (239.) Gherkin. Four days after inoculation (between 10 a. in. and 1 p. m.) the tip of the pricked leaf changed color and commenced to wilt. The ninth day the plant, which was a small one, had collapsed. (240.) Cow-pea. The twenty-fourth day there was no wilt or death of the tissue. Up to March 4, there had been no result from the inoculation. (241.) Muskmelon. The third day there was a small wilted spot including each of the five sets of pricks. An enormous number of bacteria were pricked in and this was supposed to account for the unusually short time between inoculation and the first signs, i.e., only 68 hours. There was little increase of the wilt, however, during the next 24 hours. On the sixth day signs were uncertain. The twenty-fourth day the leaves were spotted and the vines were not thrifty but it was doubtful if the bacteria were still alive. Up to March 4 there had been no other result from the inoculation. (242.) Cucumber. This was a large plant. The fourth day the vine showed no signs of the wilt. Two days later the case was doubtful but the ninth day the blade of the pricked leaf was dry-shrivel- ing. The twelfth day the leaves above and below were turgid. Twenty-four days after inoculation (December 31), the stem was collapsing, all the leaves having shriveled some days ago. The stem was now cut open and the interior found to be full of a white sticky mass of bacteria. Inoculations into 244 and other vines were made direct from the interior of this plant. A loop of slime taken from the interior of 242, with bacteriological precautions, and spread on slant agar, yielded a pure culture which was used to inoculate No. 251, etc. (243.) Cucumber. Thiswas a big plant. The fourth day (1 p. in.) the inoculated leaf had changed to a pale green and had begun to wilt around one of the five groups of pricks. The period of incubation was nearly four days. Two days later one-fifth of the pricked blade hung flabby and had become a dull green. The ninth day the blade of the pricked leaf was shriveling. The twelfth day three leaves above the pricked one were wilted. They had been turgid the day before. The first leaf below was normal. The twenty-fourth day the stem was collapsing. All the leaves had shriveled some days ago. Remarks. — The cow-pea did not contract the disease. The three closely related species of curcurbits contracted the disease. Inoculations of December 31, 1895. Three cucumbers {Cucumis sativus) and four squashes {Cucitrbita maxima) were inoculated in the hothouse, at 9 a. m., with a white sticky mass of bacilli from the interior of cucumber- vine 242 (cucumber-strain). The bacterial slime was pressed out on the leaves and then pricked in with a sterile steel needle. Many pricks were made. In some cases the bacteria were put on the dorsal side, in others on the ventral side of the leaf. Each of the cucumbers was inoculated on one leaf, each of the squashes was inoculated on one green cotyledon. (244.) Cucumber. The eighth day (10 a. m.) there was wilt on one margin of the inoculated leaf extending outward from three groups of pricks — 5X 1 cm. The blade of this leaf was 4 inches broad. There had been no wilt the preceding afternoon. The tenth day the whole of the pricked blade was wilted. ( (ther leaves were normal. The sixteenth day the petiole of the pricked leaf was still turgid to the tip hut the blade had dry-shriveled. All the leaves above the pricked one, four in number, were wilted. The blade of the first leaf down was also wilted but the others were turgid. The twenty-fifth da) the apical part of the stem had shriveled and the disease was slowly passing downward. The vine finally shriveled to the ground. (245.1 Cucumber. There were no si^ns until the ninth day. Then the apex of the pricked leaf was wilted. The following day the apical half of the leaf had changed to a dull, faintly yellowish green and had wilted, the wilt clearly commencing in the pricked areas (see figure 74). The rest of the leaves were normal. ( )n the sixteenth day the blade of the pricked leaf was shriveled; the petiole was turgid except toward the top where it was a trifle flabby. The upper five leaves were wilted and also the WILT OP CUCURBITS. 265 two next below the pricked leaf. The rest were normal. The twenty-fifth day the apical 10 inches including three internodes below the pricked leaf were dry-shriveled. The wilt was slowly extending downward. Another big leaf-blade, the fourth down, had drooped the preceding day. Farther down were five good leaves separated by long internodes. The stem was horizontal. The whole vine finally shriveled to the ground. (246.) Cucumber. The eighth day (10 a.m.) there was wilt of the apex and margins of the pricked leaf, extending outward mostly from the groups of pricks, involving 10 to 12 sq. cm. The leaf was 4 inches broad. There had been no wilt up to the preceding afternoon, therefore the period of incuba- tion was nearly eight days. The tenth day the terminal three-fourths of the pricked blade had wilted and was a dull yellowish green (it was more green than yellow but looked faded). The petiole and remaining leaves were normal. By January 16 the blade of the pricked leaf had shriveled. The upper half of the petiole was flabby and the extreme upper end was shriveling. The terminal six leaves (6 inches of stem) had wilted and also the blade of the first leaf down, the separating internode being 4 cm. long. The rest were normal. The twenty-first day the upper half of the vine had wilted. It was now brought into the laboratory and by direct transfer four slant agar-cultures were made from it (tubes 1 to 4, January 21, 1896) as follows: No. 1 frombase of wilted part of stem (interior); No. 2 from interior of a young shriveling fruit near the top of the vine; No. 3 from middle of wilted part of the stem which had begun to shrivel; No. 4 from interior of stem not far from the lowest external sign of wilt (in leaf) and where the stem was sound externally but sticky within. Nos. 1 and 4 yielded pure cultures of Bacillus tracheiphilus. (247.) Squash. Groups of pricks were made on a big cotyledon. Up to March 4 there had been no general wilt. The pricked cotyledon was dead and the pricked part was thicker than the rest as if from develop- ment of cork-tissue. (248.) Squash. One of the cotyledons was pricked. There was no result from the inoculation. Even the cotyledons did not wilt. There seemed to be cork in and around the pricked area. (249.) Squash. The pricks were made in one of the cotyledons. Up to March 4 (63 days) there were no general signs re- sulting from the inoculation, although 22 days after inoculation there was a decided wilt at the tip of the cotyledon which con- tinued for several days extending very slowly from the group of pricks outward. (250.) Squash. This vine was also pricked on one of the cotyledons but with no result. March 4 the pricked part looked as if cork-tissue had been formed there. Remarks. — In the cucumbers the bacterial wilt progressed upward faster than down- ward, i. c, two or three times as fast. On the sixteenth day in the squash all the pricked cotyledons were large, thick and green. One of the pricked cotyledons wilted at the tip after 22 days. One of the cotyledons, collected March 4 (probably from 247 or 249) was afterwards infiltrated with paraffin and sectioned. Bacteria were present in some of the bundles and some of the latter were a little disorganized but not much. The impression one gets from these sections is that the bacteria have multiplied in the vessels very slowly. A definite corklayer was not made out. *Fig. 74. — Leaf of Cucumis sativus (plant No. 245) inoculated with B. tracheiphilus and shaded to show progress of wilt. The needle-pricks were made Dec. 31, 1895. First sign of disease appeared at apex of leaf on ninth day. About 2^1 hours later wilt hnd pvtpnHpH n^ indir.ii nl hv liVhtpr shaHinp- Dr.nwn bv the writer. Fig. 74/ On Drawn by the writer. 266 BACTERIA IN RELATION TO PLANT DISEASES. Inoculations of January 7, 1896. A set of inoculations was made in the hothouse at noon on cucumber {Cucitmis sativus), pumpkin (Cucurbita pepo), squash [Cucurbita maxima), cow-pea (Vigna), tobacco (Nico- tiana tabacum), common gourd (Lagenaria vulgaris), balsam apple (Momordica balsamina) , and the vegetable sponge (Luffa acutangula) . The bacteria used were from a thin, nearly colorless, wet-shining, very sticky growth (cucumber-strain) covering nearly the whole surface of a slant agar (tube No. 4, December 31, which was inoculated from the interior of vine 242 and yielded a pure culture) . The inoculations were made with a sharp-pointed steel-needle and the culture was so sticky that an enormous number of bacteria must have been inserted into the leaves. (25 1 .) Cucumber. This was a large vine. Many pricks were made on one leaf. The wilt appeared on the seventh day in the pricked area to which it was then confined. Two days later the wilt had moved slowly from the pricked area, and now involved about one-seventh of the blade — the terminal portion from the pricked area to each side and outward. The wilt had extended more rapidly up than down. The rest of the vine was normal. The eighteenth day the terminal part of the stem had dry- shriveled and the wilt was slowly passing downward. The whole vine finally shriveled to the ground. (252.) Cucumber. No pricks were made but a big loop of the sticky bacterial slime was put into a fresh pistillate flower on and below the stigmas. The flower was near the apex of the stem. The eighteenth day there was some possibility that the bacteria had passed into the fruit. It had changed color throughout, i. c, it had become a dirty green. The fruit was 0.5 to 0.75 inch long at this time. There had been no wilt of the leaves as yet. Up to March 4, however, there were no further signs. (253.) Pumpkin-seedling. Many pricks were made on a cotyledon. The inoculation failed. Not even the whole of the cotyledon wilted. (254.) Pumpkin-seedling. Many pricks were made on the blade of the tender first leaf. The thirteenth day the pricked part of the leaf was wrinkled and yellowish but not wilted. Five days later one side of the pricked leaf had wilted. By March 4 (56 days) the pricked leaf had wilted to the stem but there were no other signs of disease and the plant was very thrifty. (255.) Pumpkin-seedling. Many pricks were made on the small, delicate first leaf. The eighth day the pricked apex of the blade was wilted and the following day the wilt involved about one- fourth of the leaf. Four days later the pricked part of the leaf had dry-shriveled and the wilt was extending, although the middle basal part of the leaf was still turgid as was also the petiole. Twenty- four hours later the upper part of the petiole had drooped over. The eighteenth day the pricked leaf had wilted and shriveled nearly to the stem and the cotyledons were beginning to droop. The whole vine ultimately shriveled to the ground. (256.) Pumpkin-seedling. Many pricks were made on the tender, partially developed first leaf. The thirteenth day about two-thirds of the blade of the pricked leaf had wilted, i. e., the pricked apical area and the margins nearly to the petiole. Five days later the pricked blade and over half of the petiole had shriveled. The whole vine shriveled to the ground after a time. (257.) Winter-squash-seedling. Many pricks were made on one of the big green cotyledons. The thirteenth day there was a distinct wilt and loss of color extending from one group of pricks to the edge of the leaf — about 1 sq. cm. Five days later there was a bad droop of the pricked cotyledon. Up to March 4, however, there had been no general wilt and the plant was still thrifty. Not even all of the pricked cotyledon had wilted. (258.) Cowpea. This plant was 6 inches high and was just developing the third true leaf. Many pricks were made on the apex of one of the first leaves. There was no result from the inoculation. (259.) Tobacco var. Little Oronoco. Many pricks were made on the apex of the blade of a bright green leaf, 3X1.5 inches. There was no result from the inoculation. (260.) Common gourd (No. 226 of December 3, now inoculated for the third time). Many pricks were made on the apex of the leaf-blade midway up. The seventh day there was a slight flabbiness at the tip of the pricked leaf. Two days later there had been no decided change. Up to March 4 there was no general wilt. (261.) Common gourd (No. 228 of December 3, now inoculated for the third time). Many pricks were made on the apex of a leaf-blade midway up on the stem. The seventh day there was a slight flabbiness at the tip of the pricked leaf. The ninth day there was no decided change and the case was considered a doubtful one. ( >n March 4 the plant was still living and there had been no general wilt. WILT OF CUCURBITS. 267 (262.) Common gourd (No. 229 of December 3, now inoculated for the third time). Many pricks were made on the blade of a leaf half-way up the stem. The ninth day there had been no wilt. Up to March 4 there was no general wilt. (263.) Balsam apple (No. 230 of December 3, now inoculated for the third time). Many pricks were made on a leaf half-way up. This was a small plant on December 3, when it was first inoculated but now it was about 16 inches long with 17 leaves. The others had grown nearly as much. The seventh day there was wilt around the pricks. This had appeared probably the day before. Two days later the wilted area extended out a short distance from the pricked area. Up to March 4 (56 days) there had been no general wilt and the plant was still growing slowly. (264.) Balsam apple (No. 231 of December 3, now inoculated for the third time). Many pricks were made on a bright green leaf half-way up the stem. The seventh day there was wilt around the pricks which, however, increased little if any during the next 2 days. Up to March 4 there had been no general infection. Not even all of the pricked leaves wilted: They only became somewhat yellowish. (265.) Balsam apple (No. 232 of December 3, now inoculated for the third time). Many pricks were made on the blade of a bright green leaf growing midway of the stem. The seventh day there was a wilted area around the pricks. Two days later there was little, if any, increase in this area. On March 4 the plant was still healthy. (266.) Luffa acutangula (No. 235 of December 3, now inoculated for the third time). Many pricks were made on a leaf-blade near the apex. On January 16 the plant was healthy. Up to March 4 there was no general wilt. Remarks. — Cow-pea, tobacco, and Luffa refused to take the disease. Pumpkin, squash, balsam-apple, and gourd showed local signs but even the pricked leaves did not succumb as a whole and no secondary or general signs appeared, i. c, there was no wilt outside of the pricked leaf except perhaps in Nos. 255 and 256. Cucumber (No. 251) contracted the disease. Inoculations of January 21, 1896. Inoculations were made in the hothouse at 10 a. m., on Cucumis sativus, Ciicurbita californica and Cucumis anguria. The bacillus was taken from tube 6 December 31, the bulk of which had been used up the preceding day in inoculating fermentation-tubes and making new agar-cultures. The bacteria were introduced by needle-pricks. The plants were under observation until March 4. (267 and 268.) Cucumis sativus. These were old vines. Many pricks were made on a leaf-blade. No signs appeared. (269.) Cucurbita californica. Many pricks were made in the fourth leaf-blade above the cotyledons. The vine was a young one. There was no result from the inoculation. (270 and 271.) Young Cucumis anguria. Many pricks were made on a leaf-blade. There was no result from the inoculation on either plant. Remarks. — These five inoculations were made into plants of three susceptible species. All failed. The reason for this failure is to be sought in the nature of the culture used. What was then puzzling is now perfectly plain. The culture was 21 days old, and was on stock 93 b, an agar to which sugar had been added. On the same date as these plant inocula- tions, four fermentation-tubes were inoculated copiously from the same culture and all failed (7 days test) showing that the culture was dead. The body of the fluid in these tubes consisted of slightly alkaline (litmus) peptonized beef-broth free from muscle-sugar. One of these fermentation tubes contained saccharose, another maltose, another lactose and the fourth dextrine, each of these very suitable foods being added in the proportion of 0.2 per cent. The early death of the culture was undoubtedly due to the injurious action of its own by-products, i. e., of acids derived from the decomposition of the sugar, this organism being extremely sensitive to acids. 268 BACTERIA IN RELATION TO PLANT DISEASES. Inoculations of February 26, 1896. A set of inoculations was made in the hothouse at 2 p.m., from a white, wet-shining, sticky culture of Bacillus tracheiphilus on steamed carrot (tube 2 February 18, from 1, February 13 which was a potato culture made from a fermentation-tube (No. 1, January 20) containing cane-sugar. The plants inoculated were: Benincasa ccrifera, Cucurbita focti- dissima, Cucurbita califoruica, cucumber, muskmelon, watermelon and Datura stramonium. Numerous bacteria were put in with each inoculation, which was made with a sharp- pointed steel needle. All of the plants were examined on February 28 and February 29, and were free from disease. (272.) Benincasa ccrifera. This was a small plant having three leaves besides the green coty- ledons. Many pricks were made on the middle leaf . The plant appeared healthy on March 5. The first signs were noted on March 6 at 2 p. m. (end of the eighth day) at which time there was a slight wilt extending outward from the pricks to the margin. The following day fully half of the pricked leaf had wilted. Two days later the whole of the blade of the pricked leaf had wilted, as also that of the first leaf above and below. The sixteenth day the leaves were badly shriveled, but the stem was normal. Material was saved in alcohol. On microscopic examination enormous numbers of bacteria were found in the vessels (slide No. 202). (273.) Cucurbita foetidissima. This vine was grown from seed planted October 11, and at the time of inoculation had eight good leaves. Many pricks were made on the under side of an upper leaf. The fifth day (9 a. m.) the leaf looked suspicious and at ih2om p. m. there was distinct wilt at the tip and a dull green color where pricked. The following day there was little change. The next day the pricked area had changed to a whitish green and the terminal one-fifth of the blade hung flaccid. The following day about one-third of the leaf was wilted and 2S hours later (close of eighth day) there was a bad collapse of two-thirds of the pricked blade. The beginning of the eleventh day the whole of the blade of the pricked leaf had wilted but the petiole was still rigid. The blades of the first and second leaf up were now inclined to droop. Two days later the blade of the first leaf down was wilted. All the petioles were still rigid. The sixteenth day all the leaf blades had shriveled. The stem and petioles (lower two-thirds) were normal. The plant was put into alcohol. On micro- scopic examination the bundles of the petiole of the inoculated leaf (fig. 77) were found filled with bacteria and badly disorganized (slide No. 254). The bacteria also extended into the bundles of the fleshy root but here the disorganization was less. (274.) Cucumber (Cucumis sativus). This was a thrifty young plant. Many pricks were made on the tip of an upper leaf. The fifth day (9 a. m.) there was wilt and change of color in an area of 10X3 mm., along one side of the pricked area which was about 10X 10 mm. in diameter. At 1 h 20™ p.m. of the same day the wilted area was twelve times as large. The following morning there was little change, but 24 hours later the terminal eighth of the leaf was drooping. The next day the change of color in the wilted area was more decided but the latter had not increased much. At the end of the ninth day there was very striking wilt confined to the terminal half of the pricked leaf, a wedge-shaped area 7.5 em. long by 2.5 cm. wide (in the widest portion). It was dull green and the tip was drooping. Twenty-five hours later the whole of the blade of the pricked leaf had wilted. The rest of the vine was normal. On March 9, the blades of the first two leaves above and the first two below had wilted but the petiole of the pricked leaf was still rigid. Five days later (seventeenth day) the vine was badly wilted and was pulled up for microscopic examination. The stem was green and turgid but its vessels contained a stick}' bacterial slime which strung out in long, delicate threads. The middle part of the stem was saved in alcohol. (275.) Cucumber. This was an old plant. Many pricks were made on the blade of an old, upper, whitish leaf. On March 3, at 10 a. m., there were no signs but 24 hours later half of the leaf was flabby and drooping, on the pricked side. It was now nearly 7 days since the leaf was pricked. The following day the whole of the pricked blade was dry shriveled and the tip of the petiole was slightly flabby. On March 6 (end of ninth day) the petiole had shriveled. The rest of the vine was normal. The following day there were constitutional signs. By 10 a.m. of March 9 everything had collapsed except the green stem and a few small basal leaves. (276.) Cucumis melo var. dudaim. This was a small plant. Many pricks were made on one li ii blade. The fifth day there was slight wilt in the pricked portion and running out to one margin of the blade. The following day there was little change, but 24 hours later the whole pricked leaf was droi iping. This, however, was favored by a dry soil because, on watering, the leaf recovered its turgor in the afternoon with the exception of the tissue immediately around the pricks. The next morning the pricked leaf was turgid with the exception of a wedge-shaped piece extending from the WILT OF CUCURBITS. >69 Fig. lb: pricks to the margin. The ninth day (2 p.m.) the wilt involved most of the pricked side of the leaf (fig- 75) and 25 hours later all of the pricked leaf had wilted but the petiole, while the other leaves were normal. The eleventh day all the leaves had collapsed. The sixteenth day the vine had shriveled to the ground. (277.) Cucumis melo var. dudaim. This was a small plant. Many pricks were made on the blade of one leaf. The sixth day ( 10 a. m.) there was a slight wilt and change of color in the center of the pricked area. Twenty-four hours later two-thirds of the pricked leaf was drooping. The eighth day the pricked leaf was turgid with the exception of a wedge-shaped area extending from the pricks to the tip and involving about one-eighth of the blade (fig. 75). The next afternoon there was only a little increase of the wilt, but 25 hours later all of the blade of the pricked leaf was wilted. The petiole was still turgid. The blades of the first three leaves up now showed a slight droop. The twelfth day all the leaves had collapsed and the sixteenth day the vine had shriveled to the earth. (278.) Benincasa ccrifcra. Many pricks were made on one of the leaf-blades of a small plant. On March 3 there were no signs, but 24 hours later there seemed to be a slight droop of the pricked portion. On March 5 there was no clear evi- dence of the wilt, but the following afternoon (end of the eighth day) there was change of color and distinct wilt. These signs were confined to an area of about 1 sq. cm. from the pricks to the tip, and the most of the pricked leaf was still normal. Twenty-five hours later the wilt was spreading slo wh- in the blade of the pricked leaf. The twelfth day six leaves besides the pricked one (part above it and part below) were wilted, some badly. Four days later the vine had shriveled to the ground. (279.) Watermelon (Citrullits vulgaris). This was a small plant. Many pricks were made on one leaf-blade. On March 3, at 10 a. m., there were no signs of the wilt, but 24 hours later the pricked leaf had changed color and wilted from the pricked area outward to the tip, about one-third of the leaf being affected. The following day the leaf had recovered its turgor except a very small wedge at the tip beyond the pricks. The bulk of the pricked area seemed normal. The next afternoon During the next 25 hours there was a slow spread of the wilt and change of color (whiter) in the blade of the pricked leaf. Two days later about one-fourth of the pricked leaf had wilted and dried out. The rest was normal. Up to the seventeenth day there was no change. Four days later the whole of the blade of the pricked leaf, which was a small one, had wilted. The petiole and remainder of the plant were normal. (280.) Cucurbita californica. Many pricks were made on one leaf-blade which was 2 inches across. The leaf was almost exactly the shape of a leaf of English Ivy. On March 3, at 10 a. m. there was a slight wilt in the pricked area. Twenty-four hours later about one-third of the pricked leaf had wilted and the following morning fully one-half of the inoculated leaf had succumbed to the wilt (see fig. 76). The next afternoon (March 6) the whole of the pricked leaf had collapsed, also the first leaf below, the neighboring cotyledon, and the first three leaves above. The stem was turgid as were also the fourth and fifth leaves up and the other cotyledon. Twenty-four hours later the signs were much aggravated. The twelfth day the leaves had collapsed including the petioles and the terminal part of the stem. The plant was now removed and put into alcohol. On microscopic examination great numbers of bacteria were found in the vascular bundles of the stem. (2S1.) Datura stramonium. This was a young thrifty plant. Many pricks were made on a leaf- blade. A great quantity of bacteria were put into the leaf, but up to the twenty-first day the plant was growing finely and there were no signs of the wilt. *Fig. 75. — Left: Leaf of plant No. 276 (Cucumis melo vox. dudaim), ninth day after inoculation with Bacillus tracheiphitus, shaded part wilted. Right: Leaf of inoculated plant No. 277 (Cucumis melo var. dudaim) on eighth day after Bacillus tracheiphilus was introduced by needle-pricks. The first wilt was a little earlier (central dark shading) . fFio. 76. — Leaf of Cucurbita calif ornica (plant No. 280) inoculated with B. tracheiphilus. The needle-pricks were made Feb. 26, 1896, and the wilt appeared in the deeply shaded part March 3. During the next two days it involved over half the leaf-blade as shown by the lighter shading. The whole plant collapsed on the twelfth day. there was no decided increase. Fig. 76.t 270 BACTERIA IN RELATION TO PLANT DISEASES. Remarks. — Datura stramonium resisted. Local signs were obtained on the watermelon but there was no general infection of the plant. Local and then constitutional signs appeared on the cucumbers, on Cucurbita foetidissima, C. calif ornica, Benincasa cerifera and on the little melon, Cucumis melo var. dudaim. The old and young cucumber proved equally subject to this desease. The bitter plant which I have called Cucurbita calijomica was grown from seeds sent to Mr. Gilbert Hicks by Prof. J. W. Tourney of Arizona. It came to me unnamed and I had much difficulty in classifying it. The plant was finally determined for me by Dr. J.N. Rose of the U. S. National Herbarium. Inoculations of April 4, 1896. Four vines of Melothria scabra, three of cucumber, one of Echinocystis lobata, one of watermelon and two of Cucumis erinaceus were inoculated with a pure culture of Bacillus tracheiphilus taken from tube 1, March 30. A big loopful of the bacterial slime was put on one leaf of each plant and pricked in with numerous fine punctures, using a small sharp steel needle. In some cases an additional loop of the slime was afterwards put on over the pricks which were protected from direct sunshine. The rods in this tube were mostly in a state of active motility as determined by examination in a hanging drop. Inoculations in each case were made on the blade of the leaf and were very thorough. (282.) Melothria scabra (from Mexico). Up to April 13 there were no signs. The eleventh day there was a yellowing of the tissue about the pricks, but no wilt of the blade. Two days later there was little change (the weather for the past six days had been very hot). The twenty-fourth day the pricked leaf had shriveled but the rest of the plant was normal. On June 16 the plant was still alive the inoculation having failed to kill it. On June 25 (82 days) there was still no result other than the local injury. (283.) Melothria scabra. There were no signs until the eleventh day. At that time a very little of the tissue in the pricked area was dead (1 to 2 sq. mm.) but there was no general wilt of the leaf. Two days later the pricked leaf was yellower but not wilted. The twenty-fourth day the pricked leaf had shriveled but none of the others showed any trace of the wilt. On June 1 6 the plant was still living. The inoculations failed to induce constitutional signs. (284.) Melothria scabra. On April 13 the plant was normal. The eleventh day the leaf-blade was yellowish green and puffed out where pricked. Two days later the pricked leaf was yellow around the pricks but not wilted. The twenty-fourth day the pricked leaf which had shown itself very resistant at first, had shriveled. The other leaves were normal. June 16 the plant was still living and did not show any constitutional signs. (285.) Melothria scabra. This plant behaved like the preceding. The inoculation did not harm the plant beyond the pricked leaf. (286.) Common Cucumber. No record earlier than April 13. On that day there was wilt on one side of the leaf around the pricks. Two days later the whole blade of the pricked leaf was droop- ing. The thirteenth day the whole of the pricked leaf-blade had wilted, also two leaves above and one leaf below. The petioles were turgid. The twenty-fourth day the plant was dead. (287.) Common Cucumber. No record earlier than April 13. On that clay there were no signs of the disease, but 2 days later there were a few square centimeters of wilted tissue in and around the pricks. The thirteenth day the whole of the pricked blade was wilted. The leaves above and below were normal. The twenty-fourth day the stem was still green, but the leaves had wilted. (288.) Common Cucumber. The ninth day the pricked leaf was normal, but 2 days later there was wilt of a few square centimeters in and around the pricked area. The thirteenth day the whole of the pricked leaf had wilted. The leaves above and below were normal. The twenty-fourth day the plant was dead. (289.) Echinocystis lobata. No record earlier than April 13. The afternoon of that day there was wilt of the tip of the leaf, i. e., of the tissue in and around the pricked area. There was only a slight change the following noon. The eleventh day most of the pricked leaf was turgid. The wilted portion had dried out and apparently the disease had come to a stop. Two days later there was no increase of the wilt and the greater part of the pricked leaf was normal. The twenty-fourth day the plant was growing rapidly and had recovered. On June 16 the plant was still living. It had made a long growth and blossomed. Only a small part of the pricked leaf succumbed. WILT OF CUCURBITS. 