LIBRARY Michigan Stat: ' University This is to certify that the thesis entitled ECOLOGY OF XAIVTHOMONAS PHASEOLI AND XANTHOMONAS PIMSEOLI VAR. FUSCANS IN NAVY (PEA) BEANS (PHASEOLUS VULGABIS L.) presented by David M. Weller has been accepted towards fulfillment of the requirements for Ph.D. degreein Botany and Plant Pathology Major professor Date December 12;, 1978 0-7639 OVERDUE FINES ARE 25¢ PER DAY PER ITEM Return to book drop to remove this checkout from your record. ECOLOGY OF XANTHOMONAS PHASEOLI AND XANTHOMONAS PHASEOLI VAR. FUSCANS IN NAVY (PEA) BEANS (PHASEOLUS VULGARIS L.) BY David M. Weller A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1978 ABSTRACT ECOLOGY OF XANTHOMONAS PHASEOLI AND XANTHOMONAS PHASEOLI VAR. FUSCANS IN NAVY (PEA) BEANS (PHASEOLUS VULGARIS L.) BY David M. Weller Common and fuscous bacterial blights of bean (PhaseoZus vulgaris L.), incited by Xanthomonas phaseoZi (E.F. Sm.) Dows and X. phaseoli van fuscans (Burkh.) Starr and Burkh. (Xanthomonas campestris), respectively, are serious diseases of Michigan Navy (pea) beans. Rifampin-resistant mutants of X. phaseoli (Xp) and X. phaseoZi var. fuscans (pr) were screened for similarities to wild-type isolates and their utility in ecological studies. A mutant (Ra) isolate of Xp and one (R10) of pr were similar to wild-types in numerous bacteriological tests, grew in viva at rates identical to, and were as virulent as the wild-types in bean leaves. The doubling time for R10 and Ra was about 11% longer than that for the wild-types in buffered yeast-extract liquid medium. A mutant (R10-86) resistant to both rifampin and streptomycin was similar David M. Weller in virulence to its RlO parent and original pr wild- type. The rifampin-resistance marker permitted selective isolation of Ra and R10 from field-grown bean plants; growth of all phyllosphere bacteria was inhibited on media with rifampin (50 ug/ml). Population trends of R10 and Ra in inoculated leaves of field-grown Navy beans were similar to a standard bacterial growth curve. Mean doubling times of R10 and Ra were 19.4 and 18.8 hours, respectively during the exponential growth phase. Leaf populations of R10 and Ra peaked during the stationary phase and remained stable until leaf abscission; a slow death phase accompanied leaf decomposition on the soil surface. Symptom development on leaves required minimal populations of 5 x 106-2 x 108 bacteria per 20 cm2 leaf tissue and usually corresponded to the early stationary phase. The ecology of Xp and pr was studied in field- grown Navy beans by monitoring populations of R10, Ra, or R10-S6 in plants grown from infected seeds or sprayed with the bacteria. All above and below ground portions of seedlings grown from infected seeds were colonized by the blight bacteria immediately after germination. Primary leaf colonization initiated vertical bacterial spread into the expanding leaf canOpy. By continually sampling leaves from the main stem of R10 or Ra inoculated plants, population profiles were established David M. Weller which described the multiplication and spread of the bacteria in the canopy from seedling to reproductive phases of plant development. The profiles were characterized as a series of growth curves displaced over time with each curve representing bacterial multiplication on an individual leaf relative to the primary leaf node. Correlation of the sequence of symptom expression with the population profile explained the phenomenon of late disease development which is characteristic of common and fuscous blights. The spread of R10 and Ra was facilitated by rain, bud colonization, and systemic movement. Ten to 50% of the R10 or Ra leaf population was removed as secondary inoculum during rainfall. Vegetative buds were continually colonized by up to 103 bacteria per bud and leaves from the buds were usually infected. Isolates R10 and Ra displayed a significant systemic phase with doubling times of 22.8 and 23.8 hours in stems. The population profiles of R10 and Ra in stems were characterized as a series of growth curves, each representing bacterial multiplication in a section of stem relative to the cotyledon node. Isolates of R10 and Ra were associated with Navy bean roots and occupied the rhizosphere throughout the growing season. Seeds externally infested with blight bacteria were shown to be an important source of primary inocula for David M. Weller common and fuscous blights and 14% of commercial Navy bean seed lots were contaminated. Surface populations of Xp and pr ranged from 0-4 x 104 bacteria per seed, however, minimal populations of 103-104 bacteria/seed were required for infected plants to be produced under natural growing conditions. Symptomless seed internally bearing Xp or pr was identified as a potential primary inoculum source. Seeds with symptoms were always associated with visibly infected pods; when pod infection results from systemic-borne bacteria, hairline suture lesions may be produced which are difficult to detect. This dissertation is dedicated to my wife ii ACKNOWLEDGMENTS I would like to express my most sincere appreciation to Dr. A.W. Saettler for his support and guidance throughout this study. His confidence in me and interest in my work has made my graduate education both enjoyable and productive. I am grateful for the time and guidance given by the members of my graduate committee, Drs. E.J. Klos, A.L. Jones, J.M. Tiedje, and M.W. Adams. My thanks to Sara Stadt, Sue Benson, and Sue Kaufman for technical assistance. I appreciate the encouragement and advice of all my friends in the Department of Botany and Plant Pathology; the many hours of professional and social interaction will long be remembered. I would like to acknowledge the Weller and Barum families for their moral support throughout my academic career. I am deeply grateful to my wife Suellen; her love, patience and support were important components in my graduate education. iii TABLE OF CONTENTS Page LIST OF TABLES.. ......... ...... ..................... vii LIST OF FIGURES........................ ...... . ...... ix GENERAL INTRODUCTION AND LITERATURE REVIEW .......... 1 GENERAL MATERIALS AND METHODS.. ..................... ll Bacterial storage and culture .................. 11 Bean culture................................... 11 Plant inoculation.............................. 12 Disease rating.......................... ....... 13 Bacterial isolation and identification.... ..... 13 LITERATURE CITED.......... ........ ..... ............. 15 PART I ISOLATION AND SCREENING OF RIFAMPIN-RESISTANT MUTANTS OF XANTHOMONAS PHASEOLI AND XANTHOMONAS PHASEOLI VAR. FUSCANS INTRODUCTION............ ....... . ................. ... 21 MATERIALS AND METHODS O O O O O I O O I 0 O O I O O O O O O O O O O O O O ..... 23 Rifampin agar medium.................... ....... 23 Isolation and culture of rifampin-resistant mutantSOOOO0.0.0............IOOOOOOOOOOOOOOO. 23 Physiological tests............................ 24 Greenhouse and field inoculation............... 24 RESULTS......... ........ ..................... ...... . 26 Comparison of rifampin-resistant mutants of Xp and pr and the wild-type parents......... 26 iv Recovery of rifampin-resistant mutants of Xp and pr from field-grown beans. .............. 30 DISCUSSION ..... . ...... ....... .......... . ............ 4O LITERATURE CITED ...... .. ...... . .......... . .......... 43 PART II POPULATION TRENDS AND DISTRIBUTION OF XANTHOMONAS PHASEOLI AND XANTHOMONAS PHASEOLI VAR. FUSCANS IN FIELD-GROWN NAVY BEANS (PHASEOLUS VULGARIS L.) INTRODUCTION ........................................ 45 MATERIALS AND METHODS ...... . ...... . ........... . ..... 50 Inoculation and isolation of R10, Ra, and R10-86 ooooooooooo oooooooooooo ..... o oooooooooo 50 Effect of simulated washing and rain on blight bacterial populations on Navy bean leaves... ......... . .......................... 52 Detection of surface-borne blight bacteria ..... 53 Production of bacterial blight infected Navy bean seeds .............................. 53 RESULTS. ..... . ......... .......... ........ . .......... 55 Multiplication of R10 and Ra in leaves of field-grown beans.. .......................... 55 Resident bacteria and yeasts associated with bean leaves...........OOOOOOOOOOOOOO. ........ 62 Multiplication and spread of R10 and R10-S6 in Navy bean seedlings....................... 65 Multiplication and spread of R10 and Ra in in leaves and buds of field-grown beans during the vegetative and early reproductive stages.......................... ........ ..... 69 Field observations of disease develOpment ...... 75 Multiplication and systemic spread of R10 and Ra in stems of field-grown Navy beans.... 75 Effect of washing on blight bacterial pOpulations of Navy bean leaves .............. 82 Rain-trapped Ra from field-grown Navy beans.... 85 Detection of blight bacteria on leaf surfaces.. 88 Effect of R10 and Ra populations on the severity of leaf symptoms .................... 89 DISCUSSION.. ........................................ 92 LITERATURE CITED .................................... 105 PART III PRIMARY INOCULA SOURCES OF COMMON AND FUSCOUS BACTERIAL BLIGHTS INTRODUCTION ........................................ 111 MATERIALS AND METHODS ............................... 113 Surface sterilization of bean seed ............. 113 Externally contaminated bean seed .............. 114 Internal seed infection ........................ 116 Symptomless internally-infected seed.. ......... 116 Overwintering of R10, Ra, and R10-86 in bean refuse..... .......... . ...... ...... ........... 117 RESULTS .............................. . .............. 119 Surface-infested seed as a source of primary inoculum ..................................... 119 Internal seed infection and bacterial populations ......................... . ........ 124 Relation of pod symptoms and seed symptoms ..... 127 Overwintering of R10, R10-86, and Ra in bean refuse .......... .... ........... .. ............ 130 DISCUSSION .......................................... 131 LITERATURE CITED .................................... 136 vi LIST OF TABLES Table Page 1. PART I Disease severity in leaves of greenhouse-grown kidney beans (cultivar Manitou) inoculated with wild-type Xanthomonas phaseoli var. fuscans (pr 16), and X. phaseoli (Xp 11). (Xp 21) compared to that produced by their respective rifampin—resistant mutants (R10, Ra and Rd) .......... . ................................. 27 Disease severity in leaves of greenhouse-grown kidney beans (cultivar Manitou) inoculated with wild-type Xanthomonas phaseoli var. fuscans (pr l6), and rifampin-resistant mutant (R10) compared to that produced by rifampin-strepto- mycin-resistant mutants (R10-82 and R10-86).... 29 Comparative growth of fungi, yeasts, rifampin- resistant isolate R10 and other (presumably nonpathogenic) bacteria from bean leaf tissue.. 35 Growth of phyllosphere bacteria on antibiotic- supplemented media.... ......................... 39 PART II Distribution of isolate R10 over Navy bean seedlings (cultivar Seafarer) grown from hilum spotted seeds.. ..... ...... ..... . ............... 66 PART III Relationship between the surface populations of R10 and Ra on Navy bean seed and the deveIOpment of blighted plants. ...... ................ ...... 120 Surface pOpulations of blight bacteria on mechanically threshed Navy bean seed ........... 121 vii Table Page 3. Frequency of surface blight contamination in commercial Navy bean seed lots... .............. 123 4. Effect of seed infection by R10 and Ra on seedling growth.... ............... ..... ........ 