27 1 (290.) Common Watermelon. The ninth day there was wilt of the tip of the leaf beyond the pricks and reaching down to them. There was only a slight change the following noon, and the eleventh day the wilt seemed to have stopped. The tip of the leaf beyond the pricks was dead (a few sq. mm.) but between them the tissue was still living, although a yellowish green color. Two days later all but about one-twenty-fifth of the pricked leaf-blade was normal. There had been little change since the eleventh day. The twenty-fourth day the wilted portion of the leaf was brown and dry but there had been no increase of the disease. (291.) Cucumis erinaceus. The ninth day there was wilt in the pricked area but the remainder of the leaf-blade was turgid. The following noon there was no change. The eleventh day there was a small dead patch in the pricked portion, but no general wilt of the leaf. The terminal portion had recovered its turgor. On April 17 the terminal one-sixth of the leaf in and beyond the pricked por- tion was wilted, but the rest of the plant was normal. The twenty-fourth day the plant had recovered and was growing. June 25 the plant had fully recovered and was making a fine growth. (292.) Cucumis erinaceus. The ninth day at 9 a.m. the terminal part of the pricked leaf-blade had wilted and in the afternoon of the same day the wilt involved the whole leaf-blade. The follow- ing day the blades of the two leaves next above and the two next below the pricked leaf had wilted. The petioles were turgid. The eleventh day the wilt was but little if any worse than on the preceding day and the petioles were still rigid. Two days later the blades of six leaves were badly wilted or shriveled. The petioles were rigid and the terminal leaf and one or two at the base were normal. The twenty-fourth day several leaves were dead but the plant as a whole seemed to be recovering and had apparently thrown off the disease. The weather was cool. June 16 the plant had recov- ered and had a good color but was a trifle stunted. June 25 the plant had entirely recovered and was making a good growth. Remarks. — Watermelon, Echinocystis lobata, and Melothria scabra developed only slight local signs. In one vine of Cucumis erinaceus only slight local signs appeared. In another vine of the same species, constitutional signs followed the local wilt, but finally the plant threw off the disease and recovered. Cucumbers first showed local signs (in the pricked parts) and then general signs ending in death. Inoculations of June 16, 1S96. Inoculations were made in the hothouse at 11 a. m., on Cucurbita foetidissima, Apodan- thera undulata, Cucurbita pahnata (?), Cucurbita digitata, Trichosanlh.es cucumeroides, Passiflora incarnata, Cucumis melo var. dudaim, Citrullus vulgaris, Echinocystis lobata, and Cucumis erinaceus. All the plants were infected from culture No. 16, May 27, a tube of peptone-water (cucumber-strain). This culture would have been old and exhausted long before but for the fact that the organism made no growth in it for the first 16 days, having been kept all of this time in the ice-box at temperatures ranging from 6° to io° C. At 4 p. m., June 12, it was removed from the ice-box and put at room-temperature (250 or 260 C). In 48 hours it showed faint clouding and for the 36 hours preceding its use for inoculation it had been well clouded with good rolling clouds on shaking. When examined in a hanging drop the culture was seen to contain numerous rods in process of division. Some of the bacteria were actively motile, darting ahead long distances ; others, slowly tumbling ; others, stationary. There were a few involution forms. A copious quantity of the fluid was lifted out on a sterile platinum loop, placed on the surface of a clean leaf and pricked in with a sterile steel needle. The loop and needle were flamed after each operation and used again as soon as cold. In most instances a second loop of the culture was rubbed over the numer- ous delicate pricks. At the time of the inoculation the hothouse was cool, there was no wind and the sky was overcast. There could not have been a better day nor apparently a more suitable culture. (293.) Cucurbita foetidissima (grown from seeds of Tourney's second sending. Looks a little different from the first sending, as if the plant were a variable one). Up to the ninth day no signs had appeared, but 2 days later (1 p.m.) there was a very slight wilt of the pricked leaf at one edge of the pricks. July 14 the plant was entirely dry-shriveled as a result of the inoculation. (294.) Cucurbita foetidissima. This plant must have contracted the disease as early as the fourtli or fifth day. On the sixth day the blades of five leaves were wilted, i.e., that of the pricked leaf and of two leaves above and two below. The seventh day the pricked leaf was wholly shriveled, the 2J2 BACTERIA IN RELATION TO PLANT DISEASES. petiole being still rigid. Two days later the plant was very sick, all of the leaves, seven in number, having wilted. The eleventh day the stem was still green and turgid. July 14 the plant was entirely dry-shriveled as a result of the inoculation. (295.) Cucurbita foetidissima (Tourney's first sending to Mr. Hicks). The sixth day (3 p.m.) there was no sign of the disease but 23 hours later the leaf on the pricked side had changed to a dull green and was wilted over an area of about 10 sq. cm. from the pricks outward, up, down, and inward to the midrib. The rest of the plant was uninjured. The ninth day the whole of the pricked leaf had shriveled except the petiole which was turgid. For distribution of the bacteria in the petiole of such a leaf see fig. 77, and for a detail from the same see fig. 78. The wilt now also showed on the blades of three other leaves — the first down and the first two up. Two days later all but one leaf was wilted, the stem, however, was still turgid and green. On July 14 the plant was dry- shriveled, with the exception of the base of the stem which was green. (296.) Apodantkera undulata (From Tourney, Tucson, Arizona). The pricked leaf was examined frequently but there was no marked result from the inoculation. On July 14 the plant was healthy. Part of the pricked area had dried out and most of the leaf- blade which was pricked was yellowish and yellow-green, but it had never shown any tendency to wilt. (297.) Apodantkera undulata. The seventh day all the foliage was slowly drying out but not as a result of the inoculation. Two days later there was a slight wilt (3X4 mm.) in the center of the pricked part. This portion later became dried out and brown but the plant did not contract the disease and on July 14 the rest of the pricked leaf was normal. (298.) Apodantkera undulata. Plant examined June 23, 25, 27, and July 3. There was no result from the inoculation. Up to July 14 the pricked leaf had not wilted and the plant was healthy. (299.) Cucurbita palmata? (Seeds received from Tourney: Said to have been collected in California). Up to July 14 there had been no wilt and the plant was growing rapidly. (300.) Cucurbita palmata (3). On June 25 the plant was normal. The eleventh day the pricked blade was yellow and shriveling, but 28 days after the inoc- ulation there was still no general wilt. (301.) Cucurbita palmata ( ?). Up to June 27 the pricked leaf showed no signs. On July 14 the pricked leaf-blade had wilted and some of those above it were yellow, but it was doubtful whether this was due to the disease because there was a line growth of healthy vine beyond the pricked leaf and no good leaves below it. (302.) Cucurbita digitata (From Tourney. The plant identified as Cucurbita calif ornica looks and tastes something like this). The seventh day there were tiny dead spots in the pricked area . 77. — Crov secti leaf-stalk of Cucurbita foetidissima, the wild gourd of the western plains of the United showing vascular system occupied l>\ Bacillus tracheiphilus. Plant grown in a hothouse in Washington from led in Arizona Inoculation from a pun culture by means of needle pricks in blade of leaf. Petiole fixed ili ohol, infiltrated in paraffin, sectioned on microtome, stained in carbol-fuchsin, and differentiated in 50 per cent alcohol Drawn from section with aid of Abbe camera. Slide 254 D 1. F,g. 77." WILT OF CUCURBITS. 273 but 4 days later there was no wilt. On July 14 the pricked leaf was shriveled but the rest of the vine was normal. (303.) Cucurbita digitata. Wilt began on the fifth or sixth day in the middle of the pricked part. The seventh day in the central part of the pricked area the tissue for 0.7 sq. cm. was dead and yellow- white. Outside of this was a narrow border of freshly wilting tissue, but five-sixths of the leaf was still normal. Two days later the whole of the pricked blade and the upper two-thirds of the petiole had shriveled. The eleventh day the second leaf was beginning to yellow and droop. On July 14 the pricked leaf and the first leaf up had shriveled. The rest of the vine was normal. (304.) Trichosanilies cucumeroides (from Agr. college in Japan). The seventh day the tip of the pricked leaf was wilting but 2 days later most of the pricked leaf was normal and the wilt seemed to be dying out. The eleventh day a fresh part of the leaf had begun to wilt. On July 14 the whole of the pricked leaf had shriveled. The rest of the vine was normal. (305.) Trichosanilies cucumeroides. The pricked leaf showed no signs up to June 25. The eleventh day about one-third of the apical (pricked) portion of the inoculated leaf-blade had wilted but was not yet dry. On July 14 the pricked leaf had shriveled, but the rest of the vine was normal and was growing rapidly. (306.) Trichosanilies cucumeroides. There was no result from the inoculation other than local injury which did not appear until after June 27 (eleventh day). On July 14 the pricked leaf had dry-shriveled but the remainder of the vine was normal and growing rapidly. (307.) Passiflora incarnala. The seventh day a whitish callous had formed around each of the pricks and there was no wilt. Up to July 14 there had been no result even in the pricked leaf. (308.) Passiflora incarnala. Like the preceding. No result. There was no wilt or change of color even in the pricked leaf. (309.) Cue tint is niclo var. dud aim (ripe fruit yellow and like a small round gourd; delightful odor; taste like muskmelon). The inoculation was made in an old leaf. The seventh day there was a distinct wilt in and around the pricked area. This had begun in a small way the preceding day. Two days later about one-fourth of the pricked leaf-blade was wilted. The blade of this leaf measured 3.5 X3.5 inches. On the eleventh day the whole of the pricked leaf-blade was dry-shriveled. On July 14 the whole plant was dry-shriveled. (310.) Citrullus vulgaris. The seventh day there was no wilt and it looked as if a cork-layer had formed around the pricks, at least there was a narrow yellow rim around each prick (estimated 0.1 mm. wide). The plant was examined June 25, 27, July 3 and 14. There was no result even in the pricked leaf. This leaf had received 67 pricks. (311.) Echinocystis lobata. Wilt appeared the sixth day in the pricked portion of the inoculated leaf. The seventh day nearly the whole of the pricked leaf had wilted and 2 days later the whole pricked blade including the tip of the petiole had shriveled. This leaf was separated from the one above by an internode of 4 inches and that leaf was still normal. The first leaf below was 4 inches down and that too was unaffected by the wilt. The eleventh day the plant had developed as fine a case of the bacterial wilt as could be desired. The pricked leaf had entirely shriveled, including Fig. 78. — Cross-section of petiole of Cucurbita foetidissima, showing a bundle occupied by Bacillus tracheiphilus as result of a pure culture inoculation on lamina of leaf. For orientation see fig. 77, which was made, however, Drawn from slide 254 B x, with the Abbe camera. Plant No. 273 from another section. 274 BACTERIA IN RELATION TO PLANT DISEASES. the petiole, and the wilt had extended to five leaves above (16 inches of stem) and to 3 leaves below. The stem was still green and looked normal; also the leaves farther up and lower down. On July 14 the whole plant, which was several feet long, had been shriveled for some days. (312.) Cucumis erinaceus. On the ninth day there was a slight yellowing of the pricked part and of the apex of the blade beyond the pricks, but no wilt. Two days later there was very little change. July 14 the pricked leaf had dry-shriveled to the stem but the rest of the plant was normal and was making a good growth. August 15 the plant was living and growing, having overcome the disease. (313.) Cucumis erinaceus. The ninth day this plant resembled the preceding. July 14 a few leaves had dry-shriveled but the bulk of the plant was uninjured and the wilt seemed to have stopped. August 15 the plant was still living and growing. It had overcome the disease. (314.) Cucumis erinaceus. The seventh day only about two-thirds of the small pricked leaf had wilted and 2 days later the signs were still slight. Part of the pricked leaf-blade had dried out and the lower edges were curved in and flabby. The leaf above resembled this one in being incurved and slightly wanting in turgor. The eleventh day the wilt was still confined to the pricked leaf and the first one above, the internode between the two being very short. On July 14 the pricked leaf was dead and the tip of the branch bearing it; also some leaves lower down. Midway, however, were three good leaves. The other branch (the plant was a small one) looked healthy and it seemed at this date as if the plant would outgrow the disease. On August 15 this plant, like the two preced- ing, was living and growing, having overcome the disease. Remarks. — The result with the watermelon confirmed the previous experiments. This plant is very resistant. Cucumis erinaceus, Apodanthera undulaia, Cucurbita palmata (?), Cucurbita digitata, and Trichosauthes cucumeroides were quite resistant. Some developed no signs; others only local ones; three developed constitutional signs, but recovered after a few weeks. Passi- flora incarnata is resistant. Cucurbita foetidissima. Cucumis melo var. dudaim, and Echinocystis lobata contracted the disease promptly and were destroyed by it. Inoculations of July 14, 1896. This set of experiments was made to determine whether the disease could be cut out. Twenty muskmelon plants (Cucumis melo) were inoculated in the hothouse from tube No. 12, July 8 (a potato culture made from a peptonized beef-bouillon culture which had been subjected to intense cold in a mixture of frozen carbon dioxide and ether) . On July 1 1 , the surface of the potato in this tube bore a good typical growth of Bacillus tracheiphilus. It was smooth, wet-shining, white, i. c., almost exactly the color of the steamed potato, and quite sticky. All inoculations were made on the leaf-blades and as far as possible from the stem. From 30 to 60 delicate pricks were made with a steel needle sterilized in a flame and cooled each time before using. The pricked area in each case was less than 1 sq. em. A loop of fluid from the bottom of the culture was taken out on a sterile platinum loop, placed on the clean surface of a leaf and the pricks were then made in and around this wetted surface, the drop being finally spread so as to cover all the pricks. A clean paper was then placed over the pricked leaf to screen off the bright sun. (315 to 334.) Twenty muskmelons. Next to the lowest leaf of each plant was inoculated on the blade 3 to 5 inches from the stem. July 31st. There has been no trace of wilt on any of these 20 plants. Remarks. — All the melons were of one variety, the New Early Hackensack. My plan was to cut away each pricked leaf at its junction with the stem as soon as a trace of wilt appeared, but the entire experiment miscarried. Up to July 31, none of the 20 plants had contracted the disease. Several hypotheses occurred to me as explanations of this failure: (1) The plants were of a resistant variety; (2) The wrong organism was used; (3) The cul- ture was dead when taken into the hothouse or, at least, that part of it in the fluid at the bottom of the tube; (4) The bacteria were destroyed by the bright sunlight in the short time which elapsed between spreading them on the leaf and pricking them in; (5) The intense heat of the hothouse detroyed them. The day was very hot and the sun bright. WILT OF CUCURBITS. 275 Possibly the plants may have been very resistant but my opinion at the time was that the bacteria were dead when inoculated. This is possible since the tube was inoculated very copiously on the start (by means of a pipette) and would consequently convert the sugar of the potato into a harmful acid sooner than under ordinary circumstances. Prob- ably the viscid bacteria on that part of the potato exposed to the air were still living, and very likely the experiment would have succeeded had slime been taken from the exposed surface, or had the latter been washed down into the fluid at the bottom of the tube by prolonged shaking, as in the experiment of July 16. The previous freezing had nothing to do with it, since freezing does not destroy the pathogenic properties of the organism (see experiment of July 16). Inoculations of July 15, 1896. A second set of inoculations was made in the hothouse to determine whether the disease could be cut out. The plants used were small muskmelons {Cucumis vnelo) and all the pricks were made on the apical part of the blade of the first or second leaf above the cotyledons. Many delicate pricks were made, covering an area not to exceed one square centimeter. My method of inoculation was to heat a platinum loop to redness, wait until cool, open the tube containing the culture and take out a loop of fluid from the bottom. I placed this loop of fluid on the leaf and pricked through it with a steel needle which was heated and cooled each time. The fluid was then spread so as to cover fully the pricked area in case any pricks extended outside of the liquid. The pricked portion was then covered from the direct rays of the sun for some hours. The infectious material used for these inoculations was taken from potato culture No. 12, July 8, i. e., the same culture which was used for the inoculations of the preceding day. (335 to 354.) Twenty muskmelons. No result. Remarks. — The melons were all of one variety — Princess. It was my intention to cut away the leaves close to the stem as soon as wilt appeared, but the experiment failed, the plants being inoculated from the same part of the same tube as the preceding. It is a good illustration of the danger of putting all one's eggs into a single basket. Merely as an ordinary precaution this set ought to have been inoculated from a different culture and transfers should have been made from each one into nutrient agar just prior to the inocula- tions so as to know whether the bacteria were really alive. Examination in a hanging drop just prior to inoculation would also have shown whether the fluid contained motile rods suitable for inoculation. As it was, two otherwise carefully planned experiments yielded only negative and disappointing results — results which have considerable interest, however, when compared with those of the next series. Inoculations of July 16, 1896. A third set of inoculations (ih to 5h 30™ p. m.) was made to see if the disease could be cut out. The plants were in a hothouse and the bacteria used were from tube 11, July 8 (a potato culture made from the bouillon culture which had been cooled to — 770 C). This culture was made at the same time and from the same tube as culture No. 12, July 8 (see inoculations of July 14, and 15 which failed). Loops of the liquid were also taken from the bottom of the tube but only after it had been shaken thoroughly in order to wash the sticky bacteria off the cylinder into the liquid. This was the only particular in which the material used for inoculation varied from that used for the preceding experiments which failed. Well-developed young, healthy, and rapidly growing cucumber plants {Cucumis sativus) were inoculated. The variety selected was White Wonder. Many delicate pricks (40 to 70) were made in the apical part of one leaf-blade of each plant, covering an area of not more than 1 sq. cm. The pricks themselves did the plant no injury. The platinum loop and the steel needle used in the operation were flamed and cooled each time before using. 276 BACTERIA IN RELATION TO PLANT DISEASES. A big loop of the fluid, containing many thousand bacteria (some of which were motile, as determined by examination under the microscope) was put on the clean surface of the leaf, spread a little and then rapidly pricked in, taking special care to make the needle- holes as small as possible. The afternoon was cloudy, rainy, and cooler than the 10 days preceding. On account of the cloudiness and moisture in the air the pricks were not covered over with papers. The plants were examined every day for the first 8 days and frequently after that. Twenty-four plants were inoculated. (355-) This plant was 18 inches high and very thrifty. The inoculation was made on the sixth leaf 9 inches away from the stem. The pricked leaf-blade was 5 inches broad. Up to the morning of July 21 there was no trace of the disease but at 3 p.m. of the same day about 0.5 sq. cm. on one side of the pricks was wilted. The following morning there was only a very slight change. By noon of the seventh day the wilt covered about 10 sq. cm., and reached half-way down the blade. The leaf was now cut off close to the stem with a hot knife. Four days later the vine was normal, apparently, except for a droop of the first two blades below and a fainter one of the first two above the node which had borne the pricked leaf. I filled the pot several times with water but an hour later the absorption of the water had not relieved the droop of the foliage. The next day in the afternoon the first two leaves below were cut away. They had not recovered their turgor. Three days later the first leaf up was gone (removed by someone), but the blades of the next four up showed the wilt. The eighteenth day the blades of the second and fourth leaves up were shriveled but the petioles were turgid. The fourth leaf was on the same side of the stem as the second. The blade of the third leaf which was on the opposite side was flabby but had not yet shriveled. The blades of the fifth, sixth, seventh, and eighth leaves up were drooping. The others were turgid The twenty-third day after inoculation all the leaves were shriveled and the stem itself was beginning to shrivel. The vine was about 5 feet long, i. e., it trebled in length after being inoculated. It was staked up. (356.) This was a thrifty plant about 28 inches high. The sixth leaf was pricked 10 inches from the stem. The pricked leaf-blade was 6 inches broad. The sixth day no signs had appeared but at noon of the following day there was a slight wilt in the pricked area and 2 days later this wilt covered about 5 sq. cm. The eleventh day (9 a.m., July 27) the pricked part had dried out and the wilt had increased only slightly (1 to 2 sq. cm.). At 3'1 30"' p.m., however, the wilt covered an area of about 10 sq. cm. The leaf was now cut off near the stem with a hot knife. Twelve days later no further signs had appeared, but the seventeenth day after the removal of the pricked leaf (28 days after inoculation), the blade of the first leaf below and of the first and second up were flabby. Examined microscopically the petioles of these three leaves were found to contain bacteria. (357-) This plant was 19 inches high. The inoculation was made 8.75 inches from the stem. The pricked leaf-blade was 5 inches broad. The fifth day, at 10 a.m., there had been no change in the appearance of the pricked leaf, but at 1 p.m., there was a slight wilt. At 3 p.m. I removed the leaf at the base with a hot knife, about 2 sq. cm. in and around the pricked area, having wilted distinctly. None of the other leaves ever showed any trace of the wilt. On September 23 (69 days) the plant was still living and free from the disease. (358.) This vine was 20 inches high and thrifty. The sixth leaf-blade was 5 inches broad. It was inoculated 7.5 inches from the stem. The sixth day the pricked leaf still presented a normal appearance, but the following day at noon there were about 3 sq.cm. of wilt in and around the pricked portion. Two days later (hot again) the wilted area had increased to 5 or 6 sq. cm. The tenth day there was a bad wilt of the pricked blade, mostly without change of color, but which I could not overcome by copious watering. The next morning the leaf-blade had changed color throughout (the characteristic dull green) and the outer three-fourths of the petiole was flabby. I did not remove this leaf. The twenty-third day the blade of the first leaf down and those of the first three leaves up had wilted. The petioles were turgid. (359-) This plant was 17 inches high and thrifty. The fifth leaf-blade which was 6.5 inches broad, was inoculated 8 inches from the stem. The fifth day, at 10 a.m. no signs had appeared but at 1 p.m. there was slight wilt in and around the pricked area. At 3 p. m. I removed the leaf with a hot knife, cutting the petiole close to the stem. About 3 sq. cm. of the leaf had wilted in the pricked area, and immediately around it. The eighteenth day (13 days after the removal of the pricked leaf), the foliage drooped a little, but it was doubtful whether this was due to the disease. The day was hot, still and cloudy. I watered the pot which had become rather dry, but this did not cause the leaves to recover their turgidity. Five days later (August S) the blades of three additional leaves were wilted. (360.) This plant was 19.5 inches high. The fifth leaf was inoculated 9.5 inches from the stem. The pricked leaf-blade was 5 inches broad. The eighth day there was no trace of the wilt but the WILT OF CUCURBITS. 277 next afternoon (2 p.m.) there was wilted tissue in and around the pricked portion, covering an area of about 5 sq. cm. Two days later the wilt involved about 10 sq. cm. and reached nearly half-way to the middle of the blade. The leaf was now cut away with a- hot knife close to the stem. The seventh day after the removal of the leaf (eighteenth day after inoculation), the first leaf up hung down flabby and the blades of the next three above drooped. Five days later more leaves were wilted. (361.) This plant was 13 inches high. The sixth leaf was inoculated 7.25 inches from the stem. Its blade was 4.5 inches broad. The sixth day there were no signs of the disease but the following day at noon there was a wilt of about 3 sq. cm. in and around the pricked part. Two days later (2 p.m.) the wilt included about 8 sq. cm. and reached half-way to the base of the blade. The leaf was now cut away at its junction with the stem, using a hot knife. Eleven days later there was a distinct wilt of the leaves to either side of the pricked one. Three days later (August 8) the next two leaves farther up were wilted. (362.) This plant was 22 inches high. The fifth leaf was inoculated 8.75 inches from the stem. Its blade was 6 inches broad. The sixth day ( 10 a.m.) there was no trace of the wilt, but the following noon there was wilt of about 10 sq. cm. of tissue in and around the pricked area. The wilt reached nearly half-way down the blade. The leaf was now removed close to the stem with a hot knife. Eight days later (15 days after inoculation), the blade of the first leaf up showed a decided droop. The leaves below had shriveled from other causes. The twentieth day after inoculation several more- leaves above the pricked one showed the bacterial wilt and three days later there were three addi- tional wilted leaves, farther up the stem. (363.) This plant was 26 inches high. The sixth leaf was inoculated 10 inches from the stem. The pricked leaf-blade was 6 inches broad. The first signs of wilt were visible the ninth day (2 p.m.) and were confined to the pricked area. Two days later the pricked area was dead and the tissue around it was yellow. There had been only a slight increase of wilt, but the whole leaf had a slightly yellow look. The thirteenth day (4. p.m.) the wilted area measured about 20 to 25 sq. cm. and extended along the midrib three-fourths of the way to the petiole. The leaf was now cut away close to the stem with a hot knife. Seven days later there was a distinct wilt of several leaves above the pricked one and 3 days after that additional leaves were wilted. (364.) This plant was 25 inches high. It was inoculated on the fifth leaf, 9.5 inches from the stem. The pricked leaf-blade was 5.5 inches broad. The plant was healthy on the afternoon of July 25. The eleventh day (10 a.m. July 27), the pricked area was dead and the surrounding tissue freshly wilted. In all there were about 3 sq. cm. of wilt. Two days later (4 p.m.) the wilted area around the pricks covered about 10 sq. cm. and was mostly dried out. The fifteenth day (2 p.m.) the wilt reached two-thirds of the distance to the petiole and covered 20 sq. cm. The leaf was now cut away with a hot knife close to the stem but 13 days later the blades of the first three leaves up were drooping. On examining each of them microscopically I found bacilli in the vessels in the base of the petioles. (365.) This plant was 22 inches high. It was inoculated on the fifth leaf 10.5 inches from the stem. The pricked leaf-blade was 6.5 inches broad. The eighth day no signs had appeared, but the following day (2 p.m.) there were about 2 sq. cm. of wilt in and around the pricks. During the next 3 days the wilted area increased to about 10 sq. cm. The leaf was now (July 28) cut away with a hot knife at its junction with the stem. No secondary signs appeared until after August 8. On August 13(16 days after removal of the leaf) the blades of several leaves up were drooping. I examined their petioles microscopically and found bacilli abundant in the vessels. (366.) This plant was 18.5 inches high. The fifth leaf was inoculated 9.5 inches from the stem. The pricked leaf-blade was 6 inches broad. The fifth day at 1 p.m. there were no traces of the dis- ease, but 2 hours later there was about 1 sq. cm. of wilt, mostly in the pricked area, but extending out a little on one side. By the next morning the wilt had increased about 1 sq. em. and 26 hours later (noon, July 23) measured about 14 sq. cm., reaching half-way to the base of the blade. With a hot knife I now removed the leaf at its junction with the stem. The plant showed no constitutional signs until after July 27. Eight days after the removal of the pricked leaf (July 31) the blades of the first four leaves above were badly wilted. The first two below had dry-shriveled from other causes. Five davs later five or six leaves above the pricked one had wilted and in 3 days more all the remain- ing leaves had succumbed to the wilt, and the vine, which was now about 5 feet long, had begun to shrivel. (367.) This plant was 18.5 inches high. The sixth leaf was inoculated 7.5 inches from the stem. The leaf-blade was 6 inches broad. By the thirteenth day (4 p.m.) there were about 2 sq. cm. of wilt extending along one side of the pricked area most of which was still sound. From this I inferred that the infection resulted from a few bacteria lodged on one side of the pricked area. Two days later (July 31, 2 p.m.) there were about 10 sq. cm. of wilt reaching more than half-way to the base 278 BACTERIA IN RELATION TO PLANT DISEASES. of the blade. The leaf was now cut away close to the stem with a hot knife. There were no signs until after August 5. On August 8 (eight days after the removal of the pricked leaf) the blade of the first leaf up hung flabby and the one below had wholly shriveled. The next three up showed a slight droop. Five days later several more leaves were drooping. (368) This plant was 20.5 inches high. The sixth leaf was inoculated, 6.5 inches from the stem. Its blade was 5.5 inches broad. At 9 a.m. of the fourth day there were no signs, but at 2 p.m. there was a wilted area extending from the pricked part toward the tip of the leaf, affecting about 2 sq.cm. At 5 p.m. the wilt was decided, involving all of the pricked area and a narrow strip extending nearly to the apex of the leaf (about 2 cm.). The pricked leaf was now cut away at thebase. This plant was examined July 27 and 31, August 8, 13, 17, 19, 28, and at later dates. None of the other leaves became affected. On September 23 the plant was still living and free from this disease. (369.) This plant was 16 inches high. The fifth leaf was pricked 9 inches from the stem. The pricked leaf-blade was 6 inches broad. The fifth day at 10 a.m. the leaf was still normal in appear- ance but 3 hours later there was a slight wilt in and around the pricked area. By 3 p.m. the wilt had spread rapidly. It then covered about 8 or 9 sq. cm. and reached nearly half-way down the blade. I now removed the leaf at the base using a hot knife. (This leaf was saved in alcohol for sections.) There were no signs until after July 27. Ten days after the removal of the pricked leaf (July 31) the blades of the first two leaves above were drooping badly. The first leaf below was normal. Five days later several more leaves above the pricked one were wilted. The eighteenth day after the removal of the pricked leaf additional leaves near the top of the vine were wilting. (370.) This plant was 19 inches high. The fifth leaf was inoculated 7.75 inches from the stem. The pricked leaf-blade was 6 inches broad. The fifth day, at 10 a.m. there were no signs but at 3 p.m. there was wilt of about 0.5 sq. cm., in the center of the pricked area. The wilt increased very little over night but the seventh day at noon it covered about 5 sq. cm. around the pricks. The eighth day was cool with heavy rains in the afternoon and the wilt was at a standstill. The following day, however, was sunny, transpiration was greater, and the wilt of the pricked leaf, at 2 p.m. covered about 12 sq. cm. and reached nearly half-way to the base of the blade. I now cut the leaf away at the stem with a hot knife. None of the other leaves contracted the disease. (371.) This plant was 20 inches high. The fifth leaf was inoculated 9.5 inches from the stem. The pricked leaf-blade was 5.5 inches broad. There were no signs up to noon of July 23. The ninth day (July 25, 2 p.m.) there was wilt of about 3 sq. cm. in and around the pricks. Two days later (10 a. m.) the pricked area was dead but there had been only a slight increase of wilt. The following afternoon (July 28, 5 p.m.) the leaf was cut away close to the stem with a hot knife. At that time there were about 8 sq. cm. of wilted tissue in the vicinity of the pricks. There were no constitutional signs until after July 31. On August 8(11 days after the removal of the pricked leaf) the first leaf below was shriveled and the blades of the second below and first above drooped a little. The sixteenth day after the removal of the pricked leaf there was bad wilt of several additional leaves above the inocu- lated one. I now cut off three leaves and examined them. On touching the cut ends with my finger the bacteria in the vessels strung out 1 to 2 cm. in numerous fine, sticky, cobwebby threads. (372.) This plant was 22 inches high. The fifth leaf was inoculated 9.5 inches from the stem. The pricked leaf-blade was 6 inches broad. Up to the fifth day at 3 p.m. no signs had appeared, but the morning of the sixth day about 1 sq. cm. of tissue in the pricked area was wilted. By noon of the following dav the wilted area was about fifteen times as large and reached half-way to the base of the blade. The leaf was now cut off at its junction with the stem, using a hot knife. None of the other leaves contracted the disease. On September 23 (69 days) the plant was still living and free from the disease. (373.) This plant was 23 inches high and very thrifty. The sixth leaf was selected for inoculation. Its blade was 5 inches broad, and the pricks were made 8 inches from the stem. On the seventh day at noon there was about 1 sq. cm. of wilt in the outer part of the pricked area. During the next 2 days the wilted area increased not more than 1 sq. cm. The eleventh day, at 10 a.m., the pricked area was dead and the surrounding tissue yellow, but there was only a slight increase of the wilt. Four days later (July 31, 2 p.m.) there were about 25 sq. cm. of wilt, reaching three-fourths of the way to the base of the blade. With a hot knife the leaf was now cut away at its junction with the stem. There were no constitutional signs until after August 5. Eight days after the removal of this leaf the blade of the first leaf up was drooping decidedly and the blades of the next two above showed a faint wilt (August 8, 10 a.m.). I now cut away the petiole of the first leaf up at its base, using a hot knife and examined it by touching the cut end with my finger. The surface ooze was sticky and strung a little. During the next 5 days several additional leaves wilted. (374.) This plant was 22 inches high. It was inoculated on the fifth leaf 10.5 inches from the stem. The pricked leaf-blade was 6 inches broad. Up to the fifth day at 10 a.m. there were no signs, but 3 hours later there was a slight wilt, and at 3 p.m. there was a decided wilt involving about WILT OF CUCURBITS. 279 3 sq. cm. This began in the pricked area and extended outward to the apex of the leaf. The leaf was now cut off at the base of the petiole with a hot knife. None of the other leaves wilted as a result of the inoculation. The basal leaves shriveled the fifteenth day, but from age, not from the wilt. September 23 the plant was dry but bore a healthy green fruit. The diseased plants had been bone-dry for weeks. (.375-) This plant was 16.5 inches high and thrifty. The inoculation was made on the fifth leaf 8 inches from the stem. The pricked leaf-blade was 6.5 inches broad. Up to the fifth day at 10 a. m. no signs had appeared. Three hours later there was a slight wilt, and at 3 p. m. this involved about 1 sq. cm. The wilt was on one side of the pricked area, and extended toward the margin of the leaf. The next morning the wilted area did not exceed 2 sq. cm. Twenty-six hours later (July 23) there were about 10 sq. cm. of wilt. I now cut off the leaf, with a hot knife, at its junction with the stem. There were no constitutional signs until after July 27, but 8 days after the removal of the leaf (July 31) the blade of the first leaf below and the first above drooped very decidedly. Between this date and August 5 several of the leaf-blades farther up had wilted. On August 8 (the sixteenth day after the removal of the pricked leaf) three more leaves above the inoculated one were wilted. (376.) This plant was 22 inches high. The fifth leaf was inoculated 10.25 inches from the stem. The pricked leaf-blade was 5.5 inches broad. The seventh day after the inoculation (July 23, noon) there were about 3 sq. cm. of wilt in and around the pricks. Two days later there was a wilted area of about 15 sq. cm. extending a little over half-way to the base of the blade. I now cut away the leaf close to the stem using a hot knife. There were no constitutional signs until after July 27. Six days after the removal of the pricked leaf (July 3 1 ) the blade of the first leaf below and of the first two leaves above drooped very decidedly. The eleventh day after the removal of the pricked leaf, I cut away five leaves with wilted blades, all above the inoculated one. Three days later (August 8) two more leaves above showed a decided droop of the blades. (377.) This plant was 14 inches high. The fifth leaf was pricked 7.75 inches from the stem. The pricked leaf-blade was 5.5 inches broad. The first wilt was noted the seventh day in and around the pricks. It then covered an area of about 2 sq. cm. Two days later (2 p. m.) about 7 sq. cm. of tissue had wilted. The eleventh day (July 27) about 12 sq. cm. of tissue extending about one-third of the distance to the base of the blade had wilted and changed color. The pricked area and the tissue immediately around it were now dead. The leaf was cut off close to the stem with a hot knife. The plant showed no constitutional signs until after August 3. Twelve days after the removal of the pricked leaf the first leaf below was shriveled, probably from age, and the first leaf up showed a decided droop of the blade due to the disease, I cut the petiole and saw the bacterial slime string out. The next two leaves up showed a slight droop of the blades. Five days later (August 13) two more leaves above the pricked one were wilting. (378.) This plant was 22 inches high. The fifth leaf was pricked 10 inches from the stem. The pricked leaf-blade was 5.5 inches broad. The first wilt was noted at noon of the seventh day. It then covered about 1 sq. cm. of the pricked portion. Two days later (July 25, 2 p. m.) the wilt involved about 10 sq. cm. of leaf surface and reached nearly half-way to the base of the blade. I now cut away the leaf at its junction with the stem, using a hot knife. There were no constitutional signs until after July 27. Six days after the removal of this leaf (July 31) the blades of the first five leaves up were drooping very decidedly. The leaves below had shriveled from age. Twenty days after inoculation I cut away six petioles of wilted leaves to put into alcohol. The twenty-third day August 8 one more leaf up showed wilt of the blade. The stem was still green and turgid. Remarks. — This experiment is in striking contrast to those of July 14 and 15. Everyone of the twenty-four plants contracted the disease, and in each ease it first appeared in the pricked area. Nineteen of the plants subsequently developed constitutional signs and died of the disease. No general signs appeared in the other five plants (Nos. 357, 368, 370, 372 and 374), i. e., the disease was stopped by the removal of the affected leaf. In eighteen eases the amputation of the affected leaf did not check the spread of the disease, but it is apparent that a prompter removal of the pricked leaves, to wit, on the day the signs first appeared, would have considerably increased the number of recoveries. This is deducible from the fact that in those which did escape, the amputations were performed very promptly. It is not likely, however, that this method of treatment will ever be recommended for general use, since, in most cases, the wilt of the leaves would not be detected in time. The experiment was practically closed the forty-third day after inoculation (August 28) but the plants stood on the bench in the hot-house until September 23. On that date all of the diseased plants were bone-dry and had been so for several weeks. Three of the plants 280 BACTERIA IN RELATION TO PLANT DISEASES. which did not develop secondary signs were still living (Nos. 357, 368 and 372) ; one was dry (374) but bore a green healthy fruit, and of the fifth (370) no record was made later than August 28. The first case (No. 368) appeared on July 20, at 2 p.m., i. e., at the end of the fourth day. Eight cases developed at the end of the fifth day (Nos. 355, 357, 359, 366, 369, 370, 374 and 375). One case (372) appeared after about 5! days. Eight cases developed at the end of the seventh day (Nos. 356, 358, 361, 362, 373, 376, 377 and 378). Four cases appeared at the end of the ninth day (Nos. 360, 363, 365 and 371), and one plant (364) came down the tenth or eleventh day. The last plant to become diseased was No. 367. This did not show any signs until the thirteenth day. With the exception of the stems of 355 and 366, which were beginning to shrivel, all of the stems were turgid and sound externally until after August 8. Throughout, the plants were free from insects. The plants were otherwise very healthy, were watered regularly and properly cared for, and the signs obtained were characteristic of the disease. They could be ascribed only to the initial bacterial infection of July 16. In many cases the removal of the inoculated leaf was delayed so long after the first signs appeared (several days) that we can not tell therefrom how short a number of days is requisite for the general infection of the plant when the bacteria are introduced into the blades of the leaves. A few, however, give us some basis for judgment. In vine 355, at the end of 7 days, the bacteria were 7 inches in advance of the signs of wilt, i. e., they had already traversed the vessels a distance of 9 inches from the point of inoculation (how much farther we can not tell) and had entered the stem, as shown by the subsequent behavior of the plant. In 356, at the end of 1 1 days, the bacteria were at least 9 inches in advance of the signs of wilt and had passed through the spiral vessels of the leaf a distance of 10 inches (or more) into the stem, as shown by the subsequent behavior of the plant. In 359, at the end of 5 days, as shown by subsequent signs, the bacteria had entered the stem, having passed through 8 inches of vascular system and being that much at least in advance of the signs of wilt. In 369, in 5 days, as shown by subsequent signs, the bacteria passed through 9 inches of tissue (vascular system) and entered the stem. In 372, on the contrary, at the end of 7 days, the bacteria had not yet entered the stem (9.5 inches distant), although at the time the leaf was cut away there were 15 sq. cm. of wilted tissue, reaching nearly to the middle of the blade, and signs had been present for at least 26 hours. From these results we may conclude that under favorable circum- stances infection of the main axis in the cucumber is comparatively prompt, the bacteria being able to pass down through the vessels of the leaf at the rate of about 0.75-inch to 2 inches (2 to 5 cm.) a day. Weather conditions have much to do with the rate of progress of this disease. Cool weather retards it, warm weather hastens it, extremely hot weather if long continued cheeks it altogether. Following these inoculations, daily weather records do not appear to have been kept, at least on the sheets bearing my pathological memoranda, but there are occasional refer- ences to the weather. It was cool on July 23, and cool with heavy rains on July 24. This weather temporarily checked the progress of the disease. It was hot on July 25, and on July 29 and 30. It was cooler on July 31, but windy so that transpiration would be rapid. It was very hot on August 5 to 10. On the unrecorded days there was probably the ordinary summer weather. Between August 5 and 8 numerous freshly wilting leaves (secondary wilt) were cut from these plants and fixed in strong alcohol to determine whether the bacteria are actually in the vessels of the leaf at the time the secondary wilt appears or whether this wilt is due simply to the plugging of the vessels of the stem. These leaves were well grown and sound externally. Thin microtome sections were made from the basal part of the petiole of 66 of these leaves after infiltration with paraffin. These sections were fixed on clean slides and carefully examined after removal of the paraffin and staining in carbol fuchsin. Bacteria WILT OP CUCURBITS. 28 1 can not be demonstrated in every one but they occur in 61 of them; no fungi are present, neither are there any insect-injuries. In most cases the bacteria are confined strictly to the spiral vessels of these petioles and they do not occur in all of these, nor in all of the bundles. They are not present in the phloem, the cortical parenchyma or the tissues between the bundles. Summarized, the amount of bacterial infection in the basal part of these petioles is as follows: (1) In a few petioles nearly every bundle is occupied and bacteria occur in many vessels, with cavities in two or three cases; (2) in 5 no bacteria detected; (3) in by far the greater number the bacteria are confined to a few vessels of a few bundles. In groups 2 and 3 the wilt can be accounted for only by bacterial occlusions lower down in the stem itself. From this experiment we may also conclude that White Wonder is a very susceptible variety and one to be rejected in regions much subject to this disease. This single experiment, purposely given in much detail, is sufficient, in my judgment to establish not only the bacterial nature of this disease but also the general movement of the bacteria through the spiral vessels of the plant. Inoculations of July 23, 1896. Another attempt was made in the hothouse to transfer the disease to plants by means of the beetle Diabrotica vittata. The vines used were cucumbers and muskmelons (Nos. 379-429). I had grown them from the seed and transplanted them some weeks before into benches filled with good earth. They occupied the whole of a small greenhouse. They had been well watered and the temperature had been high (hot July weather). As a result the growth had been rapid and at the time of inoculation the vines presented a very thrifty appearance. There were about an equal number of each. The cucumbers were from 4 to 6 feet high with hundreds of leaves, some of which were 9 inches broad. There was not a yellow, dwarfed, or fungus-spotted leaf in the whole house and the plants had been remark- ably free from aphides. No water had been put on the foliage. The muskmelons were equally healthy but were younger plants and had not made as much growth. When ready to make the inoculations I collected fifty to a hundred specimens of the striped cucumber- beetle (Diabroctica vittata) from squash-vines in Mr. Curtis's field south of Anacostia, where no wilt had yet appeared. Most of them were taken from the interior of squash flowers where they were in hiding through the day. These were colonized (July 23, p.m.) on five cucumber-leaves cut from plants infected July 16 (third set of inoculations to see if the disease could be cut out). Each of these leaves contained from 10 to 15 sq. cm. of freshly wilted leaf surface. Only the petioles, the wilted part of the blades and a narrow border of the blade surrounding the wilt was put in, so that the beetles would be compelled to feed on diseased tissues or else attack the hard petioles. These beetles feed mostly by night or in the early morning. At 5h 30™ the next morning the beetles had riddled the wilted parts with holes. They seemed to have fed exclusively on the parenchyma of the blade and must have consumed enormous quantities of living bacteria. The mouth parts of every one must have been infected. They were now turned loose on the cucumber and melon vines and began to feed at once. The day was cool and rainy. Several additional coloni- zations were made since it was believed that there were many chances for failure. Such additional colonizations were made on July 24, 3 p.m., July 26 (12 beetles), July 28 (7 beetles), July 29 (150 beetles fed 12 hours on freshly wilted leaves and turned loose at 5 a.m., after the leaves were riddled with holes), August 1 (20 beetles), August 4 (200 beetles fed over night on freshly wilted cucumber leaves and set free at 511 30™ a.m., when the leaves were riddled with gnawings). The beetles were allowed to feed on diseased material from 10 to 19 hours and were then turned loose in the greenhouse. On August 4 (511 30™ p.m.) there had been signs of the wilt for some days but inas- much as the plants had been sprinkled with fine tobacco dust for aphides I thought possibly the injury was due to that and I neglected to examine any of the wilting leaves for bacteria (a serious omission). 282 BACTERIA IN RELATION TO PLANT DISEASES. On September 22 the experiment was closed. It failed, apparently because of the extreme heat. The plants were examined frequently but I could find no secondary wilt. August 10 was the hottest of 4 very hot days. On that day it was 370 C. for some hours and might have been considerably hotter in this small house for a short time, possibly as hot as 43° C. (the thermal death point of the organism). Thermometers about town regis- tered 980 to 1030 F. and it would of course be warmer in the glass house exposed to the sun. In this connection see thermal tests on p. 293. If I were to repeat the experiment I would liberate the beetles at midnight and divide the house, keeping one part cool. Field Observations on August 3, 1896. On July 22, 1896, at Mr. Curtis's place 7 miles southeast of Washington, I tried to show Mr. Henry G. Hubbard, the entomologist, my wilt disease and failed. Mr. Curtis had about an acre of pumpkins and squashes of various ages, some covering the ground and in full bloom, others not yet in bloom and only just commencing to "run." There were at that time a very few wilted plants which Mr. Curtis attributed to borers and I to the bacterial wilt but as I could not find sticky slime inside the stems and had no microscope with me my diagnosis was unsatisfactory. The field was revisited 12 days later (August 3). There were then numerous well-developed cases of the bacterial wilt — long shoots with all the leaves drooping, and more interesting still, there were dozens of big vines which showed no general wilt, but had single leaves, usually toward the base of a stem, which bore characteristic wilt patches, i. c, pale-green flabby spots varying from a few square inches to areas as big as the palm of my hand. In all such cases this wilt appeared to have spread from gnawed places.* Moreover I saw the striped cucumber-beetle {Diabrotica vittata) eating holes in such wilted spots exactly as if this part of the leaf were a little tenderer or otherwise more desirable food (see page 215). There were on the vines great numbers of this beetle. A few specimens of Diabrotica duodecem punctata , and a very few of Coreus tristis were observed. My prediction that 90 per cent of Mr. Curtis's squash-vines, especially the older ones, would have the wilt by September came true. They were planted on the ground used for squashes 2 years before. Then nearly the whole field contracted the disease although the vines looked well at a date about corresponding to this time. The chief mischief was being done by Diabrotica vittata. This insect feeds mostly in the evening, night, and early morning. In the middle of the day it hides away from the hot sun. It specially likes to take shelter inside squash and pumpkin flowers which have recently opened. The older flowers are not to its taste. Query: Why can not squash flowers be used as traps for this insect? Hand picking of the staminate flowers in the middle of the day when they inclose from one to a dozen of the beetles would greatly reduce the numbers of this pest and, in conjunction with removal of wilted vines as fast as they appear, would go far toward checking the spread of the wilt disease. This disease may certainly be controlled by destroying the insects which distribute it and to that end their habits should be studied more carefully. Inoculations of July 26, 1S97. Four cucumber-vines (Cucumis sativus) and two vines of a cucurbitaceous plant from Mexico were inoculated in the hothouse with bacteria from a well-clouded tube of beef- broth (tube 2, July 19) made directly from the sticky interior of a diseased cucumber-stem from Virginia (near Norfolk). No statement as to the size or age of the plants. The inocu- lations were made by means of numerous needle-pricks on two leaves of each vine. There *In nearly a thousand cucumber leaves examined in July, 1893, on this same farm, for very early stages of the wilt, the gnawings of Diabroticas occurred in the diseased areas and seemed to have preceded the appearance of the wilt. i. e., the gnawed part was dried out as if older than the rest of the wilt. (See plate 1, fig. 1.) WILT OF CUCURBITS. 283 were no signs on any of the plants until after July 30. The Mexican plant is Dr. Edward Palmer's No. 1801a, and was grown from seeds of his collecting. (430.) Cucumber. Up to August 2 (seventh day) there was no result from the inoculation. (431.) Cucumber. The fifth day there was a distinct wilt of several square centimeters around the pricks on each leaf. (432.) Cucumber. The fifth day there was a distinct wilt of one square centimeter in the pricked area on one leaf while the whole blade of the other leaf was wilted and collapsing. (433.) Cucumber. The fifth day there were several square centimeters of wilt in the pricked area on both leaves. (434.) Cucurbitaceous plant (collected in Mexico). No result by the sixth day. No further record. (435.) Duplicate of 434. The sixth day both leaves were normal. No further record. Remarks. — Three out of the four cucumbers contracted the disease promptly. The ill-scented Mexican plant bore yellow flowers; long, warty fruits, and leaves suggestive of Momordica. Inoculations of August 21, 1897. A series of inoculations was made on watermelon vines (Citrullits vulgaris) in a garden atHubbardston, Michigan. The bacteria were taken from a potato-culture of B.tracheiphilus (tube 2, July 23) which had been re-inoculated July 26 with 0.2 cc. out of tube 1, July 19. This potato-culture was very sticky and in fine condition on August 12 when it was exhibited at the Detroit meeting of the American Association for the Advancement of Science, but at the time of the inoculations it had been considerably exposed to the light and was past its prime, although I believed it to be alive. It was not, however, tested under the microscope or by transfer to other media as it would have been had I had laboratory facilities. The inoculations were made in the ordinary way, i. c, by means of a dozen or two needle-punc- tures on the leaf-blade and within an area of 2 or 3 sq. cm. The extreme upper part of the potato bore the stickiest slime and this was used for most of the inoculations, but some were made in the following way: Numerous punctures were made and then the tube was tilted until the cotton plug was wetted. This plug was then taken and mopped over the pricked area. The fluid in the bottom of this tube was very milky. The inoculations were made at sunset so as to avoid the immediate evil effect of light. As a check on the virulence of the culture eight muskmelon leaves belonging to five plants were inoculated in the same way. On four leaves the bacteria were pricked in; on the others they were mopped in after the pricks were made. (436-439.) Watermelons. Sixteen healthy leaves belonging to four vines were inoculated, eight in the ordinary way, four by the second method described. There was no result. (440-444.) Muskmelons. Eight leaves belonging to five plants were inoculated, four by the second method. The inoculations failed. Remarks. — The observations were continued until September 12. The weather was very dry. Probably the potato culture was dead. It will be remembered that it was 26 days old and that it had been very copiously inoculated on the start. Inoculations of July ii, 1898. A series of inoculations was made on the wild bur cucumber (Sicyos angulatus) growing in a garden at Anacostia, D. C. The plants were large and covered the ground, the leaves being 4 to 8 inches broad, and all were perfectly healthy. All of the inoculations were made by pricking in the bacteria (cucumber-strain) with a sharp steel needle. (445-456.) About a dozen leaves belonging to several different plants were inoculated in the blades from a pure agar-culture (tube 3, June 30, re-inoculated July 7.) The seventh day there were seven beautifully typical cases in as many of the pricked leaves. These were in all stages from one just beginning to change color and wilt around the punctures to one wholly collapsed. There was 284 BACTERIA IN RELATION TO PLANT DISEASES. now an enormous tangle of foliage and all the inoculated leaves which I could find were the seven diseased ones. Signs appeared only in the pricked leaves. Several of these leaves were being gnawed by the striped cucumber-beetle (Diabrotica vittata) but almost exclusively in the wilted, softened parts. A week later the disease was progressing typically. The blade of the leaf which showed only a tiny wilted area around the pricks on July 18 was now two-thirds wilted. Several of the inoculated plants now showed constitutional signs, i. e., other leaves up and down the stem were flabby or wholly collapsed. One of these plants was brought in and examined under the microscope, bacteria being found in the vessels. The bacteria also strung out one centimeter when the sticky cut surface was touched with the finger. The plant was saved in alcohol. The disease had also been transmitted to at least one healthy vine (one leaf) by the bites of the Diabrotica. This beetle was observed feeding on the wilting leaves the week before and also to a slight extent, on the sound ones and one of the latter now showed several square centimeters of wilt around the bitten part. (457-477.) About twenty leaf-blades belonging to several different plants were inoculated directly from muskmelon petioles showing the sticky bacterial exudate. There was no definite result until after the seventh day. On July 25 several plants showed typical signs in the pricked leaves. Remarks. — This experiment adds another plant to the list of possible hosts. These observations also tend to confirm the belief already expressed that the striped cucumber- beetle prefers the wilted leaves and is consequently admirably adapted to spread this disease. The foregoing experiments were all made by the writer. Numerous additional ones, partly by the writer and partly by his assistants (plate 14) have been made since this date, but need not be mentioned here since they are not contradictory. Some of them have already been referred to in the first part of the discussion of Etiology, where also are some important observations on the distribution of the disease by Diabrotica vittata. Mention should be made, however, of some watermelon plants inoculated November 17, 1903. Of these, two plants (No. 530, variety Phinney's Early, and No. 534, variety, Mountain Sweet) showed wilt of the inoculated leaf and were brought in and put into alcohol on December 9 and 10. Nos. 526 and 528 (variety, Triumph), inoculated at the same time, showed secondary wilt on December 5 and December 7, respectively. These also were preserved in alcohol. Sections from the stems of the ones last mentioned show the presence of bacteria in all of the vascular bundles. An attempt to plate out the organism miscarried. From the foregoing inoculations we may conclude that in wilting cucumbers the organ- ism present in the tissues is sometimes B. tracheiphilus and sometimes B. trachciphihts f. cucumis, whereas in squashes it is always or nearly always the first and more virulent strain. This being true we might then expect some isolations from cucumbers to be infectious to squash and others not, whereas all isolations from squash should infect both cucumbers and muskmelons. I can not say whether there are any tangible cultural differences. In one test in litmus milk at the end of two weeks the squash strain looked exactly like the check tubes, while the cucumber strain was slightly darker. I cultivated out both on potato in typical form. EFFECT OF WATER ON CUCUMBER-WILT. Experiments in the greenhouse in May, 1895, showed that when water is withheld or given sparingly the bacterial wilt becomes visible sooner than when the plants have an abundance. If water be given to such diseased plants in abundance the leaves least wilted will frequently become turgid again for a few hours. The same thing occurs in the field, especially with squashes. VARIETIES ATTACKED. The writer has observed this disease or received reports of its occurrence in the follow- ing varieties of cucumbers: White Wonder, White Spine, Long Green Japanese, Long Green, Fordhook pickling, Telegraph, Large English (few seeded sort). WILT OF CUCURBITS. ?85 It has been observed or reported in the following varieties of muskmelons: Early Haekensaek, Shumway's Giant, Dudaim, Rocky Ford. It has been observed in the following varieties of squash: Hubbard, yellow Crookneek, Long Island White Bush, Early Yellow Bush Scallop, White Summer Crookneek, Boston Marrow. July 14, 1909, the disease was observed near Washington on Venetian squashes, grown from imported seed. MORBID ANATOMY. There are no hyperplasias in connection with this disease. It is principally a disease of the spiral and ring-vessels, and their entourage in the stem of the host. These vessels are arranged in a group toward the inner part of each bundle. They are embedded in a mass of thin- walled living parenchyma, which is separated from the inner phloem by a thin band of tissue somewhat resembling cambium in structure, and sometimes called pseudocam- bium. By its outer face, the tissue containing the spirals and ring-ves- sels joins on to the lignified tissue or xylem proper which contains the large pitted vessels embedded in pitted, lignified connective tissue. The spiral-vessels and ring-vessels are always the first part of the stem to be occupied by this bacillus. The reason for this is not far to seek. It lies in the fact that they are the only part of the xylem- portion of the bundle which passes out from the stem into the leaves to form the xylem-part of the veins of the leaf. Since the infections are entirely through the leaf-sur- face (so far as we yet know*) and since the organism passes downward into the stem exclusively by way of the spiral-vessels of the petiole, it is at once apparent why the spiral-vessels of the stem are the first part of that organ to be invaded. In the stems of the cucurbits subject to this disease there are nine or ten separate vascular bundles and, consequently, on cross-section there are, or may be, nine or ten distinct bacterial foci corresponding to as many groups of spiral-vessels and ring-vessels (figs. 67, 77, 78, 80). The organism always appears in these spiral-vessels in enormous numbers, soon filling them completely (fig. 79). The next stage in the progress of the disease is the destruction of the walls of these vessels. This appears to take place by solution, but perhaps also by rupture, these walls consisting of a very thin non-lignified membrane the only lignified parts being the spiral-thickenings or ring-thickenings. The bacteria now invade in great numbers the thin-walled living parenchyma surrounding these *Further studies should be made to determine whether infection may not also take place through the root- system. tFic. 79. — Cross-section of a cucumber stem from field, showing a small spiral vessel filled with Bacillus trachei- philus, contents of non-parasitized vessel-parenchyma-cells omitted. Drawn from a photomicrograph of a thick, unstained glycerin mount made in 1893 (the year I discovered the disease), x 1000. Fig. 79. t 286 BACTERIA IN RELATION TO PLANT DISEASES. vessels. The cells are separated from each other, crushed and dissolved (?) their place being taken by the rapidly multiplying bacillus (fig. 62). For an earlier stage of bundle infection see vol. 1, fig. 9. In this way cavities arise which by fusion with other cavities lead to the honey-combing and more or less complete destruction of this part of the bundle and consequently to the interruption of its function, viz., the movement of water. The bacteria also pass outward through the lignified tissues (by way apparently of the pits) into the large pitted vessels, several to many of which are often filled partly or completely before there is any destruction of the phloem or of the general connective tissue of the stem. In the end, the bacteria may be found also in the phloem and outside of the bundles in the surrounding tissues. For an especially good example of a late stage in which the bacteria have passed beyond the limits of the bundle and may be seen occupying the intercellular spaces and the interior of parenchyma cells see fig. 81. By this time, however, the stem begins to shrivel from loss of water, and the activities of the organism cease, so that the phloem and the tissues lying between the bundles, or beyond them toward the periphery of the stem, are seldom occupied to any great extent. Frequently pitted vessels at the outer angles of the xylem become filled in advance of those in the middle. The lignified tissues are not dissolved, but the thin non-lignified membrane separating the pits on contiguous vessels must be ruptured or dissolved by the bacteria — otherwise it is impossible to account for their diffusion into the connective tissue of the xylem and from one pitted vessel to another. It is very easy to demonstrate microscopically the presence of the bacteria in the vessels, to cultivate them therefrom (when the right methods are used) and by means of sections made from pieces embedded in paraffin to show all stages in the destruction of cells and in the formation of cavities in the bundles. The organism occurs also in the green fruits of cucumbers and produces therein the same occlusion of vessels and breaking down of neighboring cells, with the formation of small bacterial cavities, as in the stem. The fruit finally shrivels and the flesh sometimes has a water- soaked look about the bundles, but there is no general disintegration of the parenchymatic tissues, i.e., no soft rot. Numerous examinations under the microscope have disclosed no tendency of the cells of the host-plant to enlarge or divide in the presence of the organism, nor have I detected any distortions or suppressions of particular systems of tissues such as we commonly find in certain other bacterial diseases. The tissues of the attacked plant seem unable to react, except that, as already mentioned, I have observed in the field, in certain squashes attacked by this organism, certain proliferations which, rightly or wrongly, I have attributed to its presence in the tissues; and also in certain inoculated squash-cotyledons a suggestion of cork-formation in the pricked area, and a very slow multiplication of the bacteria in the bundles. THE PARASITE. Bacillus tracheiphilus EFS. — The cause of this disease is a short, straight rod with rounded ends (figs. 57 and 82). When growing rapidly in the plant or on culture-media it commonly measures 1.2 to 2.5/z by 0.5 to 0.7,11, but it may be longer or shorter or thicker or *Fig. 80. — Cross-section of a squash petiole, showing 12 vascular bundles occupied and destroyed by Bacillus tracheiphilus. Tissue between bundles and toward surface is free from bacteria. Inoculation was made Aug. 10, 1905 (Colony E, House 4), on blade of leaf by needle-pricks Prom a photomicrograph. Slide 354-3. For detail see fig. 81. Fig. 80.' WILT OF CUCURBITS. 