126 5. Detection of R10, R10-S6, and Ra in symptom- less Navy bean seeds...... ..................... 128 6. Relation between pod symptoms and seed symptoms ...... .... ......... ........... ......... 129 viii Figure LIST OF FIGURES Page PART I Growth of rifampin-resistant Xanthomonas phaseoli var. fuscans, R10, and wild type pr 16 on primary leaves of greenhouse- grown Navy (pea) beans (cultivar Seafarer). Fourteen-day-old plants were lightly sprayed to runoff with an aqueous suspension (5 x 107 cells/ml) of R10 or pr 16. The bacterial populations were sampled by vigorously shaking six leaves (average leaf area, 30 cm2) in 100 ml of phosphate buffer. Data are averages of three replications. The samples were plated on YCA or RAM ............. 32 Growth of rifampin-resistant Xanthomonas phaseoli, Ra and wild type Xp 11 in primary leaves of greenhouse-grown Navy (pea) beans (cultivar Seafarer). Thirteen—day-old plants were lightly sprayed to runoff with an aqueous suspension (5 x 107 cells/ml) of Ra and Xp 11. The bacterial populations were sampled by homogenizing six leaves (average leaf area, 30 cm2) in 75 m1 of phosphate buffer. Data are averages of three replications. The samples were plated on YCA or RAM ............. 34 PART II Population trends of pr isolate R10 and resistant bacteria and yeasts in first and second trifoliate leaves of field-grown Navy beans (cultivar Seafarer). Nineteen-day—old plants were Sprayed to runoff with an aqueous suspension (1 x 108 cells/ml) of R10. Bacterial populations were sampled by homogenizing 15 leaflets (average leaflet area, 20 cm ) in 75 ml of phosphate buffer. Samples were plated on RAM + cycloheximide, and YCA. Data are means of four replications 1 standard error.. ....... 57 ix Figure Page 2. POpulation trends of Xp isolate Ra in primary leaves of field-grown Navy beans (cultivar Seafarer). Eleven-day-old plants were sprayed to runoff with an aqueous suspension (1 x 108 cells/ml) of Ra. After abscission of primary leaves, groups of 21 brown and dry leaves on the soil surface were covered with a double layer of cheese cloth to aid in recovery. Populations of Ra were sampled by homogenizing 21 leaves (average area, 20 cm ) in 105 ml of phosphate buffer. Samples were plated on RAM (100 ug/ml rifampin) + cycloheximide (100 UQ/ml); PCNB (100 Ug/ml) was added to the medium for leaf samples from the ground. Data are means of two replications : standard error .............................. 59 Effect of temperature on the doubling times of R10 in primary leaves or first and second trifoliolate leaves of field-grown Navy beans (cultivar Seafarer). Five plots of beans were planted between June 5 and July 25, 1976 and inoculated with an aqueous suspension (1 x 108 cells/ml) of R10 when primary leaves or first and second trifoliate leaves were expanded. Four replications of 12 primary leaves or 15 trifoliolate leaflets (average area, 20 cm2) were homogenized in 75 ml of phosphate buffer and plated on RAM + cyclohex- imide. Doubling times were calculated for pOpulation changes during the exponential growth phase between each sample. Mean temperature is the average of daily maximum and minimum temperatures for days between sampling times . Temperatures were measured 200 meters away...... ............... . ........ 61 Effect of doubling times of R10 in primary leaves or first and second trifoliolate leaves of field-grown Navy beans (cultivar Seafarer) on the bacterial yield at first symptoms during the stationary growth phase. Five plots of beans were planted between June 5 and July 25, 1976, and inoculated with an aqueous suspension (1 x 109/m1) of R10 when primary leaves or first and second trifoliolate leaves were expanded. Four replicates of 12 primary leaves or 15 tri- foliolate leaflets (average area, 20 cm2) Figure Page were homogenized in 75 ml phosphate buffer and plated on RAM + cycloheximide. The mean doubling time was the average of the doubling times calculated between each sample during the exponential growth phase ........... 64 5. Population trends of isolate R10-S6 in Navy bean seedlings (cultivar Seafarer). Hilum- spotted seeds, infected with R10-S6, were planted and seedlings emerged after 12 days. Bacterial populations were determined by homogenizing the various parts of the seedling in a mortar and pestle with phosphate buffer and plating aliquots on RAM (100 ug/ml rifampin) + cycloheximide (100 ug/ml) + PCNB (100 ug/ml) + streptomycin sulfate (250 ug/ml). Primary leaf and first trifoliolate leaflet, average area 20 cm2; the stem, average weight 0.3 9, included the above ground portion of the hypocotyl and the epicotyl up to the terminal bud; below ground portion of the hypocotyl, average weight 0.1 9, later constituted portion of tap root; root mass, average weight 0.3 9, included all fibrous roots and adhering soil. Data are means of two replications with ten primary leaves, 15 trifoliolate leaflets, and five of other parts per replication ................... 68 6. Population trends of isolate R10 in the leaves and buds of field-grown Navy beans (cultivar Sanilac). Sixteen—day-old plants with expanding first trifoliolate leaves were inoculated with a suspension (1 x 108 cells/ml) of R10. Bacterial populations were individual- ly sampled from the primary to the seventh trifoliolate leaves of the main stem by homogenizing 14 primary leaves or 21 trifolio- late leaflets (average area, 20 cm2) in 105 ml phosphate buffer. Seven terminal buds from the main stem and 10-50 axillary buds from both main and lateral stems were homogenized with a mortar and pestle with 10 m1 of phOSphate buffer. Samples were plated on RAM (100 ug/ ml rifampin) + cycloheximide (50 ug/ml). Data are means of three replications. Symptoms were initially detected as leaf chlorosis; flower buds were first noted as clusters of xi Figure Page swollen buds at the growing tip; bloom represents the stage where all lower canopy flowers were open and some upper canopy flowers were closed; flat green pods indicate pods with no visible seed filling; 1—12 indicates the number of trifoliolate leaves expanded along the growing tip of the main stem .......................................... 71 7. POpulation trends of isolate Ra in the leaves and buds of field-grown Navy beans (cultivar Sanilac). Sixteen-day-old plants with expanding first trifoliolate leaves were inoculated with a suspension (1 x 108 cells/ml) of Ra. Bacterial populations were individually sampled from the primary leaves to the eighth trifoliolate leaves of the main stem by homogenizing 14 primary leaves or 21 trifoliolate leaflets (average area 20 cm2) in 105 ml of phosphate buffer. Seven terminal buds from the main stem and 10-50 axillary buds from both main and lateral stems were homogenized with a mortar and pestle with 10 ml of phos- phate buffer. Samples were plated on RAM (100 ug/ml rifampin) + cycloheximide (50 ug/ml). Data are means of three replications. Symptoms were initially detected as leaf chlorosis; flower buds were first noted as clusters of swollen buds at the growing tip; bloom represents the stage where all lower canopy flowers were open and some upper canopy flowers were closed; flat green pods indicates pods with no visible seed filling; 1-12 indicates the number of trifoliolate leaves expanded along the growing tip of the main stem .......................................... 73 8. Population trends of isolate R10 in Navy bean stems and roots (cultivar Seafarer). Twenty- day-old plants with third trifoliolate leaves just expanding were inoculated by jabbing the cotyledon scar with a syringe containing 1 x 108 cells R10 per m1. Bacterial populations were sampled from the roots up to the eighth trifoliolate leaf node of the main stem by homogenizing stem portions or roots from five plants with phosphate buffer and plating on RAM + cycloheximide. All data are based on an average node + internode fresh weight of xii Figure Page 0.35 g and are means of two replications. ROOT, included both the tap and the fibrous roots; COT, included the cotyledon node and internodal region from the soil line to the primary leaf node; PRI, included the primary node and the internodal region up to the first trifoliolate leaf node; lsT, included the first trifoliolate leaf node and the internodal region up to the second trifolio- late leaf node; 2ND-3RD, included the second and third trifoliolate leaf nodes and the internodal region up to the fourth trifoliolate leaf node; 4TH-5TH, included the fourth and fifth trifoliolate leaf node and the internodal region up to the sixth tri- foliolate leaf node; 6TH-7TH, included the sixth and seventh trifoliolate leaf node and the internodal region up to the eighth trifoliolate leaf node. Symptoms were noted as the first appearance of a redding of the node or internode; flower buds were first noted as a cluster of swollen buds at the growing tip; bloom represents the stage where all lower canopy flowers were open and some upper canopy flowers were closed; flat green pods indicate pods with no visible filling; 3-9 indicates the number of trifoliolate leaves expanded from the main stem ............ 78 9. POpulation trends of isolate Ra in Navy bean stems and roots (cultivar Seafarer). Twenty- day-old plants with third trifoliolate leaves just expanding were inoculated by jabbing the cotyledon scar with a syringe containing 1 x 108 cell R10 per ml. Bacterial populations were sampled from the roots up to the eighth trifoliolate leaf node of the main stem by homogenizing stem portions or roots from five plants with phosphate buffer and plating on RAM + cycloheximide. All data are based on an average node + internode fresh weight of 0.35 g and are means of two replications. ROOT included both the tap and the fibrous roots; COT included the cotyledon node and internodal region from the soil line to the primary leaf node; PRI included the primary node and the internodal region up to the first trifoliolate leaf node; lST included the first trifoliolate leaf node and the internodal xiii Figure Page region up to the second trifoliolate leaf node; 2ND-3RD included the second and third trifoliolate leaf nodes and the internodal region up to the fourth trifoliolate leaf node; 4TH-5TH included the fourth and fifth trifoliolate leaf node and the internodal region up to the sixth trifoliolate leaf node. Symptoms were noted as the first appearance of a redding of the node or internode; flower buds were first noted as a cluster of swollen buds at the growing tip; bloom represents the stage where all lower canopy flowers were Open and some upper canopy flowers were closed; flat green pods indicates pods with no visible filling; 3—9 indicates the number of trifoliolate leaves expanding from the main stem .................. 80 10. Comparison of the total R10 populations in field-grown primary leaves of Navy beans (cultivar Seafarer) and R10 populations readily removed by leaf washing. Eleven—day- old plants were sprayed to runoff with an aqueous suspension (1 x 108 cells/ml) of R10. Total bacterial pOpulations were sampled by homogenizing 12 leaves (average leaf area 20 cm2) in 75 m1 of phosphate buffer, pH 7.2; washed populations were sampled by gently washing 12 leaves in 100 ml of phosphate buffer for 2.5 minutes. Samples were plated on RAM + cycloheximide. Data are averages of four replications : standard error ................ 