287 thinner, according to age, culture-medium, and kind of stain used. Flagella-stains in par- ticular, affect an outer part not stained by ordinary methods (figs. 83 and 84). The organism has a distinct capsular portion, the solution of which appears to give rise to the viscidity. This viscidity occurs during active growth, and may continue for some time. When taken directly from the plant (fig. 52) this bacillus is usually viscid but not always. Often with care the slime may be stretched out to the distance of 20 to 40 cm. (once 76 cm.). The resulting cobwebby threads generally yield pure cultures of this organism and when stained on a slide and examined under a microscope are seen to be made up of bacilli embedded in a tenuous slime which separates them by considerable intervals (Vol. I, figs, 13 and 14). The Fig. 81.* organism is also sticky, at least in some of the stages of its growth, on agar, gelatin, potato, carrot, sweet potato, coconut-flesh and various other solid culture-media. In one instance the slime from a potato-culture was drawn out 53 cm. before it gave way. On potato, up to the sixth day and sometimes longer, the organism is actively motile, even when examined from very viscid cultures. Potato cultures 10 to 18 days old are usually as sticky as younger ones. This slime does not dissolve readily in water and hence failure may occur in making poured plates. Cultures in potato-broth and in sugared fluids finally become ropy, and then most of the individuals or all of them are dead. *Fig. 81. — Cross-section of one bundle of fig. 80. At bottom and left-hand side are numerous intercellular spaces occupied by bacteria. The bundle has been hollowed into a cavity, and at .v, and y parenchyma cells are also occupied by the bacteria. These can be followed through a whole series of sections, but the method of entrance into these cells has not been made out clearly. Fixed in Carnoy 48 hours. Drawn with a Zeiss 16 mm. apochromatic and No. 12 compensating ocular. Slide 354-2, middle section, middle row. 288 BACTERIA IN RELATION TO PLANT DISEASES. a- \ \ i\ Fig. 82* The organism occurs singly, in pairs, and more rarely in fours joined end to end. Long chains and filaments have not been observed. Pseudozoogloeae or flocculent particles do not occur quickly in bouillon, but compacted masses have been observed on plates, etc., and they occur in the plant. Usually in fluid cultures no rim or pellicle is formed, but there may be flocculence in some liquids, e. g., Uschinsky's solution, though this is often wanting. Spores are not known to occur. This organ- ism is a white, wet-shining, schizomycete, motile by means of 4 to 8 peritrichiate flagella (fig. 84). In young cultures motility is easy to observe. It has also been observed in slime taken directly from the vessels of the cucumber, melon, etc., and diluted with sterile water. In the plant motility is easier to observe in rods taken from tissues recently invaded than in those taken from crowded vessels. Involution forms occur in the plant and in various media — beet-juice, cucumber-juice, peptone-water, potato broth (vol. I, fig. 21), etc. On steamed potato the growth is white and so closely like the substratum in color that it is scarcely to be distinguished therefrom except by its smooth, moist, glistening appearance. It has very little action on potato- starch, even very old potato-cultures reacting strongly with iodine. It does not soften the middle lamella of potato-cells. Potato-cylinders may or may not be grayed. In one experi- ment the organism remained alive on potato over seven months (160 to 180 C), but usuallyit is dead muchsooner. In one set of experiments no growth was obtained on red turnip-rooted table-beets, nor on turnip-roots, radish-roots, or cauliflower-petioles. When these experiments were repeated some years later the fol- lowing results were obtained: On the beets growth was delayed but finally appeared. It was visible in one tube on the sixth day, in another on the fifteenth day, and in a third on the twenty- second day; the fourth tube remained sterile. Similar results were obtained by a repetition of the experi- ment: There was marked retardation of growth as compared with that in the agar-stab, none being visible on the beets the fourth day; on the thirteenth day there was a visible growth in one of the tubes but not in the other, on the twentieth day a typical growth appeared in the second tube. Upon radishes similar results were obtained, viz., marked retarda- tion but final growth in most of the tubes. No growth was obtained on turnips (two sets) and only a doubtful growth on cauliflowers. In 1896, inoculations were made into the juice of red table-beets filtered sterile and used both with and without the addition of calcium Fig. 83.t fj* Fig. 84.: *Fig. 82. — Bacteria from interior of cucumber-stem at time of general wilt of foliage, but while main axis was still green and normal in appearance (plate i, tig. 2); bacteria were present in vessels in enormous quantities. The great mass of them were of size and shape of 2, those of size 1 being seen only occasionally. Drawn unstained with Abbe camera, Zeiss 3 mm. 1.40 n. a. apochromatic objective, and No. 18 compensating ocular. Anacostia, D.C., July 15,1893. tFio. 83. — Free-hand drawings of paired rods of li. tracheiphilus which have lost their flagella. Stained by van Hrmcngem's nitrate of silver method in April, 1895, and measured after lying in balsam until Oct., 1895. Size of organ- ism appears greater when stained in this manner than when stained without flagella-mordants. Very careful measure- imnts of one member of each pair gave following result in microns: 2.03X1.05; 2.10X0.98; 2. 55X1. 10; 2.18X1.05; 2.75X1.20; average 2.32 X 1.08. Circa X1500. G. 84. — Camera drawing from cover-glass preparation of young culture of B. tracheiphilus stained by van Ermengem's nitrate of silver method. Flagella distinct; hundreds on cover-glass. Some rods with only one or two left, others with as many as 8; most bear about 6. An occasional flagellum is 10.5 // long; most are shorter. Oct. 4, 1895. Circa X 1500. WILT OF CUCURBITS. 289 carbonate ; also into boiled sterile beet-juice and the same rendered moderately alkaline by various small amounts of sodium carbonate, but in all cases growth was absent, feeble or long delayed. Involution forms were present. The result of this series of experiments and of those which preceded it goes to show that the red beet either lacks some nutrient element or contains some substance which, while not destroying the germ, almost completely inhibits growth, and this whether the juice is acid or alkaline or whether sterilized by steam heat or by filtration. The result in tube 5, where after 25 days there was a considerable multiplication, seems to show that this inhibition is due to the presence rather than the absence of some substance. The organism is white in the plant and also on a variety of culture-media. It produces no pigment other than the occasional gray stain on potato common to many bacteria. In agar and gelatin the best growth along the line of the stab is at the top ; the surface growth above the stab is thin, gray-white, and may eventually cover two-thirds of the surface (fig. 856) or even the whole surface. Fig. 86 shows the appearance of a streak-culture on gelatin. The surface colonies on agar (figs. 87 and 88) and on gelatin are small, circular, slow of growth, gray-white, smooth and usually wet-shining. Internal striae may often be seen by careful manipulation of reflected light (fig. 89). They are not on the surface and can not be seen by transmitted light (fig. 90). Plate-cultures incubated at 250 C. are seldom in good condition for study before the sixth day. Often streak- cultures show a discrete growth either throughout or on the margins (fig. 91). It does not grow on strongly acid gelatin, or on the same after it has been made neutral or feebly alkaline to litmus but is still decidedly acid to phenolphthalein. This organism is facultative anaerobic. It will not grow in the closed end of fermentation- tubes in sugar-free, petonized beef-broth, nor in the same fluid with addition of milk-sugar, maltose dextrine (fig. 926), glycerin (Vol. I, fig. 48), etc. When, however, grape-sugar, cane-sugar, fruit- sugar or mannit (?) are added to the beef-broth, the inoculated fluid becomes clouded in the closed end of the tube in the absence of air (see fig. 92a and Vol. I, fig. 47) : No gas is formed in fermentation-tubes or in any of the common media, but inoculated culture-media containing grape-sugar, fruit-sugar, or cane-sugar become acid. This acid is not volatile, but rather is concentrated by boiling, and the writer found that it could be extracted from cultures in bouillon by prolonged shaking with ether. In 1909, flask cultures several months old,f set in 700 cc. filtered river water containing 35 grams grape-sugar, 14 grams Witte's peptone, and 35 grams calcium carbonate, were submitted to Dr. Carl L. Alsberg for analysis. He did not succeed in identifying the acid, owing to insufficient amount of material, but made the following interesting negative determinations: Fig. 85.1 *Fig. 85. — Gelatin stab cultures of Bacillus tracheiphilus after 10 days at 25°C: a, inoculated June 14:6, June 18, 1904. No liquefaction. Figure a photographed under water, b photographed in air. fThe fluid was well clouded after some days and remained so for several months (more than 4). No pellicle formed, only small floating islands which were most abundant near the walls of the flask. There was considerable flocculent precipitate and the fluid became browner than that in an uninoeulated flask. The organism at end of 4 months was still pure, living and virulent in flask A, as shown by cultures on agar and on potato, and by inoculations therefrom into cucumber. Eight plants were inoculated by needle-pricks in leaf-blades Nov. 29, and all contracted the disease within 10 days, while 35 check plants remained sound. The flasks B and C presented the same appearance as A, but were not tested until the end of 7 months and 14 months respectively, when the cultures were dead. 290 BACTERIA IN RELATION TO PLANT DISEASES. Sufficient acid was produced in 500 cc. [700 cc] of the fluid to carry into solution 2.5 grams of CaO. This acid is not formic, acetic, propionic, butyric, or any other volatile fermentation acid. Wither did lactic, succinic, or oxalic acid occur in any appreciable amount. No amino acids were found. No alcohol and no volatile aldehydes were present. Glycerin is fermented in presence of Witte's peptone with production of an acid, but this does not occur in the absence of air. Maltose-bouillon became feebly clouded in the closed end of fermentation-tubes after a week, but growth was so feeble that it was ascribed to impurities in the sugar. Similar results were obtained with mannit. In 1906 the experi- ments with maltose in fermentation-tubes were repeated using the melon-strain of the bacillus in peptone water with a maltose 3 times recrystallized in the Bureau of Chemistry. The result was clouding in the open end and U, but none in the closed end. In beef-bouillon and on steamed potato, in an atmosphere of hydrogen or of carbon dioxide, the organism makes some development but grows less well than in the air (fig. 93). Con- firmatory results were obtained with buried agar-streak-cultures (fig. 94). The growth on agar is thin, white, wet-shining, with a distinct margin. The colonies do not grow rapidly. The best growth is usually at the bottom of the streak and in the upper part of the stab. Gelatin (figs. 85, 86), coagu- lated egg-yolk and egg-albumen, and Loeffler's solidified blood-serum are not liquefied. The growth on Loeffler's blood-serum was thin and poor. There was more growth in Dunham's solution made with Savory and Moore's brown peptone, which contains some muscle-sugar, than in that made with Witte's peptone. In Dunham's solution there was less growth than in corres- ponding tubes of Bacillus amylovorus. An organism identified as Bacillus cloacae grew well in Dunham's solution after B. trachciphilus had ceased to grow in it. No indol reaction was observed (3 days) nor was any obtained from Dunham's solution after 15 days' growth. In McConkey's bile-salt agar there was a moderate amount of smooth, wet-shining, surface growth, and some growth in the top of the stab, but no change in the color of the neutral red (20 days). The organism is unable to grow in asparagin-water ;f asparagin-water with dextrose; or asparagin-water with dextrose and nutrient mineral salts (asparagin being the only nitrogen compound). The growth in peptone- water (Witte's) with asparagin was not sensibly greater for some weeks than in simple peptone-water, but finally became greater in one tube, which after- ward yielded a pure culture of B. trachciphilus. This would seem to show that under some circumstances it may get its carbon food from asparagin, but not its nitrogen food. The experiment should be repeated. The organism refused to grow in filtered, boiled river-water containing 1 per cent sodium asparaginate, 1 per cent dextrose and 2 per cent glycerin. Neither would it grow in the same medium when ammonium tartrate or ammonium lactate was substituted for the sodium asparaginate. In another experiment, using distilled water, sodium asparaginate, dextrose and glycerin, and inoculating from potato very copiously, it clouded the fluid in the open end of the fermentation-tube slightly after a week, but never made a good growth. B. trachciphilus will not grow in Cohn's solution, nor in acid (+33) peptonized beef- bouillon (acid of the beef-muscle). It grows in Fermi's solution and in Uschinsky's solu- tion, but rather feebly and with more or less flocculence. Potassium nitrate in p2ptonized beef-bouillon is not reduced to nitrite. The growth of the organism in peptonized beef-bouillon is sensibly retarded by 1 per cent c. ]). sodium chloride, and very decidedly by 1.5 per cent or 1.7 per cent. In most *Fig. 86. — Streak culture of B. Iracheiphilus on gelatin. Drawn from a photograph. Age about 14 days. No liquefaction. tRepeatedin 1006 with same negative result (20 days at 230 C, approximately). Fig. 86.! WILT OF CUCURBITS. 291 instances 2 per cent sodium chloride in peptonized bouillon inhibited growth, but in one instance one tube out of five clouded thinly at the end of 12 days. The alkalinity of the bouillon used for inoculation seemed to play some part, i. e., when inoculated from acid bouillon (+20) even 1 per cent sodium chloride inhibited growth (cucumber strain?). The organism is also sensitive to acids (fig. 95). Fig. 88. t Fig. 89.{ *Fig. 87. — Petri-dish poured plate of B. tracheiphilus after incubating 6 days at 220 to 25° C. On Jan. 27, 1904 with the usual precautions against surface contaminations, a tube of bouillon was inoculated with a little white, sticky slime from the interior of the inoculated cucumber-plant No. .552. Gradually the bouillon clouded thinly after the manner of a pure culture and on Feb. 2, plates were poured demonstrating its purity. The largest colonies were on the surface of the agar; the very small ones, in body of agar; and the thin ones, on lower surface of agar next the glass. Plates made from this same tube of bouillon on Jan. 27, i.e., soon after introducing the viscid slime, remained sterile. fFlc. 88. — Portion nf an agar-plate of B. tracheiphilus showing appearance of buried colonies, surface colonies, and colonies breaking through to surface, after being incubated for 7 days at 30°C. Drawn Nov. 18, 1904, from plate 1 Nov. 11. X2. Descended from rods which withstood freezing. JFig. 89. — Colony of Bacillus tracheiphilus drawn by reflected light and magnified to show markings in the colony, its surface being smooth. This appearance is not visible by transmitted light, nor by looking across colony at an acute angle. Actual size of colony indicated by small circle. Plate I, Nov. 1 1, 1904, made from frozen bouillon and incu- bated for 7 days at 300 C. §Fig. 90. — Surface growth of Bacillus tracheiphilus as seen under a low power by transmitted light, after 7 days at 300 C. Colony from same agar poured-plate as fig. 89. Actual size of colony shown by small circle. Drawn Nov. 1 8, 1904. 292 BACTERIA IN RELATION TO PLANT DISEASES. The optimum reaction for growth in peptonized beef-bouillon is about +8 of Fuller's scale. Growth on the acid side (acid of beef-juice) takes place up to about +28 (?) ; on the alkaline side (sodium hydrate) growth ceases at about —4 (?). These statements are to be taken only as general indications, for here again much depends on the original reaction of the culture-fluid used for the inoculation. Growth may be pushed farthest on the acid side by inoculating from acid bouillons, and on the alkaline side by inoculating from alkaline bouillons, e. g. — 2 bouillon will cloud when inoculated from + 2 or o bouillon, but not when inoculated from +25 or +20 bouillon. The strain used for these experiments was the one which did not infect squash (1905). In milk, growth continues for quite a long time but with no precipitation of casein or change in the appearance of the fluid. In litmus-milk there is little or no change of color, i. c, no decided reddening or bluing of the fluid, or loss of color (reduction) not even after several months. It has seemed to me at times that I could distinguish a slight change in the color of the litmus-milk (bluing), but if any it is so slight as to be readily overlooked. Milk is, therefore, a good differential culture-medium. In old litmus-milk-cultures (dried out two-thirds), on the walls of the tubes above the fluid, very small branching fern-like crystals (fig. 96) occurred after wetting the walls with the fluid, and these crystals did not appear in the 3 check tubes. On litmus-lactose-agar there is no change at first. After some time there may be a gradual deepening of the blue color, but never any reddening. This experiment was repeated in 1906 with the same result. Only one bacillus more sensitive to dry air is yet known, viz., B. carotovorus Jones. In the writers' experiments, portions of solid cultures or fluid cultures were spread in thin layers and dried at room-temperatures on clean sterile cover-glasses and then tested by dropping from time to time into tubes of a bouillon known to be well adapted to the growth of the organism. In all cases the organism was found to be dead one-half hour to one hour after dryingout,andinsome instances when taken from bouillon drying for so short a time as 15 minutes sufficed to kill it. Covers inoculated from the same bouillon and dried only for 10 minutes yielded cultures of the organism when thrown into bouillon. Much seemed to depend on the thinness of the layer. Possibly also the surface on which the bacillus is dried may exert an influence. The bacillus is also sensitive to sunlight (fig. 97.) *Fig. 91. — o, streak culture of Bacillus tracheiphUus on litmus-lactose agar after 7 days, showing frequent tend- ency of organism to grow in discrete colonies. Tube 15, June 30, 1904. Photographed July 7. Xi H- b, Same 8 days old at 220 to 26°C. In this tube it is also growing in the form of a streak. Tube 4, June 16, 1903. Photographed, June 22. X2. Fig. 91. WILT OF CUCURBITS. 293 The organism produces little or no odor. I have never been able to detect any, but the striped cucumber beetle seems to be able to do so. Successful transfers of this organism were made from cultures exposed for twenty minutes to a very low temperature using a mixture of sulphuric ether and frozen carbon dioxide ( — 77° C). At another time several experiments with liquid air ( — 1190 C.) gave the same results, but quantitative tests showed the majority to have been killed. When exposed in bouillon over night to the temperature of liquid air, poured plates showed about 65 per cent of the organisms to have been destroyed (Vol. I, figs. 68 and 69). Tests a half year later, exposing in liquid air for half an hour, showed over 50 per cent to have been destroyed. The minimum temperature for growth is (?) 8° C. In peptone-water, inoculated and kept in the ice-box at 6° to io° C, there was no clouding for 16 days, but on removal to room-temperature (250 to 260 C.) the tube clouded thinly in 48 hours and was subsequently used successfully for the infection of plants (page 271). At another time there was no clouding in 30 days at n° C. to 130 C. The optimum temperature is, roughly, 250 to 300 C. The maximum temperature is 34 to 350 C. (?) i. e., not determined ac- curately but somewhere around this point. In October, 1905, eleven tubes of +15 peptone bouillon (stock 1622) were inoculated with the cucumber strain (from acid bouillon?) and exposed in the thermostat for 5 days at 330 C. and lower, gradually rising to 360 C, but most of the time under 350 C. (The thermostat was stable but the night temperatures were not recorded, only assumed to have been like the day tem- peratures.) All of the tubes remained clear during the exposure and none of them clouded when removed to room temperature (250 C). Two checks held at 250 C. clouded in 48 hours. This experiment was repeated three days later, paying careful attention to the night temperatures. Twelve tubes of bouillon were inoculated and exposed in the same thermostat for 445 hours, after which they were removed and placed at 250 C. Two tubes were held as checks. The latter clouded in 48 hours. The heated tubes never clouded. The recorded thermostat tem- peratures were as follows : Oct. 15, Oct. 16, Fie. 92.! Oct. 14. 4 P-m. (after opening 34 io°C 6 p.m. 34 5O°0 9 p.m. 34 5QUC 12 p.m. 34 ,io"C Oct. 15, 4 a.m. 34 50" C 6 a.m. 34 50UC Noon 34 6o°C 5 p.m. 3550° C g p.m. Midnight 4 a.m. 6 a.m. 3550° C 3550° C 36° C. 36° C. 1 1 a.m. 12:30 p.m. 355o° C 3580° C *Frc. 92. — Growth of B. Iracheiphilus in fermentation tubes: a. facultative anaerobically in cane-sugar bouillon, and, b, aerobically in meat-infusion with 1 per cent dextrine. The dextrine used was readily soluble in water and did net give a red reaction with iodine. Tube containing cane-sugar was inoculated Feb. 3, 1896, very copiously from tube 4, Jan. 2 1 , a slant tube of sugar-agar inoculated from plant No. 246. The tube was doubtfully clouded in the open end on Feb. 6, and plainly so on Feb. 7. On Feb. 8 it was thinly clouded in the whole of closed end but the fluid was still alkaline. On Feb. 13 clouding was uniform in open end and closed end (nearly so on Feb. 10) and fluid was acid to litmus. On this date a transfer to potato yielded a pure culture of B. Iracheiphilus. On Feb. 27 culture was dead, having been destroyed by its own acid. The dextrine-bouillon was inoculated May 8, 1895, and was clouded in open end May 13 but clear in closed end. On May 18 it was still clear in closed end and fluid was alkaline to litmus. 294 BACTERIA IN RELATION TO PLANT DISEASES. The thermal death-point is 430 C, the lowest yet recorded for any organism infesting plants, and until recently the lowest known. In an experiment made in June, 1896, an exposure of one hour to 410 C. killed all, i. e., no colonies developed on the agar plate poured from the heated bouillon (1 1 days incubation) whereas the check-plate, i. 1 Cross section of petiole of cabbage, showing every bundle blackened by Bad campestre, the paren- chyma being free. Slide 8, plant No. 42. PLANT BACTERIA, VOL. 2. PLATE 18 Ragged leaf disease of cabbage associated with a slow bacterial decaj- of margins of leaves (not due to Bad. campestrc). No heads were formed. Hothouse Jan. IN, 1906. / BLACK ROT OF CRUCIFEROUS PLANTS. 305 In Wisconsin, Russell & Harding repeated the experiments of Pammel, and of Smith, and confirmed their conclusions respecting the bacterial nature of the disease. Harding also studied the morphology and physiology of the organism quite carefully. The number of their successful infection experiments amounted to several hundred. In New York and in Europe, Harding subsequently continued his studies. On his return from Europe he obtained numerous successful infections on cabbage and cauliflower with a culture which he isolated from a diseased cabbage-plant found by him in Switzerland. Comparative tests were also instituted and neither in its cultural characters nor in its infectious proper- ties was any difference detected between the Swiss organism and the one from New York or Wisconsin. Hecke subsequently discovered the disease in kohlrabi in Austria and published two instructive papers on it, the number of his infection-experiments exceeding 100, Rus- sell, however, was the first to obtain the disease in kohlrabi by inoculations (Bull. 65, p. 22). Van Hall then studied it in cabbages in Holland. More recently Brenner in Basel, under direction of Dr. Alfred Fischer, investigated the etiology of this disease, and after experimenting for two seasons states that he can only confirm Smith's conclusions respecting the cause of this disease. The writer has himself isolated this or- ganism from diseased plants obtained from Illinois, Wisconsin, Michigan, Ohio, Pennsylvania, Western New York, Long Island, Maryland, Alabama, Florida and Texas, and has studied the disease in the field in half a dozen different States. He has also produced the disease in cabbage by inoculating with a pure culture of the organism received from Hecke, who iso- lated it from kohlrabi grown in Southern Austria and himself proved its infectious nature on a variety of crucifers. In most of the above mentioned isolations by means of poured plates, Bacterium cam- pestre was found in the vessels of the plants in pure culture. Only occasionally were mixtures obtained and even then the yellow organism was the preponderant form. Plants are attacked by this disease in all stages of growth from seedlings in the seed- bed to plants ready for the market. In all this class of plants most of the infections, perhaps all of them in plants beyond the seedling-stage, are through the parts above ground and generally by way of the leaves. The writer, who has spent many weeks in cabbage-fields, has never seen anything in midsummer or later suggestive of infection through the root- system, i. c, roots diseased and parts above ground not diseased, and Hecke's experiments of growing plants in soil mixed with tissues swarming with the organism, yielded only nega- tive results. So did my own. Brenner also made similar experiments with similar results: R? *~ c '!-■ ^^^ ■ .jr.. ■ -■'■-. 3H x; . .'/^B WW *'* In *""" 1 m ^•^^^■^r ^^^■1 ^^^^r R* * W^^W^m miv' W ' m **m-: |U£. * .3 &^m ;: -. Fig. 103.* *Fig. 103. — Longitudinal section and cross-sections of stems of cauliflower diseased by Bacterium campestre, showing stain and cavities in the stems. Miami, Fla., March, 1904. 306 BACTERIA IN RELATION TO PLANT DISEASES. The plants did not contract the disease even when the root-system was wounded. Russell also states that the disease does not find its way into the plant through the root-system. Stewart and Harding believe, however, that it may be communicated through the root- system, and this is not at all unlikely in early stages of growth when the tissues are soft. Later the woody stem offers an impassable barrier. Potter states that the disease has occurred repeatedly in Northumberland, England, in Swedish turnips, but that he observed it only on the root after it was well developed and always beginning in a local root-injury of some sort. Infections above ground take place readily through wounded surfaces and the organism which causes the disease may be disseminated by a variety of leaf-eating insects (fig.108), ■J p^^5"*1^! F-V /* M' ^ L V ^^ W I w • H W" "* W ^O H • > W) ~ \ ■>.:«•>"; -t /.?;:■ \--<~-' ;.• .J [p** -' ■ *j • . ' V' , /' ->- :i * — Bl^- m l ^H V^pf^ ^^B 1 " - ^ .'jp* • •.. ..B ■l,-' ^^mmm^M Fig. 104.* either by being introduced directly into wounds or more often perhaps by being left here and there on the uninjured margins of the leaves subsequently to find its way into the plant in the manner next to be described. Probably the leaves of plants are occasionally infected from the dust of the fields. The disease does not appear, however, to be one which is spread widely through the medium of the air. At least, as already recorded (Farmers' Bulletin, January, 1898) the writer has seen fields nearly free from the disease with only a fence separating them from fields in which half the plants were badly diseased and had been for many weeks, while multitudes of new infections were taking place right and left. This field also contained a multitude of infected weeds (charlock). In hothouse experiments the writer succeeded in transmitting the disease by means of the larvae of the cabbage butterfly (Plusia) and by slugs (Agriolimax). Brenner con- *Fig. 104. — Sections of kohlrabi, showing blackened vascular bundles due to Bacterium campestre. Photographed by the writer at Miami, Fla., March, 1904. The right and left were from different plants. See fig. 107. BLACK ROT OF CRUCIFEROUS PLANTS. 307 firms infection by mollusks and reports successful transmissions by aphides. He recovered a yellow Schizomycete from the body of an aphis which had punctured a diseased spot, and with this organism he induced the disease on sound plants. A curculio (Contrachelus) , which lays eggs in the stems of young cabbage plants, is also open to suspicion. Any creature which gnaws diseased leaves or stems and then gnaws or even crawls over healthy ones is very likely to transmit the infection. It is desirable that further studies should be made, especially on plants in the seed-bed and soon after transplanting, particularly with reference to underground infection. Wounds are not necessary, however, for the transmission of this disease, nor do the majority of cases arise as a result of trauma. The greater part of the infections (Smith, Russell, Hecke) unquestionably take place through certain natural openings of the plants, known as water-pores. These are modified stomata which occur in groups on the teeth of the leaf and through which excessive moisture taken up by the root- system is extruded from the plant. When the air is warm this moisture is given off as an invisible vapor, but during cool nights it gathers on the leaf-serratures in drops like dew, and may persist for hours after sunrise. It was experiments with slugs which led the writer to the discovery of water-pore infections. On leaves which had been bitten and infected by Agriolimax, a few belated infec- tions appeared on the leaf-margins where no wounds could be detected. A study of these infections showed that they began in the leaf serra- tures. This placed the question of natural infection in a wholly new light and led to further experiments with the results already known (vide Centralb. f. Bakt. 2 Abt., 1897, page 411). Up to the time of the preparation of that paper the writer had not studied this disease in the field and his conclusions were based only on experimental data. It was therefore observed with the greatest interest, in the summer of 1897, how well thefield observations bore out the conclusions of the laboratory and hothouse. In the last (second) edition of his " Vorlesungen " Fischer has questioned the possi- bility of general infection through the water-pores on two a priori grounds: (1) There is very little nutrient material in the fluid extruded from the water-pores; and (2) it would Fig. 105.* *Fig. 105. — Stems of collards (Brassica oleracea f gemmifera) from Tampa, Fla., badly affected by the black-rot. Pure cultures of Bad. campestre were plated from the interior of young shoot at point marked X (see fig. 128). Photo- graphed Sept., 1902. Natural size. 3o8 BACTERIA IN RELATION TO PLANT DISEASES. require quite a good manyhours for a dust-dry bacterium to beeoraesufficientlymoistenedso as to multiply and enter the sub-stomatie chamber. A sufficient reply is that : (i) the organ- ism does multiply considerably in this extruded fluid, as the writer demonstrated in vitro in 1897, "very little" nutrient material being sufficient, and (2) the hypothetical, dust-dry, wind-borne bacterium requiring a half day or more to moisten it, is probably not the one that usually enters the water-pores and induces the disease, but rather a fresh germ recently come from the interior of some affected leaf as an extrusion from some water-pore already diseased, or left in the vicinity of the water-pore by some wandering insect, which during its feedings on diseased leaves has first contaminated its own body and then various un- injured parts of the same plant and of other plants; such a bacterium would be ready to grow as soon as it found lodgment in a moist place. These mountains of difficulty therefore disappear as soon as the actual conditions are known. Water-pore infections take place only when the weather conditions are such that the extruded fluid from the plant remains over the water-pores for some hours in the form of drops. Moist weather with a day temperature of 200 C. appears to be very favorable for in- fection. Under these circumstances if any living rods of this organism happen to be lying in the vicinity, so as to be wetted, they are stimulated into growth and, being motile, they find their way readily into the substo- matic chamber. Proof of water-pore infections was furnished by the writer in August, 1897, and subsequently by Russell, by Hecke, and by Brenner.* Hecke made water-pore inoculations on 14 kohlrabi plants, of which only 2 were entirely negative while 8 were very successful. The period of incu- bation, that is, the time from the entrance of the organism to the ap- pearance of the disease in the veins of the cabbage-leaf, is usually several weeks (11 to 20 days in kohlrabi, Hecke) this being the period required for the multiplication of the bacteria in the substomatic chamber and their passage through the intercellular spaces of the epithem into the vessels of the leaf. Generally, however, in artificial inoculations there is a slight darkening of the infected leaf-tooth as early as the sixth to tenth day. For illustrations see Vol. I, figs. 76, 77, 78, 79, 87, 115, 116, 1 17. These infections were obtained by atomizing upon the plants in inoculation cages (Vol. I, fig. 95) agar cultures diffused in water. Russell placed the bacteria in drops of water extruding from particular water-pores. Hecke tried both methods successfully. The writer's first successes were by plunging leaves into water containing the bacteria and allowing them to remain for some hours. Brenner likewise obtained waterpore infections by this last method, and also by placing the bacteria *In fluid collected from water-pores Brenner found the organism multiplied twentyfold in the course of 10 days. Russell also collected several cubic centimeters of fluid from the water-pores, inoculated it with Bacterium campeslre and made poured plates. The second series of poured plates made 12 hours later "showed many more colonies of the specific germ, thus indicating that the bacteria originally seeded were able to grow in the water." tFlG. 106. — A small turnip root, showing center rotted out by Ba< terium campeslre. An old plant, but no normal expansion of root. From a field near Baltimore, Md., Sept., 1896. x circa 8. ^^S ^ ■ . ■■-■'/// M-.t WW BLACK ROT OF CRUCIFEROUS PLANTS. 309 on particular water-pores which were extruding fluid. Once past the incubation period, the downward progress of the disease is comparatively rapid (1 cm. or more per day), especially if the weather is warm and the soil is moist enough to induce a vigorous growth of the host- plant. The progress of the disease in cool weather and in plants making a slow growth is less rapid (plate 2). On September 7, 1897, during weather very favorable to the progress of the disease, the writer examined a cabbage-plant at Racine, Wisconsin, which bore 170 separate water-pore infections all spreading rapidly. As yet there was no disease of the root, stem, or interior of the well-formed head, nor was there any black stain in the base of any of the petioles. These infections, therefore, probably took place not much earlier than the first of August. Such a plant might be expected to be badly rotted in stem and head in course of another six or eight weeks. Brenner states that when he inoculated cabbage-plants on a single leaf-tooth many other groups of water-pores subsequently contracted the disease, the organism being pre- sumably carried along the moist surface of the leaf-margin to the other groups of water-pores by capillary attraction. As noted in 1898, the writer saw one large cabbage-plant which bore more than 400 distinct marginal leaf-infections, while many other plants in the same field and in other fields showed from 50 to 150 such separate infections. Sometimes nearly every serra- ture on a leaf would be infected. The writer has dissected hun- dreds of cabbage-leaves which were attacked at the time of observation only on their mar- gin, and other hundreds in which the organism was already well into the middle of the leaf or had already entered the stem, as determined by cutting. An extensive marginal infection obtained by spraying, with the entire absence of infection by way of the ordinary stomata is shown in fig. 10. For an early stage of water pore infection see fig. 130a and Vol. I, fig. 87; for a late stage with disorganization, this volume, fig. 11. The leaf surface of many crucifers is covered with a waxy bloom repellant to water, and this preserves the leaf from injury when submerged for some hours, and also undoubtedly makes it difficult for the bacteria to enter through the ordinary stomata. At least, infections through such stomata have not been observed. Russell's observations and experiments agree with those of the writer. In case of leaf-infections by way of the water-pores the general progress of the dis- ease is downward in the spiral vessels of the leaf (Vol. I, fig. 76, 77, 78). These are com- monly filled densely, many of them at least, by the rapid multiplication of the organism. From the blade of the leaf the bacteria pass into the leaf -traces of the petiole (fig. 109, no) Fig. 107.* *Fig. 107. — Cross-section of a small portion of a blackened kohlrabi bundle, showing cavity due to presence of Bacterium campeslre. Section from enlarged edible part of plant. Material collected at Miami, Fla., March, 1904. x 500. Slide 293 b 18. 3io BACTERIA IN RELATION TO PLANT DISEASES. and thence into the stem where, in turn, the vessels of the stem are occluded and browned. vSubsequently if the tissues are soft enough the organism passes up and down the stem and out into other leaves, always by way of the vascular system. The rapidity of movement in stems depends on their texture. In hard woody stems the bacteria move with extreme slowness or are entirely hemmed in; in soft juicy stems progress resembles that in the petiole. In leaves which are infected from the stem (fig. 98), the entire leaf-blade may be attacked almost at once and in that case may show signs of wilting. The writer has fre- quently seen cabbage leaves become flabby, unjoint and fall off while the bacteria were still confined to the petiole, such leaves having been infected by way of the leaf-traces as the result of stem-inoculations. In these cases so many leaf-traces were involved that the leaf was unable to obtain the necessary water-supply. More often some of the leaf-traces are not involved and the leaf manages to function more or less imperfectly for a considerable time. In such a leaf a part of the veins in the leaf-blade are always blackened considerably in advance of the remainder and wilting may not occur. The writer tried passing 1 per cent eosine water up such petioles by transferring them to the red fluid after cutting them under water. In many cases the eosine only passed up the unob- structed vessels, but whether failure to pass up the bacterially occluded vessels was due simply to the occlusion, or must be ascribed in part to the destruction of the vessel- walls by the bacteria, was not determined. Whether the first signs on the expanded portion of such leaves are basal or terminal, or on one side or the other of the blade, depends en- tirely on which leaf-traces are entered first, dif- ferent ones ramifying to different parts of the leaf (Smith, Hecke). In the end, such leaves are so badly affected that they unjoint and fall from the stem, without, however, any signs of soft rot. It is a slow dry-rot even in turnip-roots. When soft rot or extensive sloughing of the parenchyma intervenes, especially if it begins at the surface, we may at once suspect complications due to the presence of other organisms (see the soft rots). When inoculations are made on the midrib of a leaf, Brenner states that the bacteria pass up- ward faster than downward. The writer recorded the same fact for inoculated cabbages in 1897, and observed it again particularly in 1906. The writer has frequently observed the inoculated side of the plant to become diseased almost to the exclusion of the other side, but has observed nothing suggestive of the rapid transportation of the bacteria for long distances in a liquid moving stream such as we sometimes conceive to be present in the vessels of a plant. In a plant inoculated on the stem under the fourth leaf Brenner observed the fourth, fifth, and seventh leaves, which were on the inoculated side of the stem, to contract the disease sooner than the sixth leaf, which was on the opposite side of the stem (see cucumber wilt, p. 219). Brenner endeavored to force stained bacteria up cabbage- *FlG. 108. — Small portion cf a cabbage-leaf near margin, showing how black venation due to Bacterium campestre is frequently restricted for a time to angular areas formed by larger veins. This infection may have started from an insect bite at a, or may have run in from a group of water-pores; at b, isanothei inseel bite. Specimen from a cabbage field in Western New York. Drawn from a photograph. About natural size. Fi«. 108/ BLACK ROT OF CRUCIFEROUS PLANTS. 311 petioles, but neither by suction nor by pressure with mercury could he get them higher than about 2 cm. The stain passed farther but not the bacteria. According to Hecke the bac- terial infection passed upward in certain inoculated kohlrabi leaves 5 cm. in the time required to pass downward 2.5 cm. This more rapid upward movement is attributed by him to the effect of the transpiration stream. Weather conditions favorable to rapid growth are also favorable to the spread of this disease. When two sets of plants are inoculated in the same manner, the one receiving large quantities of water and the other less amounts, the former contract the disease more readily and the organism also passes through the tissues with greater rapidity (Russell). Varie-tal resistance to this disease, and the resistance of particular individuals within the variety, are subjects deserving of careful attention. Russell states that "In all probability there is but very little dif- ference in susceptibility, all varieties readily yielding to the disease, if the causal organism is once present." Many cabbage growers think differ- ently, but so far as the writer has had opportunity to examine into the matter himself he has found the statements of particular growers that this or that variety was specially subject based only on isolated obser- vations and easily overthrown by other observations in the same lo- cality or some other. In experiments with kohlrabi Hecke found that the "Vienna Glaskohlrabi," which ma- tures quickly, is less likely to show the disease in the fleshy part as the result of leaf-inoculations, than is the slowly maturing variety known as "Goliath." Generally the farmer thinks that variety most subject which he happens to have planted on infected land, while as a matter of fact on the very next farm the con- ditions may be reversed. Nothing here said should be construed into a denial of difference in behavior, but only regarded as a reason for caution in drawing conclusions, it being quite in line with what we know of other diseases to suppose there are tendencies to resistance, especially in particular plants, which might be increased and made of economic importance. According to Russell, cauliflower is the most susceptible plant, while turnips and rutabagas are not very susceptible. The writer found radishes rather resistant. In North Holland, according to Ritzema Bos, the disease was most prevalent in red cabbage in 1900, but in Savoy cabbage in 1901. Fig. 109.* *Fig. 109.— Portion of a cauliflower-petiole in cross-section, showing destruction cf the xyJem portion of several bundles by Bacterium campestre. In lower left-hand part of figure bacteria may alsc be seen wedging apart parenchyma cells. Re'sultof a pure culture inoculation made into blade of leaf by means of needle-pricks. Material fixed in alcohol, infiltrated in paraffin, sections stained with carbol-fuchsin. and drawing made from a photomicrograph, x 190. Slide 1 18-5. For a small portion of this section more highly magnified see fig. 1 10. For a longitudinal section through a similar cauliflower-petiole see vol. 1, fig. 7. 312 BACTERIA IN RELATION TO PLANT DISEASES. Since the above paragraphs were written S. F. Edwards has reported (1908) that the Houser cabbage "is practically immune to black-rot under field conditions." Even when pure cultures of the bacteria were inoculated into the cabbage the inoculations were either without result or the disease advanced so slowly as to do but little injury. The period of incubation is variable. Hecke, inoculating by needle-puncture, obtained the first signs of the disease in from 7 to 28 days on leaves, and in from 9 to 31 days on stems. He made t,^ inoculations on kohlrabi leaves by needle-puncture, every one of which was successful; he likewise inoculated 23 kohlrabi plants in the stem by needle-puncture, and Fig. 110.* of the whole number there was not one which did not give some indications of disease. Exclusive of sprays, plunge-experiments, and the use of insects, etc., almost all of my own inoculations were made by needle-puncture without hypodermic injection, and the first distinct signs of disease were generally visible in 14 to 21 days. Brenner also found this period of incubation correct for most of the plants he experimented with. On cotyledons, however, he obtained signs of the disease in 8 days, and the entire plants soon contracted the disease and were destroyed or greatly injured. *Fig. no. — Cross-section of a cauliflower-petiole showing bacterial cavity in a small bundle (lower one at left in fig. 109) due to presence of Hiiitcriuni itimpeslre. From a pure culture inoculation. A paraffin section stained with carbol-fuchsin. Enlarged from a photomicrograph. BLACK ROT OF CRUCIFEROUS PLANTS. 3*3 The greatest contrast to this prompt destruction ever obtained by the writer was on cabbage-plants dwarfed and forced to make a very slow growth by keeping them for a long time in 4-inch pots. These plants, which were inoculated in the autumn, developed the disease, became stunted, and then appeared to grow out of it, but on some of them it reappeared the next summer at the top of the plants in young leaves which were unques- tionably infected from the stem by way of the leaf-traces. Fifteen months from the date of inoculation, and more than 30 cm. above the point where the needle entered, the organism was recovered in pure culture from the woody stem of one of these tall spindling plants. There can be no doubt that the severity of the disease varies with varying seasons. In moist warm seasons the disease often makes a clean sweep on fields which may yield a crop the following season, provided the weather is dry enough to induce slow growth and to prevent wholesale infection by way of the water-pores. Owing to its wide distribution, and the ease with which infections may be obtained on a variety of plants, this is a very good disease for the use of classes in schools. For experimental or demonstration purposes there is a choice in the parts to be inocu- lated. Young rapidly growing leaves and plants are better than old or slow growing ones, and infection by needle-pricks generally succeeds best when made into the upper fleshy part of stems immediately under leaves or when made into the midrib. Cabbage plants when inoculated on the lower leaves often throw off these leaves before stem-infection has taken place. Brenner notes that when he inoculated into the petiole of the leaf this was frequently thrown off in course of a few weeks. He found inoculations on secondary veins or peripheral veins less successful than those on primary veins. *Fig. in. — Cross-section of root of inoculated turnip plant (No. 53), showing two reticulated vessels, one of which is occupied by Bacterium campestre while the other is free (nearly). Surrounding parenchyma cells are free from bacteria. Drawn from a photomicrograph which was made from material fixed in alcohol, infiltrated with paraffin, sectioned on the microtome and stained with carbol-fuchsin. Contents in cells at right are protoplasmic, x 1000. 3H BACTERIA IN RELATION TO PLANT DISEASES. In experiments with Bacterium campestre on kohlrabi Heeke found that when the plants were kept excessively moist under bell-jars some parts of the leaf-edge became water- logged (glossy and darker colored) from an excess of water. These parts subsequently died but the plants did not contract the disease from water-pores situated in such suffocated areas. Consult chapter on Angular Leaf-spot of Cotton for similar observations by the writer. The leaves of maize seedlings also frequently become water-logged without con- tracting Stewart's disease. Query. — Why do the bacteria multiply so abundantly in the vessels? MORBID ANATOMY. This organism causes no hyperplasias. After it has gained an entrance, which must be ordinarily through parenchymatic tissues (epithem, etc.) the parasite is confined for some time pretty closely, although not exclusively, to the vascular system and even to particular leaf-traces or bundles especially to the spiral and reticulated vessels which are very often filled with incalculable numbers of this organism (figs, in, 112). When such a state of occlusion exists, especially in juicy parts, the walls of the vessels are destroyed in places (dissolved?) and the organism finds its way into the surrounding parenchyma, but never or almost never to the surface of the plant. Progress through the parenchyma is slow, apparently on account of its acidity. Often the intercellular spaces are first occupied (figs. 113, 114); the middle lamella is then dissolved (figs. 115, 116) and the elements are separated and squeezed into all sorts of shapes by the multiplication of the bacteria (fig. 117). Subsequently the wall proper of the cell becomes thinner and vaguer and finally seems to disappear altogether. There may be some doubt, however, on the latter point, i. e., as to the final complete solution and disappearance of the cellulose. The dark band at the bottom of fig. 1 10 is probably formed of com- pacted cell walls crowded out of the cavity. Similar cell-walls crowded out of the center of the cavity may be seen in fig. 1 1 7 as white lines. Lignified tissues are not dissolved although Brenner states that they are. This statement probably rests on some misinterpre- tation. The spiral threads and other distinctly lignified portions of the bundle persist. It is the destruction of the surrounding non-lignified tissues which gives rise to the large cavi- ties in turnips and other susceptible tissues. The formation of cavities by this organism is very common in a number of host-plants (figs. 109, 117, 118 and Vol. I, figs. 6 and 7) and all stages of the separation and destruction of the cells may be studied to good advantage in turnip-roots and cabbage-leaves or cauli- flower-leaves, especially the petioles. The occlusion of the vessels and the formation of cavities may also be made out very satisfactorily in kohlrabi and rape (figs. 107, 119). These cavities always begin in the vascular bundles and are occupied by the bacteria in incalcu- lable numbers, scarcely anything like it being observed in the animal kingdom. Small parenchyma-cavities sometimes appear around the smaller blackened veins in leaf-blades, but more conspicuous ones are to be found in the fleshy midrib and petiole. As a whole the parenchyma of the leaf-blade appears to be too dry or too acid for this organism. The pith Fig. 112.* *FlG. 1 12. — Bacterium campestre occupying a reticulated vessel in a turnip-root as result of a pure culture inocula- tion on blades of leaf. Same vessel as shown in fig. 120 but a little farther down. Drawn from a photomicrograph. x 500. BLACK ROT OF CRUCIFEROUS PLANTS. 315 offers a more favorable substratum. Occasionally in cabbage and collards the entire pith of a stem disappears (fig. 105) and in turnips it is not uncommon for cavities in the roots to occupy a large portion of the interior (fig. 106). Although, as stated, parenchyma is destroyed by the wedging apart of its cells, there are other ways, i. c, it is not infrequent, as shown in fig. 120 and on cross-section in Vol. I, fig. 5, for non-lignified cells surrounding a vessel to be entered and filled by the bacteria rather than to be crushed and crowded out of the way by external multiplication. The cell-wall appears to be intact as shown in the drawings and clearly no great amount of it can be dissolved. It is not easy, therefore, to make out exactly the method of entrance. Probably the bacteria enter these particular cells by way of pits, the vessel being first filled by the organism which then either dissolves the thin separating membrane of the pit or softens and ruptures it. Harding and Brenner both mention the occasional presence in the bacterial cavities of granules, which do not stain like bacteria, and the origin of which is somewhat doubtful. Hecke states that he found such granules in undestroyed vessels just beyond the advancing margin of the bacterial mass, as determined by serial sections. This substance stained with magdala red but did not retain the iron-haematoxylin. The material was fixed in a mixture of formalin, wood-vinegar and wood-alcohol. Brenner is inclined to consider these granules as in the nature of bassorin and derived from the decom- position of the host-tissues. The writer be- lieves some of them to be dead and more or less disorganized bacteria, which for this reason do not take stains well. Such granules occur in great numbers in sugar-cane attacked by Bac- terium vascularum. See also Symbiosis, page 1 1 1 . They offer a good subject for further research. There are no gaps in the bacterial occupa- tion of particular vessels and, consequently, as Hecke suggests, movement of the organism in the tissues, probably may be assumed to be due to growth rather than to self-motility. The black stain always follows the bacterial occu- pation, rather than precedes it, but the bacteria are rarely more than 1 or 2 cm. in advance of the pigmentation, so that to a good degree absence of brown stain in the vascular bundles may be taken to denote absence of the bacteria. f This black or brown stain may be located *Fig. 1 13.— Cross-section of small portion of cauliflower-petiole, showing parenchyma to left of fig. 109. Inter- cellular spaces occupied by Bacterium catnpestre. The cells themselves are entirely free. Section stained with carbol fuchsin. Slide 1 18-5. The bacteria entered this tissue by way of one of the bundles in which there is a large cavity. jRussell says "the causal organism can frequently be isolated at a point 2 or 3 inches in advance of the darkened tissue." (Bull. 65, p. 23.) This I have not been able to verify and am inclined to think it is a mistake. Errors may easilv occur, since in one of the author's own examinations of leaves infected on the blade no brown stain was detected in a fresh-cut petiole on cross-section, at a certain level, using a good hand-lens, but was plainly visible to the naked eye more than 2 inches farther down (away from the point of inoculation) in a few vessels of one bundle on mashing the tissues for poured plates, and was then detected on the original cross-section, the surface of which had meanwhile become dry and somewhat lighter colored. These browned vessels which contained numerous bacteria, were 1 1 cm. below the place of inoculation, and 6 cm. below the point where the brown stain in the vessels was at first supposed to have entirely disappeared. The poured plates yielded numerous colonies of Bacterium campestre. j$3 A m Igljt A^Sv jpMpr?. -J; -V y S- -y ffcw. c fcQ\ 0Z$2. ,u,uo III 1 : ' w s£v » 'ion mm Fig. 125.* \ 'loo mm / » * Fig. I26.f Fig. I27.| of Bact. phaseoli of the same age Bact. campestre showed more liquefaction and stain, but not more growth : By reflected light the contrast in color of the serum, white vs. brownish, was decided (tests of May , 1 909) . Subsequently this contrast became less. It is an organism rather sensitive to acids, even those derived from plants. Complete data (quantitive) are not available. Further experiments should be made. According to Harding the vitality of the organism is lessened by long cultivation, i. c, it liquefies gelatin more slowly and is less resistant to heat and to desiccation. It destroys the middle lamella of cell-walls and possibly(?) on prolonged action the cellulose of crucifers, but not lignified tissues. It has no solvent action on Swedish filter-paper. It produces indol slowly in sugar-free peptonized beef-bouillon or peptonized Uschinsky's solution ; it does not reduce potassium nitrate to nitrite when grown in peptonized bouillon or with cane-sugar in Fischer's solution. The organism had no characteristic odor, except (Harding) the strong odor of crucifers when grown on these substrata and in bouillon an odor of sweet corn. *Fig. 125. — Bacterium campestre from a potato culture kept 4 months in ice-box at 120 C. Fluid at bottom of culture was filled solid with bacteria, but the growth was still a fresh yellow color. Drawn unstained from a hanging drop with 2 mm. objective, 1.30 11. a., and 12 compensating ocular, Mar. S. 1905. |Fio. 126. — Bacterium campestre, drawn unstained from a hanging drop Mar. 9, 1905, after 48 hours in beef bouillon at 3o"C. The organism was actively motile and short; termo-like paired rods were the common form. No long rods were observed. Bouillon was thinly clouded. It was made from a typical culture on potato, i.e., thai which furnished material for fig. 125. JFio. 127. — Bacterium campestre, from a very old potato culture (brownish yellow slime stirred up in water) made Mar. 8, 1905, from a culture inoculated Oct. 4, 1904, and kept in a refrigerator a long time at 12° C. A larger proportion of the rods were vacuolate than are here shown. BLACK ROT OF CRUCIFEROUS PLANTS. 