84 11. Comparison of total Ra populations in field- grown Navy beans (cultivar Seafarer) and rain trapped populations of leaf runoff water. Eleven-day-old plants were sprayed with an aqueous suspension (1 x 108 cells/ml) of Ra. Bacterial pOpulations were sampled by homo- genizing 14 primary leaves or 21 trifoliolate leaflets (average area, 20 cm2) in 105 ml 0.01 M phosphate buffer, pH 7.2. Total Ra pOpulations were determined by summing populations on primary through fourth trifoliolate leaves present at the time of rainfall. Rain trapped bacteria were collected in wax cups placed under single bean plants. All samples were plated on RAM (100 ug/ml) + cycloheximide (50 ug/ml). Total plant xiv Figure Page populations are averages of two replications : standard error and rain trapped pOpulations are averages of six replications + standard error ............................ T ............ 87 12. Effect of R10 and Ra on the severity of disease in the second through fifth trifoliolate leaves of the main stem from field-grown Navy beans (cultivar Sanilac). Beans were planted on 6/10/77 and inoculated 17 days later with an aqueous suspension of (l x 108 cells/ml) of R10 and Ra when first trifoliolate leaves were partially expanded. Twenty-one trifoliolate leaflets (average area, 20 cm2) from the same level in the canopy were homogenized in 105 m1 of 0.01 M phosphate buffer, pH 7.2, and plated on RAM (100 ug/ml rifampin) + cycloheximide (50 ug/ ml). Disease in each sample was determined by counting individual lesions on 21 leaves per replication .............................. 91 xv GENERAL INTRODUCTION AND LITERATURE REVIEW Common blight incited by Xanthomonas phaseoZi (E. F. Smith) Dowson (Xp) and fuscous blight incited by Xanthomonas phaseoli var. fuscans (Burkh.) Starr and Burkh. (pr) are important diseases of most commercial bean (PhaseoZus vulgaris L.) cultivars. Common blight was first reported by Beach (4) in 1892, about the same time, Halsted (27) reported a similar disease of bean pods and seeds. Erwin F. Smith in 1897 described the etiology of common blight and named the causal bacterium Bacillus phaseoli E. F. Smith (47, 48). Smith (49) further characterized the pathogen and transferred it into the genus Pseudomonas in 1901 then into the genus Bacterium in 1905 (50). Bergey's Manual ed II (5) renamed the blight bacteria Phytomonas phaseoli (E. F. Smith) Bergey et aZ.; Dowson (22) in 1938 placed the bacterium in the currently accepted genus of Xanthomonas. Fuscous blight was discovered by Burkholder (11) in 1924 on an unknown bean cultivar from Switzerland. Disease symptoms were identical to common blight and the only means of distinguishing the diseases was by culturing the causal bacterium. pr was initially known as Phytomonas phaseoZi var. fuscans but later it was placed in the currently recognized genus of Xanthomonas. Xp and pr are obligately aerobic, gram negative, straight rods, which produce a yellow nondiffusible pigment and are motile by a single polar flagellum. Both bacteria produce hydrogen sulfide, liquefy gelatin, proteolize milk, hydrolize starch and Tween 80 and produce an alkaline reaction in phenol red dextrose agar (7). Both bacteria are differentiated on the basis of a brown diffusible pigment produced by pr (2, 3, 28) and are otherwise physiologically and biochemically identicaL Bergey's Manual 8th ed. currently recognizes that Xp and pr are nomen species of Xanthomonas campestris (7). Nomen species terminology will be used in this paper. Xp and pr produce visible disease symptoms on all plant parts except the roots; symptomatology of the two diseases is essentially identical. However, Zaumeyer (58) reported that pr may cause a slight hypertrophy in tissue around a stem lesion and in seedlings, a darkening of the stem around the point of inoculation. Leaf symptoms are the most diagnostic feature of common blight or fuscous blight. Infection begins as minute water-soaked spots on the abaxial surface of the leaf with yellow discoloration opposite the spot on the adaxial side. Lesions then enlarge irregularly, dry out and become brown and brittle; occasionally a slight crust of bacterial exudate is present. The necrotic area is surrounded by a distinctive yellowish halo-like zone. Several lesions on a leaf may coalesce, and eventually occupy most of the leaf area and cause premature leaf drop (11, 57). Stem lesions begin as water-soaked dark green spots which eventually become dry, sunken and reddish brown in color. The lesion usually extends longitudinally up the stem but shows little downward movement. Stem symptoms are most common on seedlings and less evident as the plant matures (ll, 58). Pod symptoms appear as dark green water-soaked spots which later become dry and sunken. Drying begins at the outer edge of the lesion and extends inward; a yellow crustation of bacterial exudate may cover the lesion (ll). Pod infection occasionally results in seed infection. On white seeded bean cultivars, Xp and pr cause a shiny yellow spotting of the seed coat, the extent of which varies from small blotches to complete seed discolora- tion. Infection of dark-seeded bean cultivars is harder to detect due to seed coat pigmentation; symptoms appear as a darkening of the seed coat. Seeds which are infected early in develOpment may be completely shrivel- led or badly wrinkled, whereas, those infected near pod maturity are only darkened at the hilum (ll). Seed infection is most probable when the dorsal pod suture is infected. The bacteria invade the funiculus, pass through the raphe, and finally into the seed coat (55). Seeds may also be infected through the micropyle if the bacteria invade the pod cavity (56). Common and fuscous bacterial blights have been reported in all bean producing areas of the world (6, 19, 53, 58). The diseases often are major limiting factors in bean production; in lesser-developed countries losses are especially severe due to inadequate control practices. Common and fuscous blights historically have been important in the United States and especially in Michigan where approximately 35% of all dry edible beans and 85% of all U.S. Navy (pea) beans are produced. According to the USDA's disease surveys the two diseases were recognized as widespread and serious threats soon after the discovery of common blight in 1892 (4). In 1919 common blight was prevalent in 75% of New York's bean fields and caused serious damage; one year later, Colorado suffered a 40% crOp loss. The most severe loss in the nation due to common blight in 1921 was in Michigan where yields were reduced by 25%; in 1927 the average loss to bean blight throughout the U.S. was 1.4% (56). Andersen (l) estimated that bacterial blight caused a 3.5 million dollar loss in 1949-1950 to growers in three Michigan counties; in Nebraska 1953 losses were estimated at greater than $1,000,000 due to blight. In 1967 (33) at least 75% of Michigan's 650,000 acres of Navy beans were damaged by common and fuscous blights with yield reduction of 10-20%. In 1962, Sutton and Wallen (51) reported 60% of the Navy bean fields in Southwest Ontario, Canada infected with fuscous blight. Wallen and Jackson (52) estimated 4.60%, 6.60%, and 0.70% of the bean acreage in Ontario infected with the two blights and losses of 1.7%, 2.5%, and .25% in 1968, 1970, and 1972, respectively. Yield losses attributable to either Xp or pr are difficult to estimate due to the similarity of symptoms produced; damage resulting from halo blight may also alter the above estimates. The large volume of research conducted on common and fuscous blights reflects the historical importance of these diseases to the American bean industry. The early research conducted on common and fuscous blight, especially that by Burkholder and Zaumeyer (9, ll, 56, 57) focused on the basic aspects of the disease life cycle and symptomatology; their research provided a foundation of data for understanding the field ecology of Xp and pr. Detailed field studies which would have expanded Burkholder's and Zaumeyer's concepts and increased the understanding of how blight bacteria act in the field have never been conducted. Lack of such basic information has prevented research on other aspects of common and fuscous blights. Research to develop control measures for bean blight has received priority among bean researchers. The greatest emphasis has been on developing blight resistant or tolerant bean cultivars. Burkholder in 1924 (10) was the first to conduct extensive screening of bean germ— plasm for resistance to common blight; he concluded that none of the cultivars tested were immune, however, some differed in the disease severity. In 1946 Burkholder (12) reported varietal susceptibility trials to pr; all cultivars tested were susceptible except two, which were slightly resistant. Major programs for breeding resistance to common and fuscous blights in dry edible beans were initiated 10-15 years ago in many areas throughout the world (39). Coyne and Schuster (17) tested 1080 PI accessions, cultivars, and breeding lines of P. vulgaris for reaction to Xp; high tolerance was detected in some lines, and correlated with late maturity (16). Susceptibility or tolerance of beans to common blight also depends upon the stage of development of the plant (18). Using Great Northern Nebraska No. 1 selection 27 as a source of resistance to common and fuscous blights, Coyne and Schuster developed the tolerant Great Northern varieties Tara and Jules (14,15) Resistance to common and fuscous blights appears to be polygenic, quantitative, and highly heritable (16, 36). Pathogenic variation exists among Xp and pr isolates in their interaction with various bean cultivars and is considered an important factor in developing tolerant varieties. Pathogenic variation was first suggested by Smale (46) who detected differences in virulence in individual colonies of stock Xp cultures. Schuster and Coyne presented definitive evidence for variation among geographic isolates of Xp (43, 44, 45); Colombian and Ugandan isolates of Xp were more virulent on the tolerant Great Northern selection 27 than the standard Nebraska Xp isolate. Ekpo and Saettler (25,40) confirmed these findings and extended the existence of pathogenic variation to pr; generally, pr isolates were somewhat more virulent than Xp isolates. Much research has been directed toward maintaining bean seed stocks free of Xp and pr contamination and developing methods to detect seed-borne infection. Most bean producing states maintain seed certificatitxiprograms which oversee the quality of commercial seed. The Michigan seed certification program is administered by the Michigan Crop Improvement Association under authority delegated to it by the Michigan Department of Agriculture (34). The first step in certified seed production is to plant foundation seed supplied by the Michigan Foundation Seed Association; such seed is usually grown in the semi-arid or arid West where conditions are unfavorable for seed-borne diseases. Seed from fields which pass visual inSpection for the presence of common and fuscous blight symptoms, and which show no contamination in laboratory tests for Xp and pr can be sold as certified. By using certified seed a grower minimizes the probability of blight in his crop. Numerous assay methods have been develOped or adapt- ed to detect the presence of seed-borne Xp and pr. Katznelson (29) detected internally-borne Xp by measuring the increase in phage titer after incubation with seed. Schuster's leaf water-soaking method (42), designed to test resistance in beans to blight, has been adapted for use in seed testing. Saettler (38) developed the seed- ling injection technique for assaying Michigan seed lots for blight. Serological techniques are used by Guthrie (26) in Idaho to detect blight bacteria. None of the seed assaying methods have proven entirely satisfactory for detecting blight bacteria and all suffer from the inability to detect low levels of the blight bacteria. Numerous chemicals have been tested throughout the years for control of Xp and pr. Dimond and Stoddard (21) reported control of common blight of kidney bean in the greenhouse with several systemic compounds applied as soil drenches. Although streptomycin inhibited symptom expression of common blight of kidney beans in the greenhouse (31, 32); it has provided only limited blight control in the field (30, 37, 54). Copper-containing compounds have been the most widely studied chemicals for control of common and fuscous blights. Bordeaux mixture was tested over many years with variable results (8, 13, 24). More recent work with modern copper formulations and non-metallic organic bactericides has yielded promising results. In several studies, Dickens and Oshima (20, 35) reported excellent control of blight with applications of c0pper sulfate, c0pper ammonium carbonate and NEMA. Similar chemicals produced only marginal control of blight in tests by Saettler (37, 41), however, recently good control of common and fuscous blights was obtained by Weller and Saettler (54) with NEMA and copper hydroxide. The purpose of this study was to increase our understanding of how common and fuscous blights develop naturally in the field. Three objectives were involved in the overall study, and each will be presented separately. The first objective was to develop a selective system whereby Xp and pr could be selectively isolated and studied in the field. The second objective was to monitor Xp and pr population trends in field grown Navy (pea) beans and relate bacterial multiplica- tion and spread to the pattern of disease development. The third objective was to evaluate the relative 10 importance of different sources of primary inoculum in the establishment of bean blight in the field. GENERAL MATERIALS AND METHODS Bacterial storage and culture. Isolates of Xanthomonas phaseoli (Xp) and Xanthomonas phaseoZi var. fuscans (pr) used in this study were stored in 40% aqueous glycerol at -10 C and in pulverized diseased leaves at 4 C. Bacteria were grown on yeast extract calcium carbonate agar (YCA: 10 g yeast extract, 15 g agar, and 2.5 g calcium carbonate per 1000 ml glass distilled water), or in buffered yeast extract (BYE: 10 g yeast extract per 1000 ml 0.01 M phosphate buffer, pH 7.2). The bacteria were subcultured no more than five times past the stock culture in order to maintain genetic stability. All isolates were periodically assayed for changes in virulence by seedling injection technique (38). Bean culture. Navy (pea) beans (cultivars Seafarer, Sanilac, and Tuscola) and Kidney beans (cultivar Manitou) were grown in the greenhouse and the field. In the greenhouse the plants were grown in the standard greenhouse soil mix in 5-cm diameter clay pots. Day- light was supplemented with 14 hours of fluorescent 11 12 lighting from 0600 and 2100 hours. The temperature generally was maintained at 24-30 C and the plants were watered alternately with Rapid-Gro (1 teaspoon per 2 liters of water) and tap water. Field experiments were conducted at the Botany and Plant Pathology Research Farm, Michigan State University and the Saginaw Valley Bean and Beet Research Farm, Saginaw, Michigan. Beans were seeded using standard planting techniques or by hand with 28 inches between rows and 2 inches between plants in the row. Plant inoculation. Greenhouse-grown plants were inoculated with blight bacteria by one of three methods: (1) seedling injection (38), in which a bacterial suspension (108 cells per ml) was injected into the primary node of a lO-day—old kidney bean seedling (cultivar Manitou); (ii) leaf water-soaking (42), in which the undersurface of bean leaves was sprayed to a water-soaked appearance with a bacterial suspension (108 cells per ml) with a DeVilbiss sprayer attached to a compressed air line at a pressure of 1.2 kg per cm2; and (iii) leaf misting, in which a bacterial suspension (5 x 107 cells per ml) was lightly sprayed from a DiVilbiss sprayer to runoff on the leaf with no visible water soaking. Plants in the field were sprayed to runoff with a bacterial suspension (108 cells per ml) using a Knapsack sprayer with no visible watersoaking. 13 Disease rating. Inoculated plants were rated for disease symptoms by one of three methods depending on the type of inoculation. Plants inoculated by seedling injection were rated on a 0-3 scale: 0 = no symptoms; 1 = stem lesion and primary leaves partially collapsed; 2 = stem lesion and primary leaves completely collapsed; and 3 = stem lesion, primary leaves completely collapsed, and apical meristem dead. Leaves from greenhouse grown beans inoculated by water-soaking or leaf misting were rated on a 0-10 scale where 0 = no symptoms and 10 = total leaf necrosis. The amount of disease on leaves from field-grown plants was rated by counting lesions on individual leaflets. Bacterial isolation and identification. Bacteria were isolated by homogenizing plant tissue with 0.01 M phosphate buffer, pH 7.2, with mortar and pestle, glass tissue grinder, or Waring Blendor; sometimes plant tissue was shaken or soaked in buffer. Plant homogenates or washings were serially diluted in phosphate buffer and 0.1 ml portion of appropriate dilutions spread onto 48 hour-old plates of agar media. The number of colony forming units was used as an estimate of the number of bacteria in a sample. Yellow bacteria suspected as being Xp or pr were identified based on growth rate, odor and color on YCA, production of brown soluble pigment, production of hydrogen sulfide, and color 14 reaction in phenol red dextrose agar. Occasionally other standard tests for Xanthomonas spp. were employed (23). Seedling injection provided positive identification if the above physiological tests were inconclusive, however, such cases were rare. Xp and pr require 72 hours growth on YCA for the formation of distinct colonies, whereas, most yellow saprophytic, phyllosphere bacteria produce colonies in 24-48 hours. Phenol red dextrose agar was especially useful in screening large numbers of yellow isolates; all blight bacterial produce a characteristic red (alkaline) reaction whereas most yellow saprOphytes produce a yellow (acid) reaction. LITERATURE CITED ANDERSON, A.L. 1951. Observations on bean diseases in Michigan during 1949 and 1950. Plant Dis. Reptr. 35:89-90. BASU, P.K. 1974. Glucose inhibition of the char- acteristic melanoid pigment of Xanthomonas phaseoli var. fuscans. Can. J. Bot. 52:2203- 2206. BASU, P.K., and V. R. WALLEN. 1967. Factors affecting virulence and pigment production of Xanthomonas phaseoli var. fuscans. Can. J. Bot. 45:2367-2374. BEACH, S.A. 1892. Bean blight. New York [Geneva] Agr. Expt. Sta. Ann. Rept. 11:553-555. BERGEY, D.H., F.C. HARRISON, R.S. BREED, and others. 1925. Bergey's Manual of Determinative Bacteriology. 2nd ed. Baltimore, Md. 462 p. BOELEMA, B.H. 1967. Fuscous blight of beans in South Africa. South African J. Agr. Sci. 10: 1059-1063. BUCHANAN, R.E., and N.E. GIBBONS (eds.). 1974. Bergey's Manual of Determinative Bacteriology. 8th ed. Williams and Wilkins, Baltimore, Md. 1268 p. BURKE, D.W., and G.H. STARR. 1949. Direct measures used on control-tests of bacterial blight of beans (Abstr.). Colo-Wyo Acad. Sci. Jour. 3:43. BURKHOLDER, W.H. 1921. The bacterial blight ofthe bean; a systemic disease. Phytopathology 11: 61-69. 15 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 16 BURKHOLDER, W.H. 1924. Varietal susceptibility among beans to the bacterial blight. PhytOpathology 14:1-7. BURKHOLDER, W.H. 1930. The bacterial diseases of the bean: a comparative study. New York [Cornell] Agr. Expt. Sta. Mem. 127, pp. 88, illus. BURKHOLDER, W.H., and E.T. BULLARD. 1946. Varietal susceptibility of beans to Xanthomonas phaseoli var. fuscans. Plant Dis. Reptr. 30:446-448. CHRISTOW, A. 1934. Einege versuche uber die bakterienkrankheit bei bahnen. Phytopath. Ztschr 7:537-544. COYNE, D.P., and M.L. SCHUSTER. 1969. "Tara," a new Great Northern dry bean variety tolerant to common blight bacterial disease. Nebr. Agr. Expt. Sta. Bul. 506:1-10. COYNE, D.P., and M.L. SCHUSTER. 1970. "Jules," a Great Northern dry bean variety tolerant to common blight bacterium (Xanthomonas phaseoli). Plant Dis. Reptr. 54:557-559. COYNE, D.P., and M.L. SCHUSTER. 1973. PhaseoZus germ-plasm tolerant to common blight bacterium, Xanthomonas phaseoli. Plant Dis. Reptr. 57: 111-114. COYNE, D.P., M.L. SCHUSTER, and S. AL-YASIRI. 1963. Reaction studies of bean species and varieties to common blight and bacterial wilt. Plant Dis. Reptr. 47:534-537. COYNE, D.P., M.L. SCHUSTER, and K. HILL. 1973. Genetic control of reaction to common blight bacterium in bean (Phaseolus vngaris) as influenced by plant age and bacterial multiplication. Jour. Amer. Soc. Hort. Sci. 98:94-99. CRISPINA, A., and J. CAMPOS. 1976. Bean diseases of importance in Mexico in 1975. Plant Dis. Reptr. 60:534-535. DICKENS, L.E., and N. OSHIMA. 1969. Protective sprays inhibit secondary spread of common bacterial blight in snap beans. Plant Dis. Reptr. 53:647. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 17 DIMOND, A.E., and E.M. STODDARD. 1952. Chemo- therapeutic investigations on the common bacterial blight of beans. PhytOpathology 42: 72-76. DOWSON, W.J. 1939. On the systematic position and generic names of the gram negative bacterial plant pathogens. Zentbl. f Bakt. (etc.), Abt. II. 100:177-193. DYE, D.W. 1962. The inadequacy of the usual determinative tests for the identification of Xanthomonas spp. N.Z. J. Sci. 5:393-416. EDGERTON, C.W., and C.C.MORELAND. 1913. The bean blight and preservation and treatment of bean seed. La. Agri. Expt. Sta. Bul. 139; pp. 43. EKPO, E.J.A., and A.W. SAETTLER. 1976. Pathogenic variation in Xanthomonas phaseoli and X. phaseoli var. fuscans. Plant Dis. Reptr. 60:80-83. GUTHRIE, J.W., and H.S. FENWICK. 1967. Sheep- produced antibodies for use in detecting plant-pathogenic bacteria. Plant Dis. Reptr. 51:50. HALSTED, B.D. 1892. A bacterium of Phaseolus. N. J. Agr. Expt. Sta. Ann. Rpt. 13:330-333. HAYWARD, A.C., and J.M. WATERSTON. 1965. C.M.I. Description of pathogenic fungi and bacteria No. 48: Xanthomonas phaseoli; No. 49: Xanthomonas phaseoZi var. fuscans. KATZNELSON, H., and M.D. SUTTON. 1951. A rapid plaque count method for the detection of bacteria as applied to the demonstration of internally borne bacterial infections of seed. Jour. Bacteriol. 61:689-701. MARLATT, R.B. 1955. Effectiveness of streptomycin as a control for common bacterial blight of pinto beans. Plant Dis. Reptr. 39:213-214. MITCHELL, J.W., W.J. ZAUMEYER, and W.P. ANDERSON. 1952. Translocation of streptomycin in bean bacterial blight. Science 115:114-115. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 18 MITCHELL, J.W., W.J. ZAUMEYER, and W.H. PRESTON, JR. 1954. Absorption and translocation of streptomycin by bean and its effect on the halo and common blight organisms. Phytopathology 44: 25-30. M.S.U. Agri. Exp. Station. 1971. Focus on Michigans bean industry. Michigan Science in Action No. 16, 6 pp. M.S.U. Coop. Exten. Service. 1976. Seed certification in Michigan. Extension Bulletin E-1019, No. 112, 6. p. OSHIMA, N., and L.E. DICKENS. 1971. Effects of c0pper sprays on secondary spread of common bacterial blight of beans. Plant Dis. Reptr. 55:609-610. POMPEU, A.S., and L.V. CROWDER. 1972. Inheritance of resistance of Phaseolus vulgaris (dry beans) to Xanthomonas phaseoli Dows. (common blight). Cien. Cult. (S. Paulo) 24:1055-1063. SAETTLER, A.W. 1970. Fungicide and Nematicide Test Results of 1970. 26:56. SAETTLER, A.W. 1971. Seedling injection as an aid in identifying bean blight bacteria. Plant Dis. Reptr. 55:703-706. SAETTLER, A.W. 1977. Breeding dry edible beans (Phaseolus vulgaris L.) for tolerance to Xanthomonas bacterial blights. Fitopatologia Brasileira 2:179-186. SAETTLER, A.W., and E.J.A. EKPO. 1975. Pathogenic variation in Xanthomonas phaseoZi and X. phaseoli var. fuscans. Ann. Rept. Bean Imp. Coop. 18:67-70. SAETTLER, A.W., and H.S. POTTER. 1967. Chemical control of bacterial blights of dry field beans in Michigan by foliage sprays applied by ground and air equipment. Plant. Dis. Reptr. 51:622-625. SCHUSTER, M.L. 1955. A method for testing resistance of beans to bacterial blight. Phytopathology 45:519-520. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 19 SCHUSTER, M.L., and D.P. COYNE. 1971. New virulent strains of Xanthomonas phaseoli. Plant Dis. Reptr. 55:505-506. SCHUSTER, M.L., and D.P. COYNE. 1975. Genetic variation in bean bacterial pathogens. Euphytica 24:143-147. SCHUSTER, M.L., D.P. COYNE, and B. HOFF. 1973. Comparative virulence of Xanthomonas phaseoli strains from Uganda, Colombia, and Nebraska. Plant Dis. Reptr. 57:74-75. SMALE, B.C., and J.F. WORLEY. 1956. Evaluation of 2, 3, 5-tripheny1tetrazolium chloride for obtaining pathogenic types from stock cultures of halo blight and common blight organisms. Plant Dis. Reptr. 40:628. SMITH, E.F. 1897. Description of Bacillus phaseoli n. sp. Bot. Gaz. 24:192 (Abstr.). SMITH, E.F. 1898. Description of Bacillus phaseoli n. sp., with some remarks on related species. Proc. Amer. Assoc. Adv. Sci. Proc. 46:288-290. SMITH, E.F. 1901. The cultural characters of Pseudomonas hyacinthi, Ps. campestris, Ps. phaseoli, and P3. stewarti-four one flagellate yellow bacteria parasitic on plants. U.S. Dept. Agri. Div. Veg. Physiol. and Path. Bull. Bul. 28, 153 pp. SMITH, E.F. 1905. Bacteria in relation to plant diseases, Vol. 1. Carnegie Inst. Wash., ZBSEKL illus. Washington, D.C. SUTTON, M.D., and V.R. WALLEN. 1970. Epidemiologi- cal and ecological relations of Xanthomonas phasecli and X. phaeoli var. fuscans in south- western Ontario, 1961-1968. Can. J. Bot. 48: 1329-1334. WALLEN, V.R., and H.R. JACKSON. 1975. Model for yield loss determination of bacterial blight of field beans utilizing aerial infrared photography combined with field plot studies. Phytopathology 65:942-948. 53. 54. 55. 56. 57. 58. 20 WATSON, D.R.W. 1970. Bean common blight and fuscous blight in New Zealand. Plant Dis. Reptr. 54:1068-1072. WELLER, D.W., and A.W. SAETTLER. 1976. Chemical control of common and fuscous bacterial blights in Michigan Navy (pea) beans. Plant Dis. Reptr. 60:793-797. ZAUMEYER, W.J. 1929. Seed infection by Bacterium phaseoli. Phytopathology 19:96(Abstr.). ZAUMEYER, W.J. 1930. The bacterial blight of beans caused by Bacterium phaseoli. U.S. Dept. Agr. Tech. Bul. 186, 36 pp. ZAUMEYER, W.J. 1932. Comparative pathological histology of three bacterial diseases of bean. Jour. Agr. Res. 44:605-632. ZAUMEYER, W.J., and H.R. THOMAS. 1957. A mono- graphic study of bean diseases and methods for their control. U.S. Dept. Agri. Tech. Bull. 868, 255 p. PART I ISOLATION AND SCREENING OF RIFAMPIN- RESISTANT MUTANTS OF XANTHOMONAS PHASEOLI AND XANTHOMONAS PHASEOLI VAR. FUSCANS INTRODU CT ION The lack of suitable culture medium selective for Xp and pr has limited basic ecological studies of blight bacteria under field conditions. Kado's medium D5 (11) reported selective for Xanthomonas spp. and Schaad's (16) medium for Xanthomonas campestris are not suitable, due to the lack of specificity and low plating efficiency for Xp and pr. To isolate Xp and pr from field-grown bean tissue, dilution plating of homogenates on nutrient medium is necessary. Field-grown beans support a large, diverse microflora of bacteria, yeasts, and fungi; blight bacteria grow slowly and, therefore, are easily overgrown by faster-growing saprophytic bacteria on nutrient media. Only when disease symptoms are apparent is isolation of blight bacteria assured because of the high bacterial populations in the tissue. Indeed, in viva population studies of bean blight bacteria are impossible by this technique. Incorporation of antibiotic-resistance into Xp and pr could provide a system for the selective isolation and quantitation of blight bacteria from field grown 21 22 plants whether or not visible disease symptoms are apparent. Antibiotic-resistant mutants could then be inoculated in the field and reisolated by plating homogenates on antibiotic supplemented media. Antibiotic-resistance has been used in the study of some plant pathogenic bacteria but not extensively in quantitative field studies. Gowda and Goodman (5) and Lewis and Goodman (14) used Erwinia amylovora resistant to streptomycin in greenhouse studies of fire blight. Hsieh used streptomycin resistance in the study of the ecology and epidemiology of Xanthomonas oryzae (6, 7, 8). Hsu (9) studied mixed infection of Xanthomonas phaseoli, X. vesicatoria, and X. campestris in host and non-hosts plants with streptomycin-resistant mutants. Rifampin and neomycin resistant Erwinia rubrifaciens have been used in the study of deep bark canker of Persian walnut (4, 10). Rifampin resistance has recently been used in ecological studies of Agrobacterium tumefaciens (l, 15). This portion of the paper: (i) describes a system for selective isolation of Xp and pr based upon bacterial resistance to the antibiotic rifampin, (ii) compares some characteristics of rifampin-resistant Xp and pr mutants with those of their respective wild types, and (iii) documents the usefulness of rifampin mutants as tools for studying Xp and pr under field conditions. MATERIALS AND METHODS Rifampin agar medium. Rifampin (99.9% active) was obtained from Calbiochem, San Diego, CA 92112. To prepare rifampin agar medium (RAM) 10 mg of rifampin was dissolved in 0.4 ml methanol and diluted to 10 ml with distilled water. The solution was passed through a fritted-glass bacterial filter and added to 190 ml of autoclaved YCA. The amount of rifampin in RAM sometimes was increased from the normal up to 250 ug/ml. RAM was usually supplemented with filter-sterilized cycloheximide in distilled water at 25 ug/ml; at times the concentra- tion was increased up to 10 fold. RAM was occasionally supplementedvfiijipenta-chloro-nitro-benzene (PCNB) or streptomycin sulfate at 100 and 250 ug/ml, respectively. The PCNB was added prior to autoclaving the YCA; streptomycin sulfate was passed through a fritted glass bacterial filter. Isolation and culture of rifampin-resistant mutants Naturally-occurring rifampin-resistant mutants were selected from wild type isolates pr 16, Xp 11 (isolated from Michigan bean seed) and Xp 21 (isolated from 23 24 Colombian bean seed) by spreading 109 cells on RAM. These wild type isolates were selected because they were highly virulent relative to other isolates. The selection of mutant colonies which developed on RAM was based upon colony size, shape and vigor of growth. Naturally- occurring mutants resistant to both rifampin and streptomycin were selected from rifampin resistant isolates of pr by spreading 109 cells on YCA supplemented with 250 ug/ml streptomycin sulfate. The rifampin resistant isolates and the rifampin-streptomycin resistant isolates were subcultured twice on RAM and RAM + streptomycin sulfate, respectively, and then stored in 40% aqueous glycerol at -10 C. Physiological tests. Physiological tests were performed as described by Dye (3) with the following modifications: (1) pigment production by Xp and pr isolates was observed on YCA, (ii) lead acetate paper was used to detect hydrogen sulfide produced by the isolates on YCA, and (iii) slime production was determined on yeast extract dextrose calcium carbonate agar (YDC: 10 g yeast extract, 10 g dextrose, 15 g agar, and 2.5 9 calcium carbonate per 1000 ml water). Greenhouse and field inoculation. To compare virulence and growth of wild type and mutant isolates greenhouse-grown bean plants were inoculated by seedling injection, leaf water-soaking, or leaf misting. Six 25 circular spots (1 cm in diameter) were water-soaked and the symptoms developing outward were rated on successive days according to the scale previously described. A randomized block design was used for all experiments. In 1976 the virulence of isolates R10 and pr 16 were compared under field conditions; isolates Ra and Xpll were compared in 1977. pr and Xp isolates were inoculated to runoff to Navy beans (cultivar Seafarer) 49 and 45 days after planting, respectively; in both years plants were in the flat green pod stage. Individual treatment plots were three rows of beans four meters long and arranged in a randomized block design, with each isolate replicated three times. pr and Xp inoculated plants were rated 24 and 32 days after inoculation by counting the number of leaflets and pods bearing lesions on 12 plants in each replication. RESULTS Comparison of rifampin-resistant mutants of Xp and Kpf and the wild-type parents. The seedling injection test was used to test the virulence of each rifampin- resistant mutant relative to its wild type; each injection was replicated three times with three seedlings per replication. Two of five mutant isolates of pr 16, two of three of both Xp 11 and Xp 21 produced the same virulence rating as the wild types. One of the rifampin-resistant isolates of pr 16 (designated R10), one of Xp 11 (designated Ra), and one of Xp 21 (designa- ted Rd) were selected for further studies. All three isolates were resistant to greater than 500 ug/ml rifampin. Five mutant isolates resistant to rifampin and streptomycin sulfate were selected from a population of R10; two of the five double-marked mutants produced the same virulence rating as pr 16 and R10 in the seedling injection test. Isolates R10 and Ra produced as much disease as their wild types in greenhouse-grown kidney bean leaves, whereas, isolate Rd produced less disease relative to its wild parent (Table 1). One 26 27 TABLE 1. Disease severity in leaves of greenhouse-grown kidney beans (cultivar Manitou) inoculated with wild-type Xanthomonas phaseoli var. fuscans (pr 16), and X. phaseoli (Xp ll), (Xp 21) compared to that produced by their respective rifampin-resistant mutants (R10, Ra, and Rd) .a Post inoculation disease rating atb Isolate 14 days 20 days 25 days 34 days pr 16 R10 LSD (P = 0.05) 1.1 1.0 Xp 11 2.6 4.2 Ra 2.9 5.0 LSD (P= 0.05) 0.7 Xp 21 2.6 4.8 Rd LSD (P = 0.05) 0.7 1.0 aBeans were grown in lS-cm diameter clay pots, two plants per pot, and the trifoliate leaves were inoculated with an aqueous suspension of 108 bacteria per milliliter sprayed at a pressure of 1.2 kg/cmz. Isolates pr and Xp were used to inoculate 32 and 50-day-old plants, respectively. Disease rating scale: 0-10, with 0 = no symptoms and 10 = complete yellowing and necrosis. Data of pr and Xp inoculations are averages of three and four replications, respectively, of 12 leaflets each. 28 rifampin-streptomycin resistant mutant designated R10-86 produced as much disease as pr 16 and R10 in greenhouse- grown kidney bean leaves (Table 2); a second isolate designated R10-$2 produced more disease, however the symptoms were somewhat atypical. The virulence of R10 and Ra was compared to that of the wild types in field- grown Navy beans (cultivar Searfarer). Twenty-four days after inoculation with R10, 16.9% of the leaflets and 5.2% of the pods were visibly infected, whereas, 15.4% of the leaflets and 4.6% of the pods from plants inoculated with pr 16 were infected. Thirty-two days after inoculation with Ra, 37.1% of the leaflets and 10.1% of the pods were visibly infected, whereas 36.4% of the leaflets and 11.5% of the pods from plants inoculated with Xp 11 were infected. In no case were there significant (P = 0.05) differences in the amount of disease produced by rifampin-resistant mutants and their respective wild types. All Xp and pr rifampin- resistant mutants and double marked mutants retained cultural characteristics such as colony size, colony shape, slime formation, and yellow and brown pigment formation similar to that of their respective wild types. The responses of R10 and Ra to standard physiological tests were identical to their wild types. The in vitro doubling times of R10 and Ra were approximately 11% longer than those of the wild types 29 TABLE 2. Disease severity in leaves of greenhouse-grown kidney beans (cultivar Manitou) inoculated with wild type Xanthomonas phaseoli var. fuscans (pr 16), and rifampin- resistant mutant (R10) compared to that produced by rifampin-streptomycin-resistant mutants (R10-82 and R10-S6).a Post inoculation disease rating atb ISOIate 12 days 17 days pr 16 2.6 5.8 R10 2.9 5.3 R10-82 3.0 7.6 R10-S6 2.7 6.0 LSD (P = 0.05) 0.5 1.5 aBeans were grown in 15 cm diameter clay pots, two plants per pot, and the trifoliolate leaves were inoculated with an aqueous suspension of 108 bacteria per milliliter sprayed at a pressure of 1.2 kg/cmz. Plants were inoculated 33 days after planting. Disease rating scale: 0-10, with 0 = no symptoms and 10 = complete yellowing and necrosis. Data are averages of four replications. 30 in BYE shake culture at 25 C, however in viva growth rates of R10 and Ra in primary leaves of greenhouse-grown Navy beans (cultivar Seafarer) were identical to those of the wild types (Figures 1 and 2). Rifampin-resistance in R10 and Ra was stable; no revertants to rifampin sensitivity were detected by replica plating (13) after 16 and 11 consecutive trans- fers of R10 and Ra, respectively, on YCA; after each transfer ten plates with 100-150 colonies per plate were tested for revertants. Moreover, all pr and Xp bacteria isolated on YCA from 11 and 15 leaf lesions of field grown Navy beans 60 days after inoculation with R10 and Ra in separate plots were rifampin resistant. No revertants to rifampin-streptomycin sensitivity were detected after ten consecutive transfers of R10-86 on YCA or in R10-86 isolated from lesions of greenhouse- grown plants. Recovery of rifampin-resistant mutants of Xp and Kpf from field-grown beans. The growth of fungi, yeasts, and bacteria from symptomless leaves of field-grown Navy beans inoculated with R10 was compared on YCA, RAM, and RAM + cycloheximide. Three replicates of two leaflets each were homogenized in 10 ml of phosphate buffer and 0.1 ml of a dilution series was plated on 1 4 each media to yield 10- to 10- dilution of the homogenate (Table 3). Phyllosphere bacteria were 31 Figure 1. Growth of rifampin-resistant Xanthamanas phaseali var. fuscans, R10, and wild type pr 16 on primary leaves of greenhouse-grown Navy (pea) beans (cultivar Seafarer). Fourteen-day-old plants were lightly sprayed to runoff with an aqueous suspension (5 x 107 cells/m1) of R10 or pr 16. The bacterial populations were sampled by vigorously shaking six leaves (average leaf area, 30 cm2) in 100 ml of phosphate buffer. Data are averages of three replications. The samples were plated on YCA or RAM. 32 20.533002. smhm< VN «N ON 2. or z Np or m m> ZO 03¢: op umXollo ZO Owhfiuasov coon >>mz 039U 35 m o vloa o o MIOH o o Nloa o o HIGH mummm» oumcmmofiom cam + amass MO do» coflusaflo o mN o 0 ON 0 CNN 0 o moa o oomA o o o o oomA o o o mwuouomn III mummox mfluouomn dz cam + “mesh mz cam ooHwaocoHowo + Edm zmm ANmwflcoHoo ywsmmflu mama coon Eoum mfiuouomn Avacomosummco: Nanmssmoumv nonuo pom cam oumHOmH ucmumwmoulcfidsmwfin .mummo> .Hmcom mo nu3onm m>HpmummEoo .m mqm<9 36 completely inhibited on the RAM and the R10 were recovered. Colonies of phyllosphere bacteria prevented detection of R10 colonies on YCA except at the 10-4 dilutions of the homogenate; the number of R10 colonies l and 10'2 dilutions was 30% less than on RAM. At 10- fungi and yeasts overgrew RAM plates; the addition of cycloheximide reduced the growth of fungi and yeasts and allowed detection of R10 colonies. Cycloheximide was neither antagonistic to rifampin activity nor toxic to R10. Pure suspensions of R10 were plated on YCA, RAM, RAM + cycloheximide, and YCA + cycloheximide; there were no statistical differences in the number of colonies developing on each media. Results were similar when Ra or R10-S6 were substituted in the above tests. The usefulness of the rifampin-resistant mutants for field study of Xp and pr population dynamics was tested with R10. Nineteen-day-old field grown Navy bean plants (cultivar Seafarer) were sprayed to runoff with an aqueous suspension of R10 (5 x 107 cells per m1). Bacteria were isolated from first and second trifoliolate leaves homogenized with 75 ml of phosphate buffer; each sample was replicated four times. The following populations of R10 were detected per leaflet (average leaflet area, 20 cm2) l, 6, 11, and 17 days 4 6 7 after inoculation: 4.2 x 10 , 9.2 x 10 , 9.8 x 10 , and 2.8 x 108, respectively. 37 The efficiency of R10 recovery from leaf tissue was determined. Fifteen healthy leaves were homogenized with 75 m1 of phosphate buffer and 108 R10 cells in a Waring Blendor and serial dilutions of the homogenate were plated on the RAM. Controls consisted of 108 cells not homogenized but diluted and plated, and 108 cells shaken in 75 ml of phosphate buffer. The same number of R10 cells was recovered after blendor treatment as in the controls; thus mechanical damage or release of toxic substances by the leaf does not occur with the blendor method. RAM + cycloheximide was sufficient to inhibit the growth of all bacterial and most fungal residents of the phyllosphere. However, this media was not selective enough for rhiZOplane isolations of R10 or Ra because of the natural resistance in soil bacteria to these antibiotics. Isolate R10 was easily isolated from root samples only when the population of the bacteria was above approximately 102-103 bacteria per 0.3 9 root tissue; by using R10-$6, populations of the bacteria were detected at levels of near 101 per 0.3 9 root tissue when root homogenates were plated on RAM (250 ug/ml rifampin) supplemented with 250 ug/ml streptomycin sulfate, 250 ug/ml cycloheximide and 100 ug/ml PCNB. 38 Streptomycin-resistance has been used in some studies as the basis of an antibiotic-selective system for plant pathogenic bacteria. The toxicity of streptomycin and rifampin to bean phyllosphere bacteria was compared (Table 4). Rifampin at 50 ug/ml inhibited the growth of all bacteria, whereas, about 6% of the bacteria were resistant to streptomycin at the same concentration. 39 TABLE 4. Growth of phyllosphere bacteria on antibiotic- supplemented mediaq No. of phyllo- sphere bacteria Growth (%) of Antibiotic Concentration 3837cgegr§eit 22:32::3 on added (pg/ml) of bean leaf medium tissue None 0 37.7 x 105 100 Streptomycin 5 sulfate 50 2.19 X 10 5,8 Streptomycin 5 sulfate 100 1.66 x 10 4,4 Streptomycin 5 sulfate 250 0°51 X 10 1.3 Streptomycin 5 sulfate 500 0.42 x 10 1,1 Rifampin 50 0 0 a . . . Yeast extract-calc1um carbonate agar + cycloheximide at 25 ug/ml. Tissue was ground in 10 m1 of phosphate buffer in a glass tissue-grinder and plated after serial dilution. Colonies were counted after four days incubation at room temperature. DISCUSSION Rifampin was selected for use in an antibiotic- resistance selective system for bean blight bacteria because of its wide spectrum of antibacterial activity and high toxicity (2, 12, 18). No bacteria from the bean phyllosphere showed resistance to rifampin; natural resistance to other antibiotics, particularly strepto- mycin, is quite common. Moreover, rifampin has only limited use in human chemotherapy, thus there is little concern about long-term effects of any transfer of resistance from blight bacteria to other phyllosphere residents. We conclude for several reasons that R10 and Ra adequately model several important aspects of wild-type activity and they should behave similarly to pr l6 and Xp 11 under natural conditions. Multiplication and disease production of R10 and Ra were identical to that of the respective wild types in bean leaves. Isolate R10-86 is expected to act similar to the wild parents based upon the comparison of virulence in bean 4O 41 seedling and leaves. The mutation of R10, Ra, and R10-S6 was stable when the bacteria were grown in culture or bean leaves. Mutation stability assures the usefulness of these mutants throughout season-long studies. That the rifampin mutants of Xp and pr can be selectively isolated and their growth in field-grown Navy beans can be monitored over several weeks indicates their potential as tools for study of bean blight ecology. Use of the mutants should permit monitoring the sequence of seedling infection by Xp and pr originating from various sources of primary inocula, such as internally infected seed, externally infected seed, and infected plant refuse. Finally, use of R10 and Ra will permit a quantitative study of the build up and dispersal of secondary inoculum. The use of antibiotic-resistant mutants of plant pathogenic bacteria to enhance selective isolation is not a technique unique to this study. Mutants resistant to streptomycin, aureomycin,and rifampin have been described. However, none of the studies which have utilized mutants have presented data indicating extensive prescreening of the isolates relative to the wild parents. Gardner (4) reported that his rifampin-neomycin-resistant isolate of Erwinia rubrifaciens was "equivalent" in virulence to the wild type. Hsu (9) reported that population changes of streptomycin-resistant mutants of 42 Xanthamanas phaseali, X. vesicataria, and X. campestris were similar to the wild types in leaves of host plants. Hsieh (6) reported a streptomycin-resistant Xanthamanas aryzae "identical with its parent isolate in virulence and in 30 physiological and biochemical characters". The above reports provided only sparse evidence for the claims of wild type-mutant similarity and gave no details of the techniques involved in the screening procedure. The results of this study indicate a need for comprehensive screening of isolates, since reduction in virulence accompanied rifampin-resistance in some isolates. Further, the contrasting reaction of isolate Rd in seedlings and leaves suggest the need for a series of pathogenicity tests in the screening procedure. A change in wild type cell physiology or structure occurs when a bacterium becomes resistant to an anti- biotic. In the case of rifampin-resistance the B subunit of the DNA dependent RNA polymerase is altered so that the antibiotic cannot bind. Normally, changes in the wild type reduce the efficiency of the bacterium; with an altered RNA polymerase RNA transcription is less efficient and bacterial growth is reduced; other secondary changes might also occur. Small reductions in the in viva growth rate or virulence of a mutant may be unimportant over a short period of time but in a season long study such differences are greatly magnified. LITERATURE CITED ANDERSON, A.R., and L.W. MOORE. 1976. Survival of Agrabacterium in soil and on pea roots. Proc. Am. Phytopathol. Soc. 3:258 (Abstr.). CORCORAN, J.W., and F.E. HAHN. 1975. Antibiotics III: Mechanisms of action of antimicrobial and antitumor agents. Springer-Verlag, New York- Heidelberg. p. 252-268. DYE, D.W. 1962. The inadequacy of the usual determinative tests for the identification of Xanthamanas spp. N. Z. J. Sci. 5:393-416. GARDNER, J.M., and C.I. KADO. 1973. Evidence for the systemic movement of Erwinia rubrifaciens in Persian walnuts by the use of double-anti- biotic markers. Phytopathology 63:1085-1086. GOWDA, 8.8., and R.N. GOODMAN. 1970. Movement and persistence of Erwinia amylavara in shoot, stem, and root of apple. Plant Dis. Rep. 54: 576-580. HSIEH, S.P.Y., and I.W. BUDDENHAGEN. 1975. Survival of tropical Xanthamanas aryzae in relation to substrate, temperature, and humidity. Phytopathology 65:513-519. HSIEH, S.P.Y., I.W. BUDDENHAGEN, and H.E. KAUFFMAN. 1974. An improved method for detecting the presence of Xanthamanas aryzae in rice seed. Phytopathology 64:273-274. HSIEH, S.P.Y., and I.W. BUDDENHAGEN. 1974. Suppressing effects of Erwinia herbicala on infection by Xanthamanas aryzae and on symptom development in rice. PhytOpathology 64:1182- 1185. 43 10. 11. 12. 13. 14. 15. 16. 17. 18. 44 HSU, SHIH-TIEN, and R.S. DICKEY. 1972. Interaction between Xanthamanas phaseali, Xanthamanas vesicataria, Xanthamanas campestris, and Pseudamanas fluorescens in bean and tomato leaves. Phytopathology 62:1120-1126. KADO, C.I., and J.M. GARDNER. 1977. Transmission of deep bark canker of walnuts by the mechanical harvester. Plant Dis. Reptr. 61:321-325. KADO, C.I., and M.G. HESKETT. 1970. Selective media for isolation of Agrabacterium, Carynebacterium, Erwinia, Pseudamanas, Xanthamanas. Phytopathology 60:969-976. KUNIN, C.M., D. BRANDT, and H. WOOD. 1969. Bacterio- logical studies of rifampin, a new semisynthetic antibiotic. J. Infect. Dis. 119:132-137. LEDERBERG, J., and E.M. LEDERBERG. 1952. Replica plating and indirect selection of bacterial mutants. J. Bacteriol. 63:399-406. LEWIS, S.M., and R.N. GOODMAN. 1965. Mode of penetration and movement of fire blight bacte- ria in apple leaf and stem tissue. Phytopathology 55:719-723. MOORE, L.W. 1977. Prevention of crown gall on prune roots by bacterial antagonists. Phytopathology 67:139-144. SCHAAD, N.W., and W.C. WHITE. 1974. A selective medium for soil isolation and enumeration of Xanthamanas campestris. Phytopathology 64: 876-880. STALL, R.E., and A.A. COOK. 1966. Multiplication of Xanthamanas vesicataria and lesion develOpment in resistant and susceptible peppers. Phytopathology 56:1152-1154. WEHRLI, W., and M. STAEHELIN. 1971. Actions of the rifamycins. Bacteriol. Rev. 35:290-309. PART II POPULATION TRENDS AND DISTRIBUTION OF XANTHOMONAS PHASEOLI AND XANTHOMONAS PHASEOLI VAR. FWSCANS IN FIELD-GROWN NAVY BEANS (PHASEOLUS VULGARIS L.) INTRODU CTION The studies of Burkholder (6, 7) and Zaumeyer (49, 50, 51) form the foundation for present concepts about the disease cycles of bean common and fuscous blights. Infected seed is the main source of primary inocula; blight bacteria are located under the seed coat and do not enter the cotyledons until germination. Imbibition of water swells the seed resulting in a pulling apart of the epidermal cells of the cotyledon. Bacteria enter rifts in the epidermis and multiply in intercellular spaces; lesions eventually develop on the cotyledons. Initial seedling symptoms usually appear on the primary leaves which become infected while folded between the cotyledons (50). As lesions enlarge, bacterial exudate may accumulate on the surface; rain blown by wind splashes the bacteria to other parts of the plant or to uninfected plants, where new infections start under favorable conditions (52). Bacteria invade leaves through stomata or wounds, enter the intercellular spaces, cause a gradual dissolution of the middle 45 46 lamella and eventually host cells begin to disintegrate with the formation of bacterial pockets. Under proper conditions the bacterial mass will ooze from the lesion and become available for secondary spread, thus continuing the secondary infection cycle (50, 52). With the onset of the reproductive phase of the bean, pod and seed infection complete the disease life cycle. Blight bacteria may move systemically in an infected plant. Systemic movement of Xp was first noted by Barlow (2) in 1904 and later studied by Burkholder (6) and Barass (4) in 1921. Zaumeyer's histological studies (49, 50, 51) revealed the systemic pattern of Xp during early disease development. The bacteria may enter the stem through the stomata of the hypocotyl and epicotyl, through the vascular elements leading from the leaf and from infected cotyledons. Under favorable temperature and moisture conditions, the bacteria move rapidly upward in the stem. Drooping of one or both primary leaves is the first indication of systemic invasion. The bacteria concentrate in the pulvinus region and subsequent invasion of the parenchyma results in loss of tissue turgidity. If the bacteria spread extensively, apical meristem or buds in the primary axils may be killed. Leaf xylem eventually become invaded and lesions may occur when bacteria break through the vessels and invade surrounding parenchyma. 47 The bacteria may also break through the stem xylem and cause stem lesions. In seedlings, lesions usually occur at the cotyledon attachment or the primary leaf node and the stem becomes girdled and lodges. When the bacteria invade systemically, symptoms may not be detected until flowering or the plant may be slightly dwarfed. An important feature of systemic infection is the ability of the bacteria to move from stem xylem through the pod suture and into the seed with no trace of pod infection (6). Appearance of bacterial blight in bean fields is closely related to the stage of plant develonment. Blight symptoms sometimes are apparent during the seed- ling phase on cotyledons and primary leaves; however, during the vegetative phase, when the foliage is rapidly expanding, symptoms generally are not seen. Typical field symptoms only appear with the onset of the reproductive phase, which is initiated by flowering (6, 7, 16, 41). In Michigan, Navy bean fields appear blight- free until late July and early August at which time the fields suddenly become blighted. Because of this type of disease development, common and fuscous blights are considered to be late season diseases. To explain this phenomena, Gloyer (16) suggested that bean plants are not as blight susceptible during the vegetative stage as during the seedling and reproductive stages; more 48 recent studies have confirmed slightly greater susceptibility of bean plants during the reproductive stage of growth (8, l3). Burkholder suggested that late symptom expression is due to environmental conditions unfavorable for disease development during the first month and a half of the growing season (6, 7). Menzier (34) suggested the use of overhead irrigation on beans grown in hot, arid areas during vegetative growth, since blight did not appear to spread readily during that time. While the life histories of Xp and pr have received considerable qualitative study; the epidemiologies of common and fuscous blights are still poorly understood. The poor understanding of blight is especially apparent in the lack of a suitable explanation for the phenomenon of late symptom expression. The first studies of blight were based primarily upon field and histological observations. More recent research on blight develOpment has usually measured lesion formation but not considered bacterial populations. In viva studies of Xp and pr multiplication have been confined to controlled conditions mainly to study the physiological and genetic controls of resistance to the bacterial blights. A quantitative study of the population dynamics of Xp and pr under field conditions has never been conducted; such a study would contribute to an understanding of blight disease development under natural conditions. 49 Absence of selective media for Xp and pr has been responsible. The recent development of a selective system based upon rifampin-resistant mutants of Xp and pr now permits field study of blight bacteria (46). In this section: (i) the population dynamics of Xp and pr in field-grown Navy beans are described during seedling, vegetative, and early reproductive phases of plant growth, and (ii) the multiplication and spread of the bacteria are related to the pattern of disease development. MATERIALS AND METHODS Inoculation and isolation of R10, Ra, and R10-S6. Several rifampin-resistant blight isolates were studied in eight separate plots of Seafarer (planted 6/4/76, 6/21/76, 7/7/76, 7/13/76, 7/25/76, 7/19/77, 8/20/77, and 6/15/78), one plot of Sanilac (planted 6/10/77), and one plot of Tuscola (planted 6/22/77) Navy (pea) beans. All plots were located at the Botany and Plant Pathology Farm, East Lansing, MI. Generally, rifampin-resistant mutants were introduced into experimental plots via a Knapsack sprayer. Isolates R10 and R10-86 were also introduced by planting internally infected seed. The mutants were isolated from field-grown bean tissue by homogenizing leaves with 0.01 M phosphate buffer, pH 7.2, in a Waring blendor, mortar and pestle or by shaking the leaves in phosphate buffer; serial dilutions of the homogenate or washate were plated on RAM + cycloheximide occasionally supplemented with streptomycin sulfate or PCNB. Rifampin, cycloheximide, and PCNB in the media varied depending on the populations of resident bean microflora. Plating efficiencies of R10, Ra or R10-S6 were not altered by 50 51 such changes. Approximately 5 ml of phosphate buffer was used for each leaflet homogenized; no less than 75 ml of buffer was used for a sample. In 1976, samples consisted of 12 primary leaves or 15 trifoliolate leaves and were replicated four times. In 1977, samples con- sisted of 14 primary leaves, 21 or 42 trifoliolate leaflets and were replicated two to three times. Leaf populations of blight bacteria are expressed on the basis of 20 cm2 leaf tissue; this area is considered the average size of a trifoliolate leaflet or a primary leaf. To study bacterial blight spread within a Navy bean (cultivar Sanilac) canopy, ten-day-old seedlings possessing fully expanded primary leaves and half ex- panded first trifoliolate leaves were inoculated with isolates R10 and Ra; successive leaves of the main stem were subsequently assayed for the presence of the mutants until the mid-reproductive phase. During the vegetative phase, leaves expanded from the main stem at approximately two to three day intervals. Twenty- eight days after inoculation the plants were in bloom and by 33 days 1-5 cm long, flat, green pods were present; little vegetative expansion continued past bloom. Isolate Ra was assayed for up to the eight trifoliolate leaf until 13 days after bloom; isolate R10 was assayed for up to the seventh trifoliolate leaf until 15 days after bloom. 52 Multiplication and spread of R10 and Ra in Navy bean (cultivar Seafarer) stems were studied in 1978 by inoculating a bacterial suspension (1 x 108 cells/ml) into the cotyledon scar of 20-day-old plants with a syringe and subsequently assaying portions of the main stem. Each stem section consisted of a node and the internodal region up to the next higher node; the root portion consisted of both tap and lateral roots; all bacterial populations were based on a 0.35 9 average stem or root fresh weight. To isolate internally borne R10 and Ra, stems were washed in running distilled water for five minutes, soak- ed in 70% ethanol for five minutes, soaked in 50% bleach for ten minutes and rinsed in sterile distilled water before homogenizing. Effect of simulated washing and rain on blight bacterial populations on Navy bean leaves. Loss of blight bacteria from leaves due to rain water runoff was studied by several methods: 1) leaves from field- grown plants inoculated with R10 or Ra were gently washed inCLCH.M phosphate buffer for two minutes; 2) l3-day-old greenhouse-grown seedlings were inoculated with R10 and misted for one to two hours with a Herrmidifier model 500 humidifier. Water drippings from the primary leaves was collected in plastic petri plates. Water on leaf surfaces was removed by a two second dip 53 into a beaker of sterile distilled water. Total runoff water was considered the sum of the runoff and the leaf dip pOpulations. In the field, ll-day-old Navy bean seedlings (cultivar Seafarer) were inoculated with isolate Ra and runoff water was collected after each rainfall. Two or three wax cups (11.5 cm diam.) were placed around the base of a plant with the lip of the cup resting against the stem and the bottom anchored in the soil. Runoff water in the cups was assayed for bacteria within four hours after each rain. Total populations of Ra on all leaves usually were assayed no more than six hours prior to rainfall. Detection of surface-borne blight bacteria. Direct leaf prints were made by gently pressing leaves onto 48- hour-old plates of RAM + cycloheximide for one minute. Indirect leaf prints (40) were made by pressing a replica plater covered with parafilm (American Can Co.) onto a leaf and then to RAM + cycloheximide. Leaves were surface-sterilized by immersion and gentle agitation for 30 seconds in a 50% bleach solution or by UV -1 irridation [253.7 nm at 3.5 x 103 ergs sec-1cm for 20 minutes (3)]. Production of bacterial blight infected Navy bean seeds. Half-filled green pods were inoculated along the dorsal suture with a syringe containing an aqueous 54 suspension (108 cells/ml) of R10, Ra, or R10-S6. Mature pods were hand-harvested and visibly infected seeds collected. RESULTS Multiplication of R10 and Ra in leaves of field- grown beans. R10 and Ra populations were monitored in inoculated primary leaves or first and second trifolio- late leaves during 1976 and 1977. Population trends for the isolates resembled a standard bacterial growth curve with a one to three day lag phase and a six to nine day exponential growth phase (Figs. 1 and 2). The mean doubling time for R10 on Seafarer beans in five separate experiments in 1976 was 18.3 hours (range, 13.6-23.8 hours) and on Sanilac bean in one experiment in 1977 it was 20.1 hours. In 1977, the mean doubling time for Ra in one experiment on Sanilac beans was 19.8 hours and on Seafarer it was 17.7 hours. The R10 doubling times were negatively correlated with temperature (Fig. 3); higher mean temperatures resulted in greater bacterial multiplication. Populations of R10 and Ra peaked during the stationary phase and remained fairly stable until leaf abscission; a very slow death phase accompanied decomposition of infected leaves on the ground. In a study with Ra (Fig. 2), 13 and 23 days after abscission 55 I Ellil’l'llld I. 1' 56 Figure 1. Population trends of pr isolate R10 and resident bacteria and yeasts in first and second trifoliolate leaves of field-grown Navy beans (cultivar Seafarer). Nineteen-day-old plants were sprayed to runoff with an aqueous suspension (1 x 108 cells/m1) of R10. Bacterial populations were sampled by homogenizing 15 leaflets (average leaf— let area, 20 cm2) in 75 ml of phosphate buffer. Samples were plated on RAM + cycloheximide, and YCA. Data are means of four replications : standard error. 57 9 lllfilttilllilIIIIIl o——0RlO o—o RESIDENT BACTERIA A ANDYEASTS ' N2 - U 8- -I 5% SY PTOMS I- III —I IL 5 7- , .I I \ S 8 III 5 6 E: US O 2 O O 5 _, . \f JllillllLlLlll I J 468101214161820 DAYS AFTER INOCULATION I 1 4I 2 58 Figure 2. Population trends of Xp isolate Ra in primary leaves of field-grown Navy beans (cultivar Seafarer). Eleven-day-old plants were sprayed to runoff with an aqueous suspension (1 x 108 cells/ml) of Ra. After abscission of primary leaves, groups of 21 brown and dry leaves on the soil surface were covered with a double layer of cheese cloth to aid in recovery. Populations of Ra were sampled by homogenizing 21 leaves (average area, 20 cm2) in 105 ml of phosphate buffer. Samples were plated on RAM (100 ug/ml rifampin) + cycloheximide (100 ug/ml); PCNB (100 ug/ml) was added to the medium for leaf samples from the ground. Data are means of two replications 1 standard error. 59 ZO_h<..DUOZ_ ~5hm< m>._._<.._.m>O¢n "an zO_mm.Umm< ... < mEOhm<<>m u .0. "202.5200 “2m.— -h- _u- n In ‘0 (zwo 0:) :IV31 /vmaiova 'SON 901 h Q 60 Figure 3. Effect of temperature on the doubling times of R10 in primary leaves or first and second trifoliolated leaves of field-grown Navy beans (cultivar Seafarer). Five plots of beans were planted between June 5 and July 25, 1976 and inoculated with an aqueous suspension (1 x 108 cells/m1) of R10 when primary leaves or first and second trifoliolate leaves were expanded. Four replicates of 12 primary leaves of 15 trifoliolate leaflets (average area, 20 cm2) were homogenized in 75 ml of phosphate buffer and plated on RAM + cycloheximide. Doubling times were calculated for pOpulation changes during the exponential growth phase between each sample. Mean temperature is the average of daily maximum and minimum temperatures for days between sampling times. Temperatures were measured 200 meters away. BACTERIAL DOUBLING TIME (HOURS) 61 28 26- 24- 22- 20- 18" 16- 14- 12- 1O Y: 76.1- 2.9x r: 0.94 L I I I I 16 l 17 18 19 20 21 22 23 MEAN TEMPERATURE (c) 62 of inoculated primary leaves, 58% and 30% of the population detected at first symptoms were still viable; the leaves disintegrated after the last sample and prevented further sampling. In a similar experiment, 55% of the R10 population at first symptoms were viable 30 days after first trifoliolate leaf abscission. Symptom development on leaves inoculated with R10 or Ra required a minimal population of approximately 5 x 106—2 x 108 cells per 20 cm2 leaf tissue and usually corresponded to the early stationary growth phase. The yield of bacteria per infected leaflet was affected by the bacterial doubling time, the initial inoculum density, and the physiological age of the leaf. In 1976 the population of R10 accompanying first symptoms on primary leaves or first and second trifoliolate leaves was negatively, linearly related to the exponential phase doubling time (Fig. 4). Also, greater initial inoculum levels of Ra and R10 resulted in higher bacterial yields. Finally, leaves in the physiological stage of senescence were unable to sustain R10 and Ra populations in the exponential growth phase. Resident bacteria and yeasts associated with bean leaves. Populations of resident saprophytic bacteria and yeasts associated with bean leaves generally ranged between 104-107 cells per 20 cm2 leaf tissue. The 63 Figure 4. Effect of doubling times of R10 in primary leaves or first and second trifoliolate leaves of field-grown Navy beans (cultivar Seafarer) on the bacterial yield at first symptoms during the station- ary growth phase. Five plots of beans were plated between June 5 and July 25, 1976, and inoculated with an aqueous suspension (1 x lOB/ml) of R10 when primary leaves or first and second trifoliolate leaves were expanded. Four replicates of 12 primary leaves or 15 trifoliolate leaflets (average area, 20 cm2) were homogenized in 75 ml phosphate buffer and plated on RAM + cycloheximide. The mean doubling time was the average of the doubling times calculated between each sample during the exponential growth phase. 64 $30132: 62:38 .2553 MN ..N or N... mp MP . u u . . - . . . a O 00.0 r. xv—dluod— u» uuvan / vmanva 'SON 901 65 resident pOpulation generally was stable while blight bacteria were in the exponential growth phase but increased as blight bacteria entered stationary growth phase and symptoms appeared (Fig. l). Multiplication and spread of R10 and R10-36 in Navy bean seedlings. Hilum-spotted Navy bean seeds (cultivar Seafarer), containing an average of 106 R10 or R10-86 cells per seed, were used as inocula sources for studies of multiplication and spread of blight bacteria during the seedling stage. In a preliminary study, isolate R10 was found associated with all seedling parts (leaves, hypocotyl, cotyledons, terminal bud and roots) very soon after seedling emergence (Table 1). Bacterial populations varied among seedling parts, indicating that infected seedlings are not uniformly colonized; no symptoms were observed on these seedlings. In a second study, seeds infected with R10-S6 were used so that blight bacteria associated withtflmaroots could be sampled without interference from rifampin-resistant bacteria that exist naturally in the rhizosphere. Due to cool weather following planting, seedlings emerged in 12 days as Opposed to the usual five days and subsequent growth was also slowed. Bacteria of isolate R10-$6 were isolated from all plant parts except the cotyledons which were not sampled (Fig. 5). Bacterial populations 66 .AH2\mn ooHv mousnxmnoaoso + Aasxm: ooac :«m co cmumHm mum3 mumcmmoaon 0:» mo mcoflusaflp Hmflumm .ummmsn mumnmmonm Suez maumom 0cm Hmuuos m ca pcsoum mum3 mmamemm mammflu Had .mvan Hmcflfiuwu xflm no mommme uoou xflm no .mm>me unmeum NH Mo mawuooomxn xwm no .mCOpmazuoo NH mo onwamwmcoo comm nufl3 .cmme mnu mo Houuw pumpcmum wsu H mcofiumoflammu moms» mo momum>w msu mum0flpcw mmsHm> .OCHocmmxm mama mumaoHHOMHuu umuflm on» cuHB Swan EU o.¢ mum3 mmcwapmwm vHOI>M©ImCHC umw>mma >umEHum pwccmmxm nufi3 sown EU o.m who: mmcfiapwmm paelhmplxfim .m m.o unmflm3 mmmum>m .moaowuumm HflOm omumfloommm nuw3 momma» noon oopsHUCfi mmME uoou “NED om mono mmmum>m .mmwa xumEHum um m.o unmwm3 smoum mmmnm>m .muoou on» on was HmcwEku on» Eoum momma» pmwsHUCH Hauooflmw + Hmuooomwmo ONV ONV x . n x . . u . mOH Hm H + mOH ON N vOH x we m + VOH x vv 5 x . a x . x . u . mOH m5 H + mOH mm m mOH hm m + mOH X CM 5 x . II x . x O I. x . 00H VH m + @0H mm m mOH NN h + moH mo H x . 1 x . x . n x . MOH Nm N + mOH mm v HOH mc m + HOH Hw h m>mo m m>M© m oCflucmam Hmuwm mEHu powwowccw um mwumuown mo umnEsz a pan Hmcweume mmmE poom mood wumawum amuOOHmm + kuooomwm zoomawuoo gamma mcflaommm .mcmmm omuuomm ESHfln Scum czoum Ammummmmm um>fluasov mwcflaommm coma >>wz um>o cam muMHOmw mo :oHusnfluumflo .H mamas 67 Figure 5. Population trends of isolate R10—S6 in Navy bean seedlings (cultivar Seafarer). Hilum- spotted seeds, infected with R10-S6, were planted and seedlings emerged after 12 days. Bacterial populations were determined by homogenizing the various parts of the seedling in a mortar and pestle with phosphate buffer and plating aliquots on RAM (100 ug/ml rifampin) + cycloheximide (100 ug/ml) + PCNB (100 ug/ml) + streptomycin sulfate (250 pg/ml). Primary leaf and first trifoliolate leaflet, average area, 20 cm2 area; the stem, average weight 0.3 g, included the above ground portion of the hypocotyl and the epicotyl up to the terminal bud; below ground portion of the hypocotyl, average weight 0.19, later constituted portion of tap root; root mass, average weight 0.3 g, included all fibrous roots and adhering soil. Data are means of two replications with ten primary leaves, 15 trifoliolate leaflets, and five of other parts per replication. 68 7 TIII1WIIIIII ...—.EIRSI TRIFOLIOLATE LEAE HSTEM o—-o HYPOCOTYL(BELOW GROUND SYMPTOMS I- Lb—I‘ PRIMARY LEAF a: 6 'fi .9. 0 E 5 a r 1 I" ll] m E — 4- 9:. / '7 B 5 l 05 3 " O 2 8 .- 2 " '1 1 —I—I_LL.L.I_L.I_I_I_I_I_L_I_I_I_LI_I_I_I_I_I_I._) 11 13 15 17 19 21 23 25 27 29 31 33 35 