323 vSo far as known it is strictly aerobic, i. e.,it does not produce gas or grow in the closed end of fermentation-tubes in peptone-water or peptonized beef-bouillon with any of the follow- ing carbon compounds: grape-sugar, fruit-sugar, cane-sugar, galactose, milk-sugar, maltose, dextrin, mannit, glycerin ; neither will it grow in the closed end in potato-broth, cabbage- broth, or cauliflower-broth; nor with nitrates (Harding). If any acids are produced, the pres- ence of air is required and theyare readily obscured by the production of alkali (ammonia). It is not conspicuous as a reducer of litmus. Its reducing powers are variable. Occasionally some hydrogen sulphide is formed. In cabbage-broth containing litmus the organism *Fig. 128. — Petri-dish cultures of Bacterium campestre, showing character of colonies and effect of crowding on size. Cultures 8 days old at room temperature. Figs. 1 and 2 contain crystals due to growth of organism in +15 agar. Small dots are buried colonies; medium-sized faint colonies, as in the center of 2, are thin expansions of the same organism between agar and bottom of dish. These poured-plates were made directly from blackened vascular ring of a young shoot of collard shown in fig. 105 (at the point marked x). Natural size. 324 BACTERIA IN RELATION TO PLANT DISEASES. reduced the litmus in course of a few days, except at the top. On some vegetable media, and with various sugars, a feebly acid reaction was sometimes detected, but the nature of the acid is unknown. Possibly sometimes the writer may have had contamination in his cultures, since Harding has found in some of his cultures a contaminating organism having the group number 211.2223532. The subject is open to further study in which connection the interesting pages 31-33 of Harding's paper (1910) should be consulted. It does not grow luxuriantly in Fermi's solution, Uschinsky's solution, or Cohn's solution, usually it does not grow at all in the latter and when it does there is no fluorescence. In Fermi's solution after 2 weeks there was thin clouding, no pellicle, and a scanty pale pre- cipitate. It will not grow in an atmosphere of hydrogen, nitrogen, or carbon dioxide. In vacuo it also grows feebly in proportion to the completeness of the exhaustion of the air. It will not grow in peptonized beef-broth to which chloroform has been added. This experiment was repeated in February 1906, with the same result (fig. 130). Four tubes inoculated February 12 remained entirely free from clouding (February 24). It is promptly killed in agar plates by direct sunlight (30 minutes or less). It produces a brown pigment soluble in water, and a yellow pigment insoluble in water but soluble in glycerin, ethyl alcohol, methyl alcohol, acetone, ammonium carbonate, or glacial acetic acid. This yellow pigment appears to be associated with a fat, i. c, it is a lipochrome. Harding states that the yellow pigment is soluble in ethyl and methyl alcohol, is unchanged in glycerin, and is darkened in carbon-bisulphide, xylene, gasoline and chloro- form. He found it slowly destroyed in dilute acetic acid, and destroyed in sulphuric ether, dilute hydrochloric acid, sulphuric acid, and nitric acid. The writer found the yellow pigment bleached by contact with carbon-bisulphide, xylol, toluol, ether and chloroform. The color is lodged in the organism itself. The brown pigment is not formed in beef-broth, or in peptone-water with grape-sugar. The minimum temperature for growth is 50 C. or thereabouts. Its optimum temperature is 300 C. or thereabouts. Its maximum temperature is 380 to 390 C. The thermal death- point is 5i° + C* It tolerates sodium hydrate in peptonized beef-bouillon up to —40, and plant acids up to -f 30 or +40 1 ?). Young cultures stain readily with various basic anilin dyes. Harding usually obtained a polar stain with Ziehl's carbol-fuchsin. In agar cultures 20 days old he says many of the individuals stain feebly with methylene blue unless it is heated or applied for a long time. In sections of the tissues the organism stains very satisfactorily with Ziehl's carbol-fuchsin (3 to 5 minutes' exposure), with nigrosin.and with Heidenhain's iron haematoxylin. Hecke reports better success with Benda's iron haematoxylin than with carbol-fuchsin, i. c, clearer sections. With this stain, by proper differentiation, it is possible to obtain fine contrasts, i. c, the bacteria remain black on a pale background. He used weak acetic acid after carbol-fuchsin for differentiating. Good contrasts may be secured also by a suitable contrast stain, e. g., methyl green (2%) in water 18 hours; solid green, sat. water sol., 1 minute. Brenner states that the organism grows readily in Fischer's nutrient mineral solutionf with cane-sugar for the carbon food and nitrate of potash as the only nitrogen food. In other words, following Fischer's classification, it is nitrobacterium. It also grows well, according to Brenner, in Fischer's nutrient mineral solution with addition of grape-sugar and asparagin or ammonium tartrate as the nitrogen food. It grew moderately in the same "Harding reports great variations in the thermal death-point, depending on age of culture, length of time grown on artificial media and temperature at which culture was grown. His highest thermal death-point is 52° C, the lowest 440 C. For the writer's method of making thermal death point tests see Vol. I, page 77. ... P- cL tDistilled water 100.00 Dipotassium phosphate 1 .00 Magnesium sulphate 0.20 Calcium chloride 0.01 BLACK ROT OF CRUCIFEROUS PLANTS. 325 mineral solution with cane-sugar and ammonium chloride as the nitrogen food. It grew feebly in these media when glycerin was substituted for the sugars, and not at all in the potassium nitrate solution when glycerin was substituted for cane-sugar. These statements are perhaps open to criticism owing to the difficulty of obtaining cane-sugar and mineral salts entirely free from organic nitrogen. If in the presence of air this organism can use glycerin as a carbon food and if it can take nitrogen from potassium nitrate it is difficult to understand why it did not grow when glycerin was substituted for the cane-sugar. Brenner offers no explanation. The writer's interpretation is that the organism obtained its nitrogen not from the potassium nitrate, as Brenner supposed, but from some unsuspected, slight impurity in his cane-sugar and consequently when glycerin was substituted growth could not take place because there was no available nitrogen food. A further reason for this conclusion is that I have repeated Brenner's experiments, using chemicals sup- posed to be pure, but certainly not entirely free from extraneous sub- stances, and have obtained somewhat contradictory results. More- over, a very slight, but distinct growth was obtained in twice distilled water, to which only the mineral elements of Fischer's solution had been added, to wit: Dipotassium phosphate, magnesium sulphate, and calcium chloride. A decidedly better growth was obtained by adding cane-sugar, but this growth was not increased or only slightly increased by the further addition of potassium nitrate. The incom- plete culture medium (supposed to be free from nitrogen but probably not altogether free) gave as good results, or nearly as good (clouding and bacterial precipitate), as the one made complete by the addition of potassium nitrate. A third objection lies in the fact that the or- ganism does not reduce nitrates to nitrites or to nitrogen, so far as can be determined by the starch-iodin sulphuric acid test. This experiment was repeated again in 1909 using for the potas- sium nitrate Merck's guaranteed reagent, with the following results: Notes of July 30, 1909, on inoculations of June 29, which were by 2 mm. loops from young bouillon-cultures into 10 cc. portions in very clean test-tubes of resistant glass. (1) Fischer's mineral solution consisting of distilled water, dipotassium phosphate, magnesium sulphate and calcium chloride. Fluid cleared, slight yellow precipitate. Earlier, i. e., during the first 2 or 3 weeks, there was a feeble clouding showing presence of N. and C. in the medium. (2) Fischer's mineral solution plus ordinary white cane-sugar. Fluid nearly cleared. A very slight distinctly yellow precipi- tate— same as No. 1 . Earlier the fluid was feebly clouded. (3) Fischer's mineral solution plus c. p. cane-sugar. Fluid cleared. Slight yellow precipitate. Closely like 1 and 2. During the first weeks there was the same amount of very thin clouding. The change in sugar made no change in the behavior. (4) The same as No. 2 plus 1 per cent KNCh, (Merck's G. R.). Thinly clouded, some pseudozoogloea?. A small amount of yellow precipitate, tinctly more growths than in Nos. 1, 2, or 3. (5) The same as No. 3 plus 1 per cent KNO, (Merck's G. R.). Like No. 4 in clouding, etc. There is a scanty precipitate without much yellow in it. More growth than in Nos. 1 to 3. vJ i r J Fig. 129.' Dis- *Fig. 129. — Stab-culture of Bacterium campestre in nutrient gelatin o to phenolphthalein, after 12, 28 and 46 days at about 23° C. Only upper part has liquefied. Lower part has not liquefied even along needle-track where some growth has taken place. In lower tube there is a heavy yellow precipitate at bottom of liquefied clear gelatin. Inocu- lated June 5, 1897. 326 BACTERIA IN RELATION TO PLANT DISEASES. Results. — There has been two or three times as much growth in tubes to which the potassium nitrate was added, but it is not a good growth, i. c, such as would take place in bouillon with peptone . Tests made some days ago, i. e., after a distinct difference developed, showed no nitrite present in inoculated tubes containing the potassium nitrate, so the puzzle is where the bacterium obtains its necessary nitrogen. Perhaps under stress it is able to assimilate N. slowly from unsuitable material just as Bad. hyacinth i under similar conditions is able to take C. from potato-starch. On July 16, for comparison, inoculations were made in Fischer's mineral solution plus i per cent cane-sugar and i per cent Witte's peptone. These tubes although they have not been inoculated as long as the ones containing potassium nitrate have given twenty times as much growth (clouding and yellow precipitate) . Further tests were made in February, 191 1, as follows: To Fischer's nutrient mineral solution 1 per cent nitrate of potash was added. The strained solution was then divided into two equal portions. To one half was added 1 per cent cane-sugar and to the other i per cent glycerin. Inoculations were made from young bouillon cultures, and for checks on the amount of growth additional inoculations were made into our ordinary + 15 pepton- ized beef bouillon. Two strains of the organism were used, one isolated in my laboratory, the other received from Harding of Geneva, N. Y. The results after 26 days were as follows : (1) N. Y. strain in Fischer's nutrient mineral solution with cane-sugar and potassium nitrate: Fluid clear, moderate pale yellow precipitate and interrupted white pellicle shaking down easily into small fragments which make the fluid very flocculent. Five tubes: About J as much growth as in the check-tubes in peptone beef bouillon which have a yellow rim and a copious yellow pellicle. (2) Washington strain in the same: Less growth, scanty pale rim, no pellicle. Scanty yellow precipitate, no flocculence on shaking, but a moderate clouding. Five tubes: About i as much growth as in the checks in the peptone bouillon. (3) N. Y. strain in Fischer's solution with Schering's c. p. glycerin and potassium nitrate: A distinct growth but less pellicle and less precipitate than in 1. A thin rim with no distinct color. Fluid well clouded on shaking with numerous small flocculent masses. Five tubes : About one-tenth as much growth as in the peptone bouillon checks. (4) Washington strain in same: Like 3 but more thinly clouded. Yellow precipitate. Five tubes: About one-fifteenth as much growth as with checks. Each tube was tested with boiled starch water. 1 :200 potassium iodide water and a few drops of sulphuric acid water. No blue reaction took place. I can not think, therefore, that this organism is a nitrobacterium in Dr. Fischer's sense of the word, since it does not reduce nitrates to nitrites, so far as can be determined by the test with starch, potassium iodide, and sulphuric acid, and obtains its nitrogen much more readily from peptone than from KNO;i. It can use glycerine in presence of KNO3. Hecke isolated this bacterium in the winter and in spring from frozen kohlrabi. The organism probably winters over in the soil, but up to this time no one has actually plated it out of soils. In Smith and Swingle's experiments ten freezings and thawings in course of about 6 hours did not destroy all of the individuals in a bouillon culture, but the first two or three freezings destroyed most of them. In experiments made some years ago, the writer found this organism much more re- sistant to dry air than Harding's first report would indicate, to-wit, in Harding's experi- ments invariably destroyed in 45 hours, and 7 out of 8 cover-slips sterile at the end of 21 hours. In my own tests the organism on 8 out of 24 cover-slips was alive after 34 days when inoculated from a potato-culture 2 days old, and on 2 out of 23 cover-slips when inoculated from bouillon. On agar the writer found the organism very resistant, to-wit, alive after 175 months. Much seems to depend on the thinness of the layer and the nature of the surface on which it is dried. Recently Harding and his associates have shown that when placed on cabbage seed and set away in test lubes, sealed and unsealed, a certain number of the bacteria were still living at the end of a year. Harding tested resistance to disinfectants by adding one drop of a freshly clouded bouillon-culture to 10 cc. of the substance and making bouillon sub-cultures therefrom at BLACK ROT OK CRUCIFEROUS PLANTS. 327 the end of 1, 2, 5, io, and 15 minutes. Lysol in 0.5 per cent solution killed in 1 minute, but in 0.25 per cent solution it failed to kill in 15 minutes. Carbolic acid in 0.625 per cent solu- tion killed in 5 minutes, but not in 2 minutes. The organism is quite sensitive to the presence of sodium chloride, but not as much so as Bacterium phaseoli. It grew very feebly in salted gelatin (o gelatin converted into + 26 by adding c.p. HC1). Further comparisons should be made. The organism is able to live in mixed cultures for a considerable time (Russell, Smith). In agar cultures at room-temperatures Harding states that the organism remained alive from 4 to 6 months. In the cool-box on potato the writer has kept it alive for a year (average temperature about 120 C). Harding injected fresh bouillon cultures (2.5 cc. subcutaneously, 2 cc. intravenously, and 4 cc. subperitoneally) into rabbits with no ill effect other than a temporary loss in weight. RESUME OF SALIENT CHARACTERS. Positive. Pathogenic to Cruciferae, dissolves middle lamella, plugs vessels, forms numerous closed cavities in the host-plant; in cabbage causes conspicuous black stain in veins of leaf. vShort rods with rounded ends, single or in pairs, and occasionally in short chains of 4 or more; sometimes much resembling a coccus (Coccobacillus) when crowded in the plant or in old cultures; sometimes slightly curved or irregular in shape. Long chains or non- septate filaments frequently occur in sugar-rich media. Pseudozoogloeae ; yellow on all media, changing to dirty yellow-brown in the plant and on cruciferous substrata (culture media) ; motile in newly diseased tissues and in young cultures, 1 -flagellate ; very resistant to drying under certain conditions. Surface colonies on agar or gelatin rather slow- growing, circular, pale yellow at first, deepening with age, smooth, wet-shining, flat, with distinct margin; buried colonies small and slow-growing; feathery X- shaped crystals of ammonium magnesium phosphate formed in beef-agar after some days; white chemical halo on nutrient agar; gelatin and Loeffler's blood serum liquefied slowly with brownish stain; colonies on gelatin feebly zoned concentrically. Rate of liquefac- tion depends upon the medium, it may sometimes begin in 24 hours in peptonized beef-broth gelatin (feebly alka- line to litmus) and be completed in 15 days, often slow. Growth in stab-cultures is usually best near the surface. Neutral or alkaline media produce the best growth while acids ( + 40) inhibit or retard it. Copious growth on steamed potato cylinders, filling the fluid in the bottom of the tube with yellow slime and convert- ing it into a solid alkaline mass turning brownish with age ; nearly complete conversion of potato-starch. Organism blues litmus milk; it throws down casein slowly, i.e., by a lab ferment; gradual digestion of casein (Smith, Harding); inverts cane-sugar (?); vitality lessened by long cultivation (Harding) ; slow production of indol in sugar-free peptonized beef-bouillon *Fig. 130. — Two tubes of bouillon inoculated with Bacterium campestre: Left, over chloroform (clear); right, check (clouded). Each tube was inoculated on Feb. 12 with a 2 mm. loop of clouded broth. Photographed Feb. 17. Fig. 130.' 328 BACTERIA IN RELATION TO PLANT DISEASES. or peptonized Usehinsky's solution; strictly aerobic so far as is known; occasional forma- tion of hydrogen sulphide; partial reduction of litmus in cabbage-broth cultures; slight production of acid on some vegetable media and with various sugars; scanty growth in Fermi's solution, Usehinsky's solution and sometimes in Cohn's solution (see Negative) ; feeble growth in vacuo; killed in agar plates by direct sunlight (30 minutes or less) ; pro- duces a brown pigment soluble in water and a yellow pigment, a lipochrome, insoluble in water but extracted by alcohols, acetone, etc. ; minimum temperature for growth is about 50 C, optimum temperature about 300 C, maximum temperature about 380 C to 39°C. Ther- mal death-point is about 510 C. Tolerates sodium hydrate in peptonized beef -bouillon to — 40, and plant acids to +30 or +40 (?). Young cultures stain readily with various basic anilin dyes; sections of tissues stain satisfactorily with Ziehl's carbol-fuchsin, with nigro- sin and with Heidenhain's or Benda's iron haematoxylin; a nitro-bacterium according to Fischer's classification (Brenner) see Negative. Most of the organisms in a test-tube culture were destroyed by two or three freezings and thawings, but a few individuals survived ten. Killed in one minute by a 0.5 per cent solution of lysol but was not killed in 15 minutes by a 0.25 per cent solution (Harding). Carbolic acid in 0.625 Per cent solution killed in 5 minutes, but not in 2 minutes (Harding). Less sensitive to the presence of sodium chloride than Bad. phascoli. Remained alive 4 to 6 months in agar cultures at room-temperatures (Harding). Lives on culture media for a year in the cool-box (Smith) and on cabbage seed for a year (Harding etal.). Able to live in mixed cultures for a considerable time. Group number 211. 3332513 (Smith, Harding). Negative. No distinct capsule; no spores; no true pellicle formed on neutral peptonized beef- bouillon (Harding) ; no acid coagulation of milk; occasionally no liquefaction of unfavor- able gelatins; no action on lignified tissues; no solvent action on Swedish filter-paper; no reduction of nitrates to nitrites; not a nitrobacterium (Smith); no characteristic odor; no production of gas or growth in the closed end of fermentation tubes in peptone-water or peptonized beef-bouillon with any of the following carbon compounds : grape-sugar, fruit-sugar, cane-sugar, galactose, milk-sugar, maltose, dextrin, mannit, glycerin; nor in potato-broth, cabbage-broth or cauliflower-broth; no growth in hydrogen, nitrogen or carbon dioxide; often no growth in Cohn's solution; no growth in peptonized beef-broth over chloroform ; brown pigment not formed in beef-broth nor in peptone water with grape- sugar. Not pathogenic to rabbits (Harding). TREATMENT. The treatment of this disease falls principally under the head of restriction and pre- vention. Seasonal variations undoubtedly play an important part in the development of the disease on lands already infected. Cool, moist lands may be expected to be more sub- ject to it than warm dry ones. Even in the same field the writer has observed the varying quality of the soil to exert a marked influence on the number of water-pore infections, the plants on the dry end of the field being nearly free. In warm autumns accompanied by frequent rains the infections are much more numerous and the disease certainly progresses much more rapidly than in cool, dry seasons. Carman believed the disease to be funda- mentally associated with hot, wet weather. Pammel (1893) says that dry weather in Sep- tember checked the progress of the disease. Russell notes (1898) that the disease varies much in intensity on the same field in different years according to varying weather con- ditions. An intelligent cabbage-grower of Racine, Wisconsin, thoroughly familiar with the black rot, recently told the writer that he lost by this disease the entire crop from a field of six acres in the rainy season of 1900, not having enough cabbages even for the use of his BLACK ROT OF CRUCIFEROUS PLANTS. 329 family, whereas from the same field in the dry season of 1901 he harvested a reasonably good crop.* Under drainage might, therefore, be advantageous in wet seasons. When the cabbage-plant is well along toward maturity before water-pore infection takes place, to wit, in late summer or early autumn, the disease may sometimes be pre- vented from entering the head by the removal of infected leaves or portions of leavesf (experiments by the writer, and by Russell), but of course it will not answer to rob the plant of all or most of its leaves and expect a crop, nor is it likely that the experiment would prove very successful when applied to a small part only of a very badly diseased field (Geneva Sta. Bull. 232). Insects which distribute this disease may, of course, be reduced in number by insec- ticidal treatments. Care should be taken that diseased plants do not find their way into the manure-heap, barn-yard, or dump-pile and thence back on to the land. .Several cases have come to the writer's attention where wholesale infection of fields or parts of fields could be accounted for only in this way. Potter also mentions one. Heavy manuring appears to favor greatly the development of this disease, but on the other hand considerable manure is required for the satisfactory growth of the plants. The most that can be done is to see that the manure itself does not become infected. Diseased plants should be piled with dry brush and burned, or effectually disposed of in some other way. Losses in winter in storehouses can be wholly or partially avoided by keeping the plants in all parts of the house at a uniform low temperature, i. c, slightly above freezing. Of course there should be a careful inspection of such heads or roots as they are stored, and those showing decayed spots or blackened bundles should be rejected. Most of the winter decay has been observed to take place in the warmest parts of the houses. Such houses should be well ventilated and uniformly cool. By far the most common method of infection of healthy fields (so far as we yet know) is through the seed-bed. This should be made with the greatest care, preferably on land not previously used for cruciferous plants, and certainly on land which has never been sub- ject to this disease, and with seeds which are not contaminated by the presence of this organism, otherwise a very considerable proportion of the seedlings may carry the disease with them from the seed-bed into the field, which thus becomes permanently infected. The writer observed one case in Wisconsin where about 20 per cent of the seedlings were diseased. The field set from this seed-bed became so badly diseased that it was abandoned by the planter in midsummer. As early as the first week in September, 58 per cent of the plants set from this seed-bed, (which was on land where the disease prevailed the year before) were diseased, many of them badly. This was a field which had not been previously planted to cabbage and consequently one which might otherwise have been expected to yield a healthy crop. In 1 897 vStewart suggested that the disease is disseminated by seedsmen. This inference was based on the fact that he found the disease in plants which had been reserved for seed and were fruiting, although not, I believe, on the seeds themselves. It is not unlikely that the disease may have been introduced into this country on seed from the old world or that it is now being spread from place to place by cabbage-seed, etc., derived from plants grown on land subject to the disease. This, however, is conjecture and must remain so until some one has demonstrated the actual presence of the organism on cabbage-seed, etc., in spring or at least has obtained diseased plants in healthy soil from the use of such suspected seed. The writer has also seen the disease in cabbage-plants set out for seed and coming into fruit. It is also a well-established fact that much of the cabbage-seed now grown in the United States comes from regions much subject to this disease. The final proof must come 'The crop on such a field well cared for should be worth $800. tWhenever possible, portions only should be removed. 330 BACTERIA IN RELATION TO PLANT DISEASES. from a study of cabbage and other cruciferous seeds in a natural state, as they come from the seedsman, and most easily before they have left his hands, i. e., in the field on maturing plants observed to be diseased. Since the above paragraph was written Harding and his associates have thrown a flood of light on this subject by actually obtaining Bacterium campestre from the surface of cabbage- seed harvested separately from four diseased plants obtained from a field of seed-cabbage on Long Island. This shows that the organism may reach and contaminate the seeds in a certain number of plants reserved for seed, and renders it likely that the entire crop of a given seedsman would be more or less contaminated if some diseased plants came to maturity in his fields and were harvested along with the sound plants, since the dust of the threshing would disseminate the organisms widely. Can the organisms thus disseminated remain alive for a long time on such seed i. c, from autumn until spring? We do not yet know, but the probabilities are in favor of such a belief, since these experimenters have also shown that when pure cultures of this bacterium are placed on cabbage-seed and dried, some of the bacteria (a small proportion it would seem) remain alive for at least 13 months. These discoveries, while not wholly conclusive, render it extremely probable that the disease is very often disseminated on seed, since the persistent vitality which has been demonstrated experimentally in the laboratory using pure cultures is quite likely to occur naturally, at least sometimes, on contaminated seeds offered for sale in the open market. If the organ- isms remain alive on such seeds over winter in any great number we have a very satis- factory explanation of the wide dissemination of this disease in the United States during the last 15 years. Harding has now made out so good a case against the seedsmen that these gentlemen should take very special precautions to avoid harvesting seed from infected plants, and from badly infested fields. Such fields should not be used for production of seed-cabbage. The growers, on the other hand, as a matter of ordinary precaution, should disinfect all cruciferous seeds before planting them. This may be done, it is said, by soaking the seeds for 15 minutes in 1 :iooo mercuric chloride water, or in full strength formalin diluted with water in the proportion of 1 : 240. It seems likely that this short treatment will prove effective only in case the organisms are confined to the surface of the seeds. If they are sometimes in the interior a longer treat- ment will probably be necessary, and to avoid this (since it might prove disastrous to germination) it would be advisable to screen out and reject all shriveled and inferior seeds before planting, since it is probable that the plump seeds will be less likely to be infected within than the thin or shriveled ones. In Farmer's Bulletin No. 68 (which may be had on application to the U. S. Department of Agriculture), the curious reader will find various pieces of evidence tending to show spread of the disease from the seed-bed, persistence of the organism in the soil, and infection of land by way of the dung-pile, and by refuse from cabbage storehouses and pits. To summarize: Avoid injected seed, soil, and manures; destroy insect carriers of infec- tion; if the plants are attacked, harvest early, and use at once, or store in a very cool house. PECUNIARY LOSSES. It is difficult of course to arrive at any very definite conclusions respecting the losses due to this disease. In September 1904, the writer received the following statement con- cerning black rot from an extensive cabbage-grower near Chicago : "I send you by mail some small samples which I hope will be sufficient in order for you to determine the trouble. In some cases as high as 40 or 50 per cent of the cabbage is diseased and in one ease which I have noted I should think that over 90 per cent of the cabbage had died from this disease. The ground from which these plants came is comparatively new ground and has only been under cultivation for 2 or 3 years, though large quantities of stable manure have been placed upon it." BLACK ROT OF CRUCIFERIOUS PLANTS. 331 The black rot in cauliflower caused a loss of $400 per acre to a planter at Beeville, Texas, in the winter of 1902, according to his statement to me. Diseased plants were received from him, the organism was plated out on agar, subcultures were made on potato, and then it was re-inoculated into cauliflowers, which became diseased with the typical black rot. Similar statements respecting losses have been received from many other places (see p. 328). Not infrequently in various parts of the country entire fields have been destroyed so that not a single plant was harvested. Russell has published a very interesting figure of such a field. At the present time the disease is destructive to a greater or less extent, according to the season, on Long Island, in Western New York, and in parts of Ohio, Illinois, Michigan, Wisconsin, and Texas. The disease has come into general notice in this country only recently, but the losses from it in the vicinity of Racine, Wisconsin, were estimated for me by good observers 14 years ago at not far from $150,000. The disease is also said to be so prevalent in parts of New York that extinction of the industry is threat- ened. The writer estimates the losses on Long Island during the last 12 years at upward of $250,000. Probably the total loss from this disease in the United States since its first appearance has been not less than $800,000. Hecke reports this disease to be common in Austria on a variety of crucifers. The loss on a kohlrabi field in Southern Austria, which furnished the first material for his studies, was very great, in fact almost complete, since it was impossible to use the black veined tissues for preserving. Van Hall's correspondent in North Holland says that the disease has existed for several years, first attacking Utrecht red cabbage but at present attacking all varieties and even cauliflower. The damage is worst on the red cabbage. The disease is so dreaded that many fear the culture of cabbage will have to be given up. Many hec- tares of cabbage are diseased this year (1900) so that from seven-eighths to nine-tenths of the harvest will be lost. In the United States cabbages are often stored in large quantities for the spring market, and the black rot frequently continues in such plants, particularly in the warmer parts of the houses usually associated with soft rots. According to Russell a quarter of the stored crop of cabbage was lost at Racine, Wis- consin, in the winter of 1 896-1 897 by the development of this disease in the stock. When taken out such cabbage heads are often sound externally although badly rotted within. HISTORY. Garman (1891-1892) appears to have been the first to comment on this disease. He saw what we may infer to have been this disease in July, 1889, at Lexington, Kentucky, but did not make out its etiology. From the diseased cabbages two organisms were isolated, anon-motile white and a motile yellow. In most of the cabbages the yellow form was the commoner one. "The size and behavior of this [yellow] species leads me to think it a form of Bacterium termo." The two species were not described. A quick decay [soft rot] was obtained by transfer of some of the diseased material to the interior of healthy cabbages, but pure cul- ture inoculations were not successful. Further opportunity for studying the disease did not present itself and the writer leaves the subject with the following remark : "Whether the disease is induced primarily by the attacks of bacteria, or by hot, damp weather, the work thus far done does not show satisfactorily. From the facts in my possession it appears to me probable that neither alone will cause the disease ; that it is only during periods of high tempera- ture and excessive rainfall that the organisms are able to invade and break down the tissues of plants." Pammel (1 893-1 895) was the first to demonstrate the infectious nature of the disease. He isolated a yellow bacterium from rutabagas and determined it to be the cause of the disease by means of pure culture inoculations. .Vv BACTERIA IN RELATION TO PLANT DISEASES. In 1896-1897, Smith verified Pammel's statements, extended his inoculations to cab- bages and other plants, carefully illustrated the disease in color, and obtained additional information respecting cultural characters of the organism. He also discovered that infec- tions take place through the water-pores. This knowledge, published in June, July, August and September 1897, was also summarized in a Farmer's Bulletin issued Januarv 8, 1898, and widely distributed among cabbage-growers. Subsequently (February and March 1898) Russell, and Russell & Harding, went over the whole ground in papers of considerable length, confirming in the main the statements of Pammel and of Smith, and making various addi- tional observations. The same year, after many additional experiments, Smith again published in "Zeitschrift fur Pflanzenkrankheiten. " In 1899-1900, Harding demonstrated the disease to be common in Europe. European investigators (van Hall in 1900 and Hecke in 1901-1902) were then moved to study this disease. Hecke in particular did a very thorough piece of work on kohlrabi, adding con- siderably to our knowledge and at the same time confirming many statements made by Smith, Russell, Harding and others. In 1 90 1, in his dispute with Fischer, Smith published a series of photomicrographs illustrating this disease. In 1903 he published another series of photomicrographs showing the effect of the black rot on turnips. In 1903, Potter reported on the occurrence of this disease in England, especially in Swedes. In 1 902- 1 904, Brenner, a special student of Alfred Fischer at the University of Basel, went over the ground once more at Dr. Fischer's suggestion. He experimented principally with cabbages, although he mentions having observed the disease in numerous other cruci- fers. He added some new facts and also confirmed many statements previously made by Smith and disputed by Fischer. In 1903-1905, Stewart & Harding, and Harding, Stewart & Prucha continued their studies, the most important new facts brought out being that the organism does actually occur in autumn on a portion of the cabbage seed grown for market in infected districts, and that when cabbage-seeds are moistened with a culture of Bacterium campestre it is able to live on their surface for more than a year, thus ren- dering it extremely probable that the disease maybe disseminated by seedsmen. The manner of threshing cabbage seed, as they point out, is such that the dust from a few infected plants would be likely to contaminate the seeds of many sound plants, if not that of the whole crop. In 1 904-1 905, Smith and Swingle showed that the organism could be destroyed in great part by repeated freezings with liquid air and with salt and ice. The few bacteria which survive the freezings are still infectious. In 1910, Harding summarized his studies on about 45 isolations of this organism in accordance with the requirements of the Descriptive Chart of the Society of American Bacteriologists. **m gfJfP *Fig. 130a. *FiG. 130a — Left: Margin of cabbage U'uf showing extrusion of fluid from the water pons stage of bacteria] infection on cabbage leaf by way of the water-pores. Right: Early BLACK ROT OF CRUCIFEROUS PLANTS. 333 LITERATURE. 1891. Garman, H. A bacterial disease of cabbages. Bot. Gazette, Sept., 1891, vol. xvi,No.9,p.265. Brief abstract of a paper read before the Am. Asso. Agric. Colleges and Exp. Stations; Washington meeting. 1892. Garman, H. A bacterial disease of cabbage. Agric. Sci., July, 1892, vol. vi, No. 7, pp. 309-312. This paper was republished in 1894 under the same title and with only slight changes in the Third Annual Report of the Kentucky Agric. Exp. Sta. of the State College of Kentucky for the year 1890. pp. 43-46, Frankfort, Ky., 1894. 1893. Pammel, L. H. Preliminary notes on a ruta- baga and turnip rot. Bot. Gazette, Jan., 1893, p. 27. Abstract of a paper read before the Am. Asso. of Agric. Colleges and Exp. Stations; New Orleans meeting. This is the bacterial disease subsequently described more fully by Professor Pammel in 1895. 1895. PammEL, L. H. Bacteriosis of rutabaga (Bacillus campestris n. sp.) Iowa Agr. College, Exp. Sta. Bull. No. 27, Ames, Iowa, 1895, pp. 130-134, 1 pi. This paper was reprinted in Am. Monthly Microscopical Journal for May, 1895 ,p. 145 . 1896. Russell, H. L. A leaf-rot of cabbage. Proc. Am. Asso. Adv. Sci., 1895, vol. xliv, p. 193 (Springfield Meeting), Salem, May, 1896. Said subsequently to have been based on a study of the black- rot. but this fact can not be determined from the abstract which is all that was ever published Infections with pure cultures had not been obtained. 1897. Smith, Erwin F. A bacterial disease of Crucif- erous plants. Science, N. S., June 18, 1897, vol. v, p. 963, Abstract of a paper read before the Biological Society of Washington in May. 1897. 1897. Smith, Erwin F. Pseudomonas campestris (Pammel), the cause of a brown-rot in crucif- erous plants. Centralbl. f. Bakt. etc., 1897, 2te. Abt., Bd. Ill, No. 11-12, July 7, pp. 284- 291; No. 15-16, Aug. 18, pp. 408-415; No. 17-18, Sept. 10, pp. 478-486, 1 colored plate (showing signs and the character of the bac- terial growth on potato). This paper gives at length the reasons for the statements previously made in Science, and adds some new facts, the most important of which perhaps is that infections of the uninjured plant can take place by way of the water-pores. 1897. Smith, Erwin F. The spread of plant diseases: A consideration of some of the ways in which parasitic organisms are disseminated. A lecture delivered before the Mass. Hort. Soc, March 27, 1897. Proc. of the Society for 1897. Boston, 1898. Also a separate. An abstract appeared in one of the Boston papers soon after the lecture, and there was also a separate of this abstract. 1897. Stewart, F. C. The stem-rot of cabbage. Vieks Illustrated Monthly Magazine, July, 1897, vol. xx, No. 9, new scries, p. 141. An editorial which includes, however, a letter from Mr. Stewart who says: "On Long Island there is a bacterial stem- rot of seed cabbage which is very destructive in some seasons." The distribution of the disease is attributed to infected seeds. 1898. Smith, Erwin F. Additional notes on the bac- terial brown-rot of cabbages. Bot. Gazette, Feb. 1898, vol. xxv, p. 107 and Am. Nat. 1898, P- 99- Abstract by the author of a paper presented at the meeting of the Society for Plant Morphology and Physiology. Dec. 28, 1897. 1898. Smith, Erwin F. The black-rot of the cabbage Farmers' Bull. No. 68, U. S. Dept. of Agric, Div. of Veg. Phys. and Path., 8 vo., 21 pp. Issued Jan. 8, 1898. 1898. Smith, Erwin F. Some bacterial diseases of truck crops. Trans, of the Peninsula Horti- cultural Society, nth Annual Session held in Snow Hill, Md., Jan. n-12, 1898, p. 142-147, Dover, Del., 1898. Also a separate. Three diseases are discussed: Wilt of the Cucumber; Brown rot of the Potato; and Black-rot of the Cabbage. 1898. Anonymous. Brown-rot of cruciferous plants. Bot. Gazette, vol. xxv, Jan., 1898. p. 67. A review and criticism. (See next number.) 1898. Smith, Erwin F. A Reply [to Criticisms of The Bot. Gazette in reference to brown-rot of crucifersj. Bot. Gazette, 1898, vol. xxv, No. 3, pp. 204-207. -Mostly polemical but one additional fact is announced i that the ability of Pseudomonas campestris to liquefy gelatin depends on how the gelatin is made, and thus the apparent contradiction, in this particular, between Pammel's results and those of the writer is explained. 1898. Barnes, C. R. Bacterial rot of cabbage and allied plants. Bot. Gazette, March, 1898, vol. xxv, p. 211. A review and criticism. 1898. Russell, H. L. A bacterial rot of cabbage and allied plants. Univ. of Wis. Agric. Exp. Sta. Bull. No. 65, Feb. 1898, 8 vo., 39 pp. with 15 figures. Distributed in March, 1898. The cultural characters of the organism were contributed by Mr II A Harding. 1898. Russell, H. I.. A bacterial disease of cabbage and allied plants. Proc. nth Annual Con- vention of the Assoc. Amer. Agr'l Colleges and Exp. Stations held at Minneapolis, July '3-'5, 1897, PP- 86-89, Washington [March] 1898. This paper was not read at the Convention (see p. S6) and the MS. remained in the hands of the author for revision until Oct. 27, 1897. The Proceedings of which this forms a part, bears no date of issue but it was received from the binders and distributed by the U. S. Dept. of Agric. March 28. 1S9S A brief synopsis of this paper appeared in the Exp. Sta. Record, 1S97-98. vol. IX, p. 319. 1898. Smith, Erwin F. Pseudomonas campestris (Pammel) Erw. Smith: Die Ursachen der "Braun" oder "Schwarz" Trocken-Faule des Kohls. Zeitschrift fiir Pflanzenkrankheiten, 1898, Bd. vin, p. 134-137, 1 pi. (showing signs). Also a separate. This paper was sent to the printer the first of March. 1898, i. c. one year after the sending away of the Centralblatt paper of 1897 and after all of the leading statements in the latter paper had been experimentally re-examined by the writer and confirmed. The halftone from a photograph of part of a leaf (enlarged 25 times and made by transmitted light) probably gives as good an illustration of the foliar symptoms as can be obtained in black and white by use of photography. 1898. Smith, Erwin F. Description of Bacillus pha- seoli n. sp. with some remarks on related species. Proc. Am. Asso. Adv. Sci., Salem, 1898, vol. xlvi, p. 288. Read in Detroit, Aug. 1897. 1898. Jones, L. R. Club foot and black-rot. Two diseases of the cabbage and turnip. Bull. 66, Vermont Agr'l Exp. Station, Sept., 1898, Burlington, Yt. The part relating to black-rot is on pp. 13-16. A popular account drawn from papers hy Russell and Smith but including a very few personal observations. 3.34 BACTERIA IN RELATION TO PLANT DISEASES. 1899. Frank, A. B. u. Sorauer, Paul. Jahresbericht des Sonderausschusses fur Pflanzenschutz, 1898. Arbeiten der deutschen Landwirt- schaftsgesellschaft, Heft 38, Berlin, 1899. V. Oel — und Gemiisepflanzen. iS Bakteriose, p. 105. Dr. Frank reports from Berlin in cabbage. " the same bacterial disease which in America does great injury." Mere mention. 1899. Smith, Erwin F. Gelatin culture media. Am. Nat., 1899, p. 214. Brief text and 1 plate showing behavior of Ps campestris in various nutrient gelatins. Abstract of a paper read before Soc for Plant Morphology and Physiology, Dec. 28, 1898. 1899. Harding, H. A. On the occurrence of the black-rot of cabbage in Kurope. Proc. Am. Asso. Adv. Sci., Aug., 1899, vol. xlviii (Columbus meeting). Published Dec, 1899, p. 294. A brief abstract. The paper was subsequently translated into Cerman and published in full (see next number). 1900. Harding, H. A. Die schwarze Fiiulniss des Kohls und verwandter Pflanzen, eine in Europa weit verbreitete bakterielle Pflanz- enkrankheit. Centralbl. f. Bakt. etc., Mai 18, 1900, 2te Abt., Bd. vi, No. 10, pp. 305-313. 2 pi., 1 fig. in text, and a map showing 1 1 places in Europe where the disease had been located by the author. Also a separate. 1900. VAN Hall, C. J. J. Twee baeterienziekten. Teijdschrift over Plantenziekten, Jaarg. VI, Afd. 5-6, 1900, pp. 169-178, 1 fig., 1 pi. Reports finding the disease caused by Pseudomonas cam- pestris in cabages sent from North Holland to the laboratory at Amsterdam. 1 901. Smith, Erwin F. Entgegnung auf Alfred Fisher's "Antwort" etc., Centralbl. f. Bakt. etc., 2te Abt. Bd. vn, No. 5-6. Also a separate. Tafeln vin and IX and accompanying text (pp. 195-197) relate to Bacterium campestre. The plates are heliotypes from photomicrographs by the writer. 1 901. Hecke, Ludwig. Eine Bacteriosis des Kohl- rabi. Zeits. f. des Landw. Versuchswesen in Oesterreich, iv Jahrg., Heft. 4, pp. 469 to 476, 1 heliotype plate. Wien, 1901. Also a sepa- rate, pp. 8. 1902. Bos, J. RlTZEMA. De Bacterieziekte in de Kool. Phytopathologisch Laboratorium Willie Commelin Scholten. Verslag over etc., in het jaar 1901. Amsterdam, 1902, pp. 13 and 14. 1902. Hecke, Ludwig. Die Bacteriosis des Kohlrabi. Zeits. fiir landwirthschaftliche Versuchswesen in Oesterreich, v Jahrg., Heft 1, pp. 1 to 21. Wien, 1902, with 1 Crayondruck plate. Also a separate, 21 pp. 1902. Keuter reports occurrence of Ps. campestris on cabbage in Denmark in 1900. Zeits. f Pflanz- enkr., 1902, Bd. xn, p. 293. [903. Stewart, F. C. and Harding, H. A. Combating the black-rot of cabbage by the removal of affected leaves. Bull. No. 232, New York Agric. Exp. Sta., Geneva, N. Y., April, 1903, pp. 43 to 65. Printed also later as part of an annual report. 1903. Potter, M. C. On the brown-rot of the Swedish turnip. With a note on the same disease of the cabbage. The Journal of the Board of Agriculture, No. 3, London, Dec, 1903, vol. x, pp. 314-318, 1 pi. in color. 1904. Harding, H. A. and Stewart, F. C, Vitality of Pseudomonas campestris (Pam.) Smith, on Cabbage seed. Science. July 8, 1904. New- Series, Vol. xx, No. 497, pp. 55-56. This organism was obtained from cabbage seed taken from plants diseased by black-rot. On cabbage seed soaked in water to which Ps. campestris was added and then dried and placed in test-tubes some of the bacteria were alive at the end of 10 months. See bull. 251 1934. Harding, H. A., Stewart, F. C. and Prucha, M. J. Vitality of the cabbage black-rot germ on cabbage seed. Bull. 251, New York. Agric Exp. Sta., Geneva, N. Y., Oct. 1904. 1905. Smith, Erwin F., and Swingle, Deane B. The Effect of Freezing on Bacteria. Science, N. S., vol. xxi, No. 535, March 31, 1905, pp. 481-482. 1907. Edwards, S. F. Pseudomonas campestris. Thirty-second Annual Report of the Ontario Agric. Col. and Exp. Farm, 1906. Toronto, 1907, p. 136. "It was observed that some kinds were more severely in- jured than others. For example the Jersey Kale was more diseased than any other kale." 1907. Kirk, T. W. Black-rot of cabbage. Ann. Report, New Zealand Department of Agri- culture, 1907, vol. xv, p. 157. Kirk reports having seen black-rot of cabbage due to Bac- terium campestre in two different years in New Zealand. 1908. Edwards, S. F. Cabbage resistant to black-rot. In 33d Annual Report, Ontario Agricultural College and Experimental Farm for the year 1907. Toronto, 1908, p. 134. 1908. JanczEWSKi. A. Pseudomonas campestris (Bacteriosis of cabbage). Jahresbericht for 1907 (Russian), St. Petersburg, 1908, p. 71. [908. FawcETT, H. S. Cabbage Disease. Black Rot (Pseudomonas campestris (Pammel) Erw. Smith). Florida Agric. Exp. Sta., Ann. Re- port for 1908, pp. lxxv-lxxx. Also Press Bulletin, 101. Reports "serious loss to cabbage, cauliflower, and ruta-baga crops in the State for several years." At Sutherland iu February, 1908. "an examination of the fields showed that the black rot was prevalent throughout the section, destroying from 25 to 75 per cent of the crops. Cultures made from diseased plants revealed the presence of yellow bacteria (Pseudomonas campestris) in specimens of the three plants named." 1909. SackETT. Walter G. Black Rot of Cabbage. Bulletin 138, Colorado Agric. Exp. Sta., Jan. 1909, pp. 15-18. 1910. Harding, H A. The Constancy of Certain Physiological Characters in the Classification of Bacteria. New York, Agric Exp. Sta., Tech. Bui. No. 13, June, 1 910, pp. 29-34. Describes cultures of Ps. campestris made as test of classifica- tion card, Soc. American Bact. ^ / H / YELLOW DISEASE OF HYACINTHS. x. Painted after 23 days. nl 49. inoculated at x. Painted plant 28, inoculated at x. Painted alter 36 days. (4) Hyacinth scape, plant 20, inoculated at ninth bulb, Haarlem, August, 1906. (6) Longitudinal section through Plateau badly diseased, cavities present and lower surface ruptured. (7) 1 Ac ' ise of I cross-section, showing bacterial ooz< d bundles. appearance of individual scales in this stage of di ate 20. All from Hau' , Henrietta Schilthuis. YELLOW DISEASE OF HYACINTHS. (Synonyms: Wakker's Bacterial Disease; Yellow .Slime — Dutch Geele Snot.) DEFINITION. This is a specific communicable disease of the common hyacinth, the most characteristic sign of which is the appearance in the bulbs of a bright yellow bacterial slime (pi. 19). On cross-sections of the bulb the disease appears in the earlier stages as small yellow dots, and on the longitudinal section as long, narrow, yellow stripes, corresponding to the loca- tion of the vascular bundles, which are the first parts to be conspicuously infected. In later stages the parenchyma is involved, the bulbs are badly decayed, and there are then various secondary infections (see plate 20, figs. 7, 11). Other signs are dwarfing and one-sided growth of the foliage, and water-soaked or brown stripes on the leaves. HOST-PLANTS. This disease is known to occur only in the common Dutch hyacinth (Hyacinthus oricntalis). It has been successfully inoculated into this species. The leaves of the inoculated plants showed the characteristic stripes and at the end of some months unmis- takable signs also developed in the interior of many of the bulbs. It has also been inoculated by the writer into the leaves of Hyacinthus albulus, Allium cepa, and Amaryllis atamasco, but only with slight local results ; in no case did the bulbs of these plants become affected. No results were obtained from inoculation into the leaves of cabbage. GEOGRAPHICAL DISTRIBUTION. So far as known the disease occurs only in the Netherlands, where the hyacinth is grown in vast gardens for export to all parts of the world. SIGNS OF THE DISEASE. The first sign in an inoculated leaf consists of stripes having a water-soaked appearance (plate 19, figs. 1 to 3). These are soon followed by the yellowing, browning and death of the tissue first attacked and by the appearance of water-soaked spots or stripes farther down, which in turn die and dry out. vSometimes the killed tissue becomes more or less trans- parent except the veins, which are feebly browned. In natural infections, these stripes usually begin toward the apex of the leaf. They extend downward rather slowly, but much more rapidly in this direction than sidewise. The result is that often the leaf will come to have a central dead stripe extending nearly or quite its whole length, while the margins of the leaf are still green and healthy in appearance. Sometimes the bulbs are infected from the flower-stalk. The signs in the scape lower down may or may not be external. When externally visible there is a water-soaked appearance (plate 19, fig. 4 and plate 20, fig. 5), followed by browning and shriveling. The signs in the bulb are so striking as to be unmistakable. In early stages of bulb-infection the disease is confined quite strictly to the vascular bundles, from one to fifty or more of these being yellow and full of bacterial slime, in a white and otherwise healthy tissue (fig. 131, and plate 1 9, fig. 8) . When the infection occurs through leaves, the scales which bear these particular leaves are the first part of the bulb to show the yellow slime, and naturally this appears first at the top of the scale in the vascular bundles. A little later the bacterial slime from these particular scales grows down into the solid base of the bulb (the plateau) where many of the numerous 335 336 BACTERIA IN RELATION TO PLANT DISEASES. anastomosing vascular bundles become yellowed, and where we often find a considerable area of the intervening parenchyma yellow and soggy (plate 19, figs. 6, 7). Subsequently the yellow slime extends upward into the bundles of other scales and sidewise slowly into the parenchyma, until finally the bulb is destroyed. In the later stages of the disease small pockets occur in the bulb-scales (plate 20, figs. 8, 9) and other bacteria frequently enter and help to complete the destruction of the bulbs but their presence is not essential. In this stage, mites {Rhizogly pints hyaciniki) may also be present (plate 20, fig. 7). Certain fungi are also met with in later stages, and notably a species of Penicil- liitiu. This is extremely common ( plate 20, fig. n). Wakker states that there may be also an up- striping of the green leaves due to their infection from the bulb. The above signs progress very slowly, several months to a year being necessary, as a rule, for the complete destruction of the bulbs. Not infrequently the disease ex- tends from the mother-bulb, by way of the plateau, into daughter- bulbs (Wakker, Smith). In such cases the daughter-bulls always shows the disease first in the basal portion (plateau), and of course, on the side next to the mother- _. ,,, „ bulb. Bulbs are frequently at- r ig. 131."" . , tacked on one side more than on the other (plate 19, fig. 9), and this may result in a curved growth of the foliage which bends over toward the diseased side. ETIOLOGY. The cause of this disease is Bacterium hyacinthi Wakker; a bright-yellow, medium- sized rod with rounded ends, motile by means of one polar flagellum, and multiplying by fission. It is this organism which causes the yellow color in the bundles of the diseased bulbs. The yellow slime in the bulbs is made up entirely of a homogeneous-looking bac- terial growth which in early stages ordinarily yields pure cultures of this organism when cultivated out with any degree of care, but which is sometimes mixed with other organisms, especially in advanced stages of the disease. Wakker, who first studied this disease criti- cally, obtained at different times a number of good cases as the result of wounds made in the leaves, scapes and bulbs, but inasmuch as most of his successful inoculations were what the writer has designated direct infections, i.e., the inoculation of raw material, there has been a tendency on the part of various writers to discount his results, and to confuse the general reader by speculations not based on any experimental data. Dr. Wakker's statements are, however, in the main, trustworthy, since the writer has obtained numerous successful con- firmatory inoculations from pure cultures of this yellow organism (for figures consult Bulletin 26). "Fig. 131. Earlystage in the destruction of a hyacinth bulb by Bad. hyacinthi. Cross-section of bulbcnlarged to show diseased vasculai bundles in ,\ scales. These i< bundles "1 re bright yellow. A seventh bundle, which does mil show plainly in the cut. was also diseased. Several of the dark spots are negligible, being shadows due to slight openings between the scales. Plant inoculated Feb. 16, 1897, on upper part of scape. Photographed June 23, 1897. Circa x 3. 1 LOW DISEASE OF HYACINTHS. (1) Cross-section of bulb, showing in center. Haarlem, 1906. (2) Check-lube of litmus milk. (3) /■ 10 days in li 1) 2 days in litmus milk at J (5) Plant 79, variety blue Baron van Tuyll, ii -mugh the flowers. Painted after 17 days. (6) Cross-section of infected bulb-scale. The cells of the parenchyma contain starch grains, but only a few are indicated. The xylem and adjacent parenchyma are disorganized ; the phloem is uninjured. After Wakker. (7) Cross-section of a bulb, showing secondary infection. The center ided by a soft while rol and by mites. Haarlem, August, 1906. (8) Outer face of a badly diseased scale. (9) Inner (ace of badly diseased bulb-scale. (10) Inner face of badly diseased bulb-scale, (II) Vertical section of a bulb, showing a secondary infection; the base is occupied by fruiting hyphae of a Penicillium. All from Haarlem, 1906. YELLOW DISEASE OF HYACINTHS. 337 The period of incubation in the writer's experiments varied from 3 to 30 days depending on the amount of infectious material employed and on the susceptibility of the variety. All of his inoculations were through the leaves and floral organs, and always at a consider- able distance from the bulb. In all cases pure cultures were used for the inoculations which were made in various ways, viz., by needle-punctures, by hypodermic injection, by placing drops of infectious fluid in the flowers, and by submerging the tips of leaves in fluid con- taining the living organism. The last method led to no very conclusive results but since the writer's experiments were not numerous and yet gave some indications of ultimate success, they should be repeated with distinctly negative results before we are warranted in asserting that it is impossible to communicate the disease by way of the stomata. The other three methods were each successful. A great many leaf-infections were obtained and forty of the inoculated plants also showed the characteristic signs in the bulbs at the end of 2 to 4 months. From the interior of the bulbs which became diseased in this manner this same organ- ism was re-isolated on several different occasions and grown in pure cultures which again produced the typical disease when re-inoculated. All the several hun- dred control plants maintained by the writer continued free from this disease. There remains, therefore, no good ground for doubting the general correctness of the statements advanced by Wakker as to the cause of this disease. It is not only a genuine bacterial disease, but one of the most peculiar and interesting vascular diseases known to the writer. The natural methods of infection (except from mother-bulbs to daughter- bulbs) are not well understood. The disease is readily induced through wounds and it is likely that the knife of the gar- dener is responsible for a portion of the infections. Inasmuch, however, as in many of the plants the signs are said to begin on the leaves at a considerable distance from the ground some other explanation must be sought, at least for a portion of the infections. On several occasions the writer succeeded in producing the disease in the bulbs by putting drops of infectious fluid into the flowers (fig. 132). It is possible, therefore, that the disease may be disseminated both by leaf-eating and by nectar-sipping insects. Signs have not been observed in the roots. According to Wakker, wet weather greatly favors the progress of the disease, while sunshine and dry weather are unfavorable to it. This is true, also, of many other bacterial diseases, c. g., tobacco-wilt and pear-blight. VARIETAL RESISTANCE. Twenty-five years ago it was common observation, according to Dr. Wakker, that some varieties were very little subject to this disease in fields where other varieties were badly attacked. He seems to have had no doubt about the "predisposition" of certain Fig. 132. *Fig. 132. — Cross-section of base of hyacinth bulb showing cavity in the bundle due to Bacterium hyacinthi. Plant No. 67 inoculated through the flowers. Slide 502 A-A9. Drawn with Zeiss 16 mm. and 12 ocular. 338 BACTERIA IN RELATION TO PLANT DISEASES. varieties to attack. He obtained lists of sensitive and resistant varieties from seven growers. A study of these lists showed a very general agreement upon the following varieties as very susceptible : Snowball, single white; De Candolle, single blue; Grand Lilas, single blur; Princess hereditaire des Pays-Bas, single whitt . La Tour d'Auvergne, double white; Grand Vainquer, single while; Duchess of Richmond, single red; Mimosa, single blue; L'Ornement de la Nature, single red; Alba Maxima, single while; Orandates, single blue; John Bright, single blue; Mirandolina, single while; The following varieties were regarded by the same growers as resistant, or only slightly susceptible : Pieneman, single blue; Argus, single blue; Norma, single red; Bleu mourant, single blue; Pucelle d'Orleans. single -white; Willem I, single blue; Voltaire, single white; Maria Catherina (Robt. Steiger), single red. These two lists are copied from Wakker's French paper published in Archives neer- landaises. In 1906 two of the leading hyacinth growers of Holland, men who have been in the business many years, were good enough to go through their catalogues and mark off for me, independently, varieties " much subject " to Bacterium hyacinthi, and varieties "very resis- tant," leaving those varieties unmarked which are intermediate in resistance, or about which there might be a difference of opinion or a lack of exact knowledge. In this way I obtained information concerning about 300 varieties. I have carefully compared the two lists and find them flatly contradictory only in the case of three varieties — single red Pelissier, single rose Maria Cornelia, and single white LTnnocence. There are, however, numerous discrepancies, many varieties being marked + or o by one grower and left unmarked by the other. This might mean either entire lack of knowledge, or difference of opinion. In case it were disagreement we might assume either difference of behavior on the part of particular varieties in the hands of different growers, something not improbable, or that one man is a closer observer than the other. Of course if a variety is much subject to the disease in one field and resistant to it in another, the observed resistance might be entirely a matter of accident and not due to any strongly inherent peculiarity, of which one might take advantage in cross-breeding or selection. Varieties to the number of 130 were marked + or o by one or other of the two growers, 73 as much subject to the disease, and 57 as very resistant. One grower reported 59 varieties "much subject" and 43 "very resistant," the other reported 22 varieties "much subject" and 32 "very resistant." The agreements are of greater interest. There are 26 of these, a number apparently too large to be purely accidental. In addition it should be mentioned that a number of the older varieties, marked + in one catalogue, are not included in the lists of the other dealer, but must at one time have been grown by him, and may have been discarded on account of disease. The writer started inquiries to determine this point and found this supposition correct. Including these, also, the number of agreements is 35. The writer subsequently received a catalogue from a third large grower, a person well- known in the trade for thirty years or more. This man marked 61 varieties "much subject" and 13 varieties "very resistant." The varieties concerning which this third grower is in agreement with the other two growers are marked with an asterisk in the following table Respecting the other varieties in this table he makes no statement. He is in contradiction with one or other of the two growers, never with both, respecting the susceptibility of three varieties: Clio, single light blue; Obelisk, single yellow; and Czar Nicholas, double rose. He reported on 41 varieties not mentioned by the other growers, 5 being marked as resistant. The varieties concerning which the two growers agree are included in the following lists: The starred varieties represent the agreements subsequently received from a third grower. YELLOW DISEASE OF HYACINTHS. Table Showing Sensitive and Resistant Varieties of Hyacinth. 339 Hyacinths much subject to the yellow disease. Hyacinths very resistant to the yellow disease. Charles Dickens, single rose *Robert Steiger, single red *La Grandess. single white Baron van Tuvll, single rose *La Neige, single white Moreno, single rose Captain Boyton, single light blue Baroness van Tuyll, single while *Czar Peter, single light bluei *Grandeur a Merveille, single bluish white *Grand Lilas, single light blue La Franchise, single bluish white "Lord Derby, single light blue *Grand Maitre, single light blue "Hermann, single yellow *La Pevrouse. single light blue *La Tour D'Auvergne, double white Regulus, single light blue To this list the following should be added *King of the Blues, single dark blue as aresultof the additional inquiry respect- *Marie, single dark blue ing discarded sorts already mentioned: King of the Yellows, single yellow La Precoce, single while Yellow Hammer, single yellow Marie Stuart, single white Princess Roval, double red Miss Nightingale, double while Boquet Roval, double rose *Mont Blanc, single while La Yirginite, double bluish while *Orandates, single light blue Crown Prince of Sweden, double dark blue Prince de Taillerand, single light blue *King of the Blacks, single blue black *La Citroniere, single yellow Minerva, double yellow fFound also accidentally by the writer in 1897 to be specially sensitive to pure-culture inoculations of Bad. hyacinth i. For comparison with the above table I have compiled the following list from Wakker's Dutch report for 1885, the same being the opinion of six hyacinth growers of that time respecting susceptibility to this disease of 34 of the 35 varieties mentioned in the preceding list. When the sum is less than six the other growers made no report. Number of growers reporting the variety as: Variety. Very susceptible Intermediate. Not susceptible. Charles Dickens, single rose La Grandess, single white La Neige, single white Captain Boyton, single light blue Czar Peter, single light blue Grand Lilas, single light blue Lord Derby, single light blue Hermann, single yellow La Tour d'Auvergne, double while- La Precoce, single white Marie Stuart, single white Miss Nightingale, single while Mont Blane. single white Orondates, single light blue Prince de Taillerand. single light blue King of the Blacks, jingle blue black La Citroniere, single yellow Minerva, double yellow Robert Steiger, single red Baron van Tuyll, single rose Moreno, single rose Baroness van Tuyll, single white Grandeur a Merveille, single bluish while Grand Maitre, single light blue La Pevrouse, single light blue Regulus, single light blue King of the Blues, single dark blue . . Marie, single dark blue King of the Yellows, single yellow . Yellow Hammer, single yellow Princess Royal, double red Boquet Royal, double rose La Virginite, double bluish while Crown Prince of Sweden, double dark blue. 34° BACTERIA IN RELATION TO PLANT DISEASES. From a comparison of these two tables it appears that none of the varieties regarded as very susceptible twenty five years ago have ceased to be susceptible, but rather, that some of them have been discarded by certain growers on this account. On the contrary, some of the varieties then regarded as very resistant continue to be very resistant, e. g., Robert Steiger, single red; Baron van Tuyll, single rose; Grandeur a Merveille, single bluish white; Marie, single dark blue. It also appears that some varieties then regarded as resistant are now classed as very susceptible, e. g., Charles Dickens, single rose; La Grandess, single white; Czar Peter, single light blue; Hermann, single yellow, and Mont Blanc, single white. Respecting these latter cases, we may assume (i) that they have changed in this par- ticular during the last two decades, or (2) that they were susceptible from the start, but had not at that time been subjected to rigorous tests calculated to bring to light their inherent susceptibility. For the sake of those who may desire to see all the discrepancies, I append also my entire list, premising that o means very resistant; + much subject; and § intermediate or no report. To find the varieties not marked by these growers the reader has only to consult any good catalogue of Dutch bulbs. Table giving complete list of varieties mentioned by one or more of the three Dutch hyacinth growers in 1906, as resistant or susceptible to the yellow disease: Single red : Amy o § § Etna § § o Garibaldi + § § Gertrude §00 Homerus §00 Incomparable + § § King of the Reds + § § Linnaeus + § § Pelissier o + § Prima donna + § § Queen Mary + § § Robert Steiger 000 Roi des Beiges § § o Von Schiller + § § Vuurbaak + § § Single rose : Baron v. Tuyll o o § Beautv of Waltham+§ Carlyle+§ § Charles Dickens-\--\-^ Dr. Livingstone + § § Maria Cornelia +0 § Gertrude § § o Gigantea o § o La joyeuse + § § Lady Derby o § § Moreno o o § Norma o § § Princess Helene § § + Rosea maxima o § § Rosine § o § Sophia Charlotte § § + Sultan's Favorite § + § Windhorst § § + Single white: Alba maxima § § + Alba superbissima § § + Albertine § + § Augenis Christiana § o § Baroness van Tuyll o o § Blanchard § § + British Queen § § + Crown Princess + § + Grand Vainqueur § § + Grande Vedette + § § Jenny Lind § § + La belle Blanchisseuse O § § La Grandesse + + + La neige + + + La precoce + §§ L' Innocence o + § Madam van der Hoop § § + Mary Stuart + § § Mina § § + Mont Blanc + § + Paix de l'Europe § § o Mr. Plimsoll § o § Queen of England +§§ Snowball + § + White Bird o § § Grandeur d Merveille 000 La Franchise o o § Lord Grey + § § Mammoth o § § Voltaire O § § Single light blue: Blondin § § o Captain Bovton + + § Clio §0 +' Czar Peter + + + Enchantress § § + Grand Lilas -j- + + Grand Maitre 000 La Peyrouse 000 La Precieuse +§ § Leonidas § + + Lord Beaconsfield § § + Lord Derby + + + Lord Palmerston § + + Orandates+ § + Pieneman § o § Prince de Taillerand + § § Princess Mary of Cambridge -\- f Queen of the Blues § H — \- Regulus o o § Sehotel + § + Turquoise § o § Single dark blue: Argus § + + Bleu mourant § o § Baron van Tuyll § o § King of the Blues 000 La nuit § § + Leopold II o § § Lord Mayo § § + Marie 000 Baron von Humboldt + § ^ General Havelock -f- § + King of the Blacks +§ + Masterpiece +§ + Mimosa § + + Pasteur + § § Sir Henry Barkley +§ + Uncle Tom § § + William I + § + William III § § + Single violet, lilac, and mauve: Distinction o § § Lady Stanhope § § + Laura § § + L'Honneur d'Overveen+ § § Lilas enorme § § + L'L;nique § § + Mauve Queen § o § Sir Edwin Landseer § § + Sir William Mansfield § o § Single yellow and orange: Anna Carolina +§ § Bird of Paradise + § + City of Haarlem § § + Ducde Malakoff§ § + Hermann + + + Ida +§ + King of Holland § § + King of the Yellows o o § La Citroniere +§ + La Grande Jaune § + § L'Or d Australie § + + Marchioness of Lome § § + Obelisque o § + Yellow Hammer o o § Double red: Eastanjebloem o § § Noble par merit e O § § Princesse Louise + § + Princesse Royale o o § YELLOW DISEASE OF HYACINTHS. 341 Double rose: Boquet royal o O § Boquet tendre § § + Czar Nicholas o § + Dagmar § § + Frederick the Great § § + Globosa § § + Grootvorst o § § Leo + §§ Le Grand Concurrent o S S Princesse Alexandra § o § Princesse Louise § § + Venus de Medici -f § § Double white: Boquet Royal § + § Flevo §00 Florence Nightingale + § + Grand Vainqueur + § + Jenny Lind § § + La Grande Duchesse § § + La Grandesse + § + La Tour D'Auvergne + + + Princesse Metternich § § + Isabella o § § Fig. 1 33. Double white — continued: La Virgin it e o o § Prince of Waterloo +§ § Triumph Blondine o § § Double light blue: Madame Monmouth + § S Rembrandt o § § Yan Speyck +§ § Double dark blue: Crcnun Prince of Sweden o o § Laurens Koster + § § Lord Raglan o § § Prince of Saxe Weimar o § § Double yellow and orange: Boquet d'Orange +§ § Minerva -f § § Sir Rowland Hill § § § Souvereign o § -f- (by letter §) Sunflower o § § Double violet: La Victoire + § § From the third grower I received the following letter and supplementary list of varieties : I have much pleasure in handing you a catalogue in which, according to your suggestion, are marked the hyacinths which are much subject to the yellow disease and such which are very resistant. It must, however, be understood that similar statements have only relative value. Many varieties which used to be among the best trade-sorts a few years ago, have now entirely disappeared in con- sequence of the yellow disease and it may be expected that the list of varieties after ten years will be very much reduced. The following list of hyacinth varieties shows those which have entirely or nearly dis- appeared in consequence of the yellow disease : Single red and rose: Emilius Ornement de la Nature Mrs. Beecher Stowe Unica spectabilis Duchesse de Richmond Mars Lord Grey Lord Percy Single white: Madame de Stael Queen Victoria Miss Nightingale Single blue: Orondates Emilius Prinz Albert von Preussen Nimrod Siam Grand Vainqueur Argus De Candolle Porseleinen Scepter Priestley Ferruck Khan von Schiller Clio Single blue — continued: King of the Blacks The Sultan Single yellow: La Pluie d'Or Ovenvinnaar La Citroniere Single violet : Haydn Jesko Sir Henry Havelock Double red: Regina Victoria Double white: Anna Maria van Hoboken Mont St. Bernard Kmperor of the double Whites Double blue: Louis Philippe Lord Raglan Frans Hals Duke of Norfolk Double yellow: Jaune supreme *Fig. 133. — Detail from a cavity made by Bad. h vac in/hi in a bulb-scale of hyacinth showing separated and crushed cells lying in a mass of bacteria — cavity like that shown in fig. 132, but larger. Slide 507 B 6, third section fromright. Material collected by the writer at Haarlem in 1906. Gram's stain modified by substitution of amylic alcohol for ethyl alcohol. Starch grains on bottom at right. 342 BACTERIA IN RELATION TO PLANT DISEASES The new varieties are also cataloged by one of the growers, and of these 7 are marked as much subject and 6 as very resistant, while 31 do not fall into either class. Wakker states that most of the leaf-infections are usually observed in the fields in the month of May, but thinks that infection must take place at least a month earlier. The writer's experiments have led him to the same conclusion. Undoubtedly the bulk of the field infections occurs during blooming time, when insects would be visiting the blossoms freely. The downward movement of the disease in the leaves is very slow (Wakker, Smith). Fig. 135* Fig. 134.* MORBID ANATOMY. There are no hyperplasias in connection with this disease. It is primarily, and to a considerable extent during its whole progress, a disease of the vascular bundles. The reasons for this the writer has attempted to set forth in his papers on this organism ( see Literature). They are noted briefly under the next head. The xylem portion of the bundle is the first part to be attacked, especially the spiral vessels which are soon filled entirely by the rapid multiplication of the organism (plate 20, fig 6). This is what causes the bright *Fig 1 ,1 Cross-section of base of a hyacinth bulb, showing cavities in parenchyma due to Bad, hyacinthi. Upper pari .1! drawing is extreme base of a bulb scale; lower is part of plateau. Starch grains are represented in outline onl> . Slulr 502 A A3, from plant No. 67 inoculated in the flowers (see Bull. 26, p. 30). •Fig. 13.5. — Bat! hyacinthi: A detail from lig. 134 at X. VKLI.OW DISEASE OF HYACINTHS. 543 yellow color so conspicuous in the attacked bundles. Often the xylem is the only part of the bundle attacked. When the vessels become thus occluded the walls give way, probably by solution, and the bacteria flood out into the surrounding parenchyma, which, however, is quite resistant. In the end, cavities are formed in the parenchyma surrounding the bundles, but the progress of the disease in this tissue is extremely slow. These cavities are filled with remnants of the vessels and cells of the host-plant and by enormous numbers of the yellow bacteria. In the formation of cavities in the parenchyma the intercellular spaces are first occupied (figs. 133, 134, 135). In all of the diseased bulb-scales examined by the writer prior to 1906 the bulk of the tissues was still sound and the organisms were either confined to the bundles, or had made compara- tively small cavities around the same. In several hundred bulbs examined in Holland, in August, 1906, the disease had made more extensive inroads (fig. 136) and large areas, especially of the inner face of the diseased scales, were yellow and gummy (plate 20, figs. 8, 9, 10). The disease seems to progress a little more rapidly in the base of the bulb, where there is a net-work of vessels in rather close connection. Here also cavities are formed in the tissues. A small por- tion of the base of a bulb in an early stage of infection is shown in fig. 132. In the end the whole plateau becomes yellow and gummy and the surface is ruptured, letting in vari- ous molds and bacteria. The writer has not attempted to cut many sec- tions of diseased leaves, but Wakker did so carefully, after fixing in ab- solute alcohol, and showed that here also the downward movement of the organism is through the spiral vessels of the xylem. The few I have cut and examined confirm Wakker's view (see figs. 137, 138). In the end, the whole plant is de- stroyed, but, so far as the writer has observed, when the disease is uncomplicated, there is never anything resembling a soft white rot, such as that described by Heinz. In none of the many bulbs examined by the writer in 1897-1901 had the disease progressed far enough for the organism to break out of indi- vidual scales and pass sidewise into the open spaces between the scales, but this phenomenon was observed in Holland in 1906. Although the action on the cell-walls is slow, there can be little doubt I think that in the end the cell-wall proper as well as the middle lamella is dissolved and disappears. I have not established this fact, however, beyond dispute. *Fig. 136. — Cross-section of 6 hyacinth bulbs from a field near Haarlem, showing advanced stages of the yellow disease due to Bad- hyacinthi. Photographed by the writer in the summer of 1906. Fig. 136.' .544 BACTERIA IN RELATION TO PLANT DISEASES. THE PARASITE. Bat terium hyacinthi* Wakker is readily isolated. In the plant and on agar and in beef- broth, etc., it is a short rod, single or in pairs, or more rarely in fours joined end to end (figs. 139, 140). Rarely short chains have been observed, c. g., on agar. It measures under these circumstances 0.4 to 0.6 yuXo.8 to 2 fi, but like many other organisms, it is longer or shorter, thicker or thinner, according to age, culture-medium, and kind of stain used. It is gen- erally slenderer than Bad. campestre or Bad. phaseoli. The following are some measurements: (1) February 5, iSg8. Slime from a daughter-bulb stained 5 minutes in a saturated solution of basic fuehsin (very weak stain). Rods short, 0.5 to 1.0 X0.4 to 0.5(1. Two minutes in saturated watery solution of Gentian violet gave a deeper stain but not deep enough. (2) February 7, / Alkaline beef -broth, No. 1, Jan- uary 29, 1898, stained 10 minutes in saturated watery solution of basic fuehsin. 2.0X0.4/1] 1 4Xo.4/n}SingIc rods. 1 .oXo.4mJ 3.6X0.4/1 two rods joined end to end. 3.2X0.4/1 two rods joined end to end. W'iilt it rods seen 0.611. (3) July 31, iSgS. Slide of March 10 from very dilute beef -broth 3 days old, Moore's flagella stain. Size 1 to 2X 0.5 to 0.7 /J.. Flagella 3 times length of rods. (4) August 3, i8g8. Slide of March 17, 1897, agar stock 207, Fischer's flagella stain : 2X0.8/1 .'Xl Of. 2 to 3 X I .Ofi. 2.5 X0.8 to 1 .o/i several. (5) August 3, 189S. Slide of June 23, 1897, made from the interior of a bulb (yellow slime). Plant inoculated on leaf February 16, 1897. Stain, basic fuehsin in water. 1.0X0.5// two; 1.5X0.5/4; 1.2X0.5/1; Most 1 to 1.2X0.5/1; Extremes, 0.9 to 1 .5X0.5/1- Pseudozoogloeae are common. No spores have been discovered by the writer, and the spores described by Wakker probably belonged to some other organ- ism, f This is the more likely because the cultures in which they developed abundantly were made directly from the bulb, i. c, not from colonies, and were kept at a temperature slightly above the maximum for the growth of this organism, as determined by the writer. Making cultures from bulb scales in the same way as Dr. Wakker, the writer has twice obtained mixed growths from what looked like an unmixed source ; the yellow organism being contaminated once by a green fluorescent organism and once by a white, gas-forming species. In the plant and in the common culture media chains and filaments do not occur, or are rare, but old cultures on media rich in sugar, c. g., streaks on dextrose-agar or saccharose-agar, often Fig. 1374 "Synonyms: Bacillus hyacinthi (Wakker) Trevisan; Pseudomonas hyacinthi (Wakker) EFS. Ill- endospores observed by Wakker were blue-shining, strongly refractive bodies, germinating equalorially. The) were cylindric with rounded ends, measuring i/t in length and being about one-half or two-thirds as thick. They sometimes appeared at temperatures lower than 35° C, but less abundantly. th'ir,. 137 C'niss section of a hyacinth-leaf showing xylem part of tin bundle occupied by a bacterial cavity, ithei side being unoccupied. Leaf inoculated at apex in 1898 with a culture of Bad. hyacinthi. To either side of the bundle are (C C) natural passage-ways through leal. Slide .so-' B-Ay. YELLOW DISEASE OF HYACINTHS. 345 show many long slender chains and also filaments 50 to 150 /x long in which no septa are visible (fig. 141). The organism is motile, and a polar flagellum has been demonstrated (fig. 142). Involution forms occur. Rods from young cultures stain readily; those from old cultures and from the overcrowded vessels take stains less freely. Wakker recom- mended Bismark brown (phenylene brown). The writer has used Ziehl's carbol fuchsin, and Gram's stain with amyl alcohol. The purest color of the organism is bright yellow (gamboge, chrome, canary, or pale cadmium). The color is best developed in the plant and near the surface of fluid and solid culture media. When the air-supply is scanty the color is pale yellow. Duller and paler yellows occur also in certain media when oxygen is abundant, e. g; in potato-broths and acid beef-broths (not alkaline ones). In peptonized beef-bouillon, neutralized by sodium hydroxide, the color was canary yellow. The color was also bright in nutrient gelatin containing malic acid. This color appears to be a lipoehrome compound, as it is associated with a fat. It is soluble in acetone, glycerin, a water solution of ammonium carbonate, or hydrogen peroxide, and slowlyalso in strongammonia water, glacial acetic acid, ethyl acetate, ethyl alcohol, and methyl alcohol. The pigment can not be extracted with petroleum ether. The acetone extract, which also removes the fat, yields a blue-green or purplish reaction (N\ Fig. 139.f Fig. I40.{ Fig. 142 '/too m m Fig. 138.' Fig. 14l.§ with concentrated sulphuric acid, and is readily destroyed by light. The yellow pigment is also bleached by reducing agents, the color returning on their removal (for further details consult Bulletin 28). The brown stain appears to be similar to that developed by Bad. campestre, but is less pronounced. It is soluble in water, and free oxygen appears to be necessary for its formation. It was best developed in hyacinth-broth, potato-broth with peptone, and on steamed turnips, radishes, and yellow banana rinds standing in distilled water. Sienna and burnt umber were the darkest shades observed (old cultures on radish and turnip). It was not observed in beef-broth, nutrient agar, starch-jelly or nutrient gelatin. It occurs in the plant, so far as observed, only in the vascular bundles of the leaves, and is not pronounced. In nutrient media it is best observed in old cultures. Bad. hyacinthi is not sensitive to dry air. Wakker made this discovery and the writer has confirmed it. Thirteen cover slips, which were spread by the writer with a thin layer of bacteria from a young potato-culture and dried for 9 days, each infected culture media when thrown into it. Two of the same covers dried for 47 days at room-temperatures also yielded pure cultures when thrown into beef-bouillon. In another experiment 17 out of 18 similar covers infected beef-bouillon after being dry for 49 days (compare in this particular with B. tracheiphilus and B. carotovorus) . *Fig. 138. — A detail of Bad, hyacinthi from fig. 137 at X. Slide 502 B-A7. fFic 139. — Rods of Bait, hyacinthi. x 4000. After Wakker, Verslag, 1883, pi. I, fig. 1. |Fig. 140. — Rods of Bad. hyacinthi: a, directly from bulb; b, from a young beef-bouillon culture, x 1000. §Fig. 141. — Filaments of Bacterium hyacinthi from a culture on cane-sugar agar.segments not visible. Stainedby van Ermengem's nitrate of silver method, x 1000. ||Fig. 142. — Rods of Bacterium hyacinthi, showing flagellum: a, stained by V. A. Moore's modification of Loeffier's method; b, stained by Alfred Fischer's method, x 1000. 346 BACTERIA IN RELATION TO PLANT DISEASES. This organism does not produce acids in milk, but there is a slowly increasing alka- linity, and after some days (3 to 10 or more) the casein is precipitated as a finely divided, voluminous, mobile mass, which settles slowly. These phenomena are best observed in litmus-milk. The litmus in such cultures is slowly reduced, but on the death of the organism it is oxydized back into a deep blue. In the end the casein is partially peptonized, but this change does not occur rapidly. The organism makes a good growth in milk and forms a bright yellow rim (plate 20, figs. 2-4), and sometimes a pellicle. In old cultures sheaf- like crystals of tyrosin occur. In April, 1898, two 10 cc. tubes of milk, which had received 4 steamings and been under observation for a month unchanged, received 200 mgs. each of thymol. One was put away as a check, the other received 8 cc. of whey from a milk-culture of Bad. hyacinthi 33 days old, after this had been heated in the water bath for 10 minutes at 51. 8° C. (40 above the thermal death point). There was no change in the check tube. In the other, there was copious precipitation of the casein in 48 hours, but no evidence of bacterial growth either then or subsequently (11 days). In October, 1898, the experiment was repeated with the same result. In this experiment (fig. 143) 10 cc. of sterile milk received 3 cc. of whey from a milk-culture 10 days old. One hour after adding the whey the tube was heated for 20 minutes at 52°C. in the water bath to destroy the bacteria. In 24 hours the milk was entirely coagulated. A small drop from this tube was now transferred to bouillon but did not cloud it (5 days). At the same time another tube of the same milk received 3 cc. of whey from an- other milk-culture of the same organism, the only difference being that in this case the milk was heated for 10 minutes at 8o°C. after adding thewhey: This, to destroy the supposed enzyme. Result: No change in the milk (7 days). These experiments indicate the presence of a lab ferment. In litmus-milk (and in bouillon) contain- ing ethyl alcohol, a volatile acid is formed, and there is a fragrant odor in the steam. Methyl alcohol is not decomposed. There is a moderate, smooth, wet-looking growth on steamed potatoes standing in dis- tilled water, but it is not prolonged or copious, the color on potato is at first wax-yellow, but after some time it is dulled to a brownish yellow. At 2o0C. to 25°C. the streak is not visible until the second or third day when inoculations are made from fluids. Growth on potato- cylinders is much increased by the addition of a little cane-sugar, dextrin, or malt-diastase. The action of the organism on starch is feeble, and the water surrounding the potato is never converted into a solid mass of slime as in case of Bad. phaseoli, Bad. campestre, Bad. juglandis and other starch-destroying organisms. On potato cylinders first soaked in pure water to remove the slight amount of sugar and acids on the cut surface and then tubed, growth did not occur or was delayed and scanty. Young cultures have no smell; in old cultures there is a feeble odor. Type of growth on potato like Bad. stewarti (PI. 17, fig. 2). 'Fig. [43. Two tubes of sterile milk to which was added equal volumes of whey from old cultures of Bad. hyacinthi in milk. Whey added to tube b after heating to 8o° C. (to destroy enzyme); whey added to tube a after heating t0 52°C. (sufficient to destroy bacteria only I. Result Milk curdled promptly in tube a and remained unchanged in the other. Oct. 1X98. 143. YELLOW DISEASE OF HYACINTHS. 347 On yellow turnips prepared in the same way, growth was very much greater than on potato. Such turnips contained much more sugar than the potato. Turnip and carrot cylinders were softened by the long continued growth of this organism (middle lamellae). Growth on nutrient starch-jelly is also very slow, even when hyacinth-starch is used. When diastase was added to the jelly, increased growth was apparent at once (see Smith, Bulletin 26, plate I, figs. 15-16), and at the end of 35 days this was estimated at 200 times the volume in the check-tubes. At the end of 62 days (water being well retained by the medium) there was a thin canary-yellow layer over the surface of the check-tubes (Stock 310, for composition see Bad. phaseoli) equal to the growth given by the other tubes at the end of 5 days. The body of the starch in the check-tubes still preserved its bluish lustre, and on testing with Soxhlet's solution for sugar more than nine hundred and ninety-nine one thousandths of the starch was found unchanged. The only copper reduction on boiling 3 minutes was in an exceedingly thin film immediately under the bacterial layer. No brown pigment was formed on this substratum, with or without the diastase, and the color of the slime was much brighter yellow than that in corresponding tubes of Bad. campestre or Bad. phaseoli. There is always a strong iodine-starch-reaction, even in old cultures on starchy media, but some of the starch gives a red reaction (amylo- dextrin). Gelatin (fig. 144) and Loeffler's blood serum are liquefied, but the change takes place slowly, does not occur in the absence of air, and is usually inhibited by the pres- ence of 5 or 10 per cent grape- sugar or cane-sugar. Dextrin stimulates growth ; glycerin in small doses does not increase growth (?); in large doses it retards growth. In moderate doses grape-sugar, fruit-sugar, and cane-sugar stimulate growth. Lactose, maltose (?) and mannitol have no marked effect on growth. Bad. hyacinihi made a very feeble growth in a synthetic medium made as follows : Distilled water 400 ; sodium acetate 2; dipotassium phosphate 0.8; magnesium sulphate 0.04; ammonium phosphate 0.04. The behavior was much the same as in Uschinsky's fluid. This organism does not produce gas, and will not grow in the closed end of fermentation tubes in peptone-water, or peptonized beef-bouillon with grape-sugar, fruit-sugar, cane- sugar, milk-sugar, maltose (?), galactose, dextrin, glycerin, mannit, ethyl alcohol, methyl alcohol, or potassium nitrate. For some days there was no growth in the closed end of the tubes containing peptonized beef-bouillon and ordinary commercial maltose, but in the end there was a very feeble clouding in the closed end. Two repetitions of the experiment gave the same result, and no air-bubble appeared in the closed end of the tube on subsequently steaming it. In a third repetition made in 1906 using a very pure maltose there was no growth in the closed end (fig. 145). The organism is therefore, as far as known, a Fig. 144.' *Fig. 144. — Old stab-cultures of Bad. hyacinihi in gelatin containing 0.6, 0.8 and 0.9 per cent malic acid showing slow liquefaction confined to upper part of gelatin. Rims and precipitate bright yellow. 348 BACTERIA IN RELATION TO PLANT DISEASES. strict aerobe, except in the presence of maltose or some unknown substance sometimes contaminating the maltose; a matter open to further inquiry. Cane-sugar is inverted. Oxydase and peroxydase are absent, i. c, cultures give no reaction with guaiac resin in alcohol, nor on addition of hydrogen peroxide. Tyrosinase perhaps occurs, i. c, the substance causing the brown stain. Catalase is present, i. e., some body yielding a copious evolution of gas when hydrogen peroxide is added to old potato cultures. This substance is destroyed at 850 C. Large doses of grape-sugar or cane-sugar in slant agar retard growth at first (9 per cent grape, 17 per cent cane) and then stimulate it greatly. On 10 grams of recently slanted nutrient agar containing 3 grams of grape-sugar no growth was obtained. A small amount of non-volatile acid is developed out of various sugars — grape-sugar, fruit-sugar, cane-sugar, galactose, maltose; and in old cultures on the following substrata: carrot ( occasionally), rutabaga, sweet potato, sugar-beet. The steam from old cultures in hyacinth-broth caused a copious rusty precipitate when conducted into Nessler's solution, indicating the presence of ammonia or amins. This organism grows readily on all ordinary culture- media except when it is too salt or too acid. It is very sensi- tive to acids, even those in the parenchyma-juice of the hyacinth retard growth (clouding) decidedly (14 days, 16 days, or more). Growth did not take place in potato-broth ( + 30), in juice of slow-growing cabbage leaves (+49), cauli- flower-broth, sugar-beet juice diluted with water, or juice of green or ripe tomato-fruits ( + 59 and +64) ; growth was also much retarded in acid beef -broth ( + 40). The bacterium would not grow in beef -broth concentrated by boiling ( +80). When the acidity of the +30 potato-broth was reduced slightly ( + 28, +26), by sodium hydrate, growth took place. Growth in beef-broth was retarded (5 to 7 days) by 1.5 per cent c. p. sodium chloride. It does not grow well in Uschinsky's solution; in this medium there was either no growth or it was long delayed and feeble (and without much yellow color) unless peptone was added to it, in which case growth was centupled. Methylene blue in Dunham's solution was reduced within a few days, but reoxydized quickly on shaking, and was bright blue on the death of the organism, the bacterial precipitate being unstained. Indigo carmine in Dunham's solution changed from a dull blue to a bright blue and retained this color for a long time, but finally became yellowish. Rosolic acid in Dunham's solution with enough c. p. hydro- chloric acid to render the medium yellowish did not redden but became colorless; the bac- terial precipitate, on the contrary, became rosy or salmon colored. Acid fuchsin in Dunham's solution bleached slowly, the color being all gone in about 4 weeks. Litmus in various media was bleached very slowly, the reduction being evident usually only at the close of the second or third week. The optimum alkalinity for growth in peptonized beef-bouillon lies between o and +15 of Fuller's scale; the maximum tolerated alkalinity (sodium hydroxide in bouillon) is more than —20 and less than —40; in bouillon the tolerated acidity is about +30 (malic acid) Fig. 145. 'Fig [45. Behavior of Bact. hyacinthi in a fermentation-tube containing water, 1 per cent Witte's peptone, and 1 m maltose (the latter 3 times recrystallized). Culture cloudj in open end and clear in closed end alter S days Drawn Oct. is. [906. YELLOW DISEASE OF HYACINTHS. 349 Fig. 146.* to +40 (lactic acid). Growth was retarded decidedly by +30 bouillon (to the 18th day or longer). This organism produces indol in peptonized beef-broth or peptonized Uschinsky's solution, but not so abundantly as Bacillus coli. Lead acetate paper was browned, indicating slow evolution of hydrogen sulphide, when kept in the top of the test tube over certain cul- tures, c. g., coconut-cylinders (fig. 146'), but notwhen kept over others, e.g., potato-cylinders, turnip-cylinders (fig. 1462). In most cases, if the culture-medium developed the brown stain the sensitive paper remained unstained ; if the culture remained free from the brown pigment the lead acetate paper was darkened. The only exception noted was yellow globe turnip : here both paper and substratum were browned. Nitrites are not produced from organic nitrogen (beef-broth, peptone), nor from potassium nitrate in peptonized beef-broth. This organism is not a strong smelling germ. It is not readily destroyed in ordinary culture-media by its own decomposition products nor in mixed growths. It grows well with a bright yellow color and without retardation on steamed coconut-flesh, standing in distilled water. On this medium in a scanty air-supply (in vacuo, mercury at 3 inches) the growth was paler yellow than on the checks (bulk for bulk, examined on white paper) ; the same result was obtained on potato. In agar and gelatin the growth is best toward the surface of the stabs. A whitish chemical halo forms slowly on the surface around the bacterial growth; this is soluble in acids, and does not appear when grape-sugar or cane-sugar is added. Growth does not occur in an atmosphere of pure hydrogen, nitrogen, or carbon dioxide, and exposure to these gases retards subsequent growth in the air, or prevents it altogether, if the exposure is longer than a few days. The organism is more tole- rant of these gases on some media than on others, e. g., subsequent growth in air after 10 days' exposure to car- bon dioxide was retarded in beef-broth, but not on coconut cylinders. Growth in vacuo is feeble or alto- gether wanting, according to the completeness of the exhaustion of the air. Buried colonies in agar and gelatin are small, grow slowly, and show no strong tendency to break through to the surface. Surface colonies on agar and gelatin are round, smooth, wet-shining, pale yellow with a thin dis- tinct margin and are not rounded up much. They grow slowly (fig. 147, and Bull. 26, plate 1, fig. 12). Streaks on sugar agars sometimes developed the surface shown in fig. 14S. The best growth in gelatin was in that made o on Fuller's scale. The growth in gelatin slightly acid or slightly alka- line to litmus was not nearly so good; growth was good, however, in o gelatin to which a small amount of malic acid was subsequently added ( +48 and +54 with 5 and 10 per cent cane sugar). Growth was very poor on acid and alkaline peptonized beef-broth gelatins ( + 40 and —20). Streak cultures came up slowly, even on the best (10 per cent) *Fig. 146. — Production of hydrogen sulphide by Bad. hyacinthi: 1, Lead acetate paper exposed (and browned) over culture on coconut flesh; 2, the same after exposure over cultures on potato or carrot. fFiG. 147. — Strips of Petri-dish agar-poured-plates of Bad. hyacinthi. showing slow growth of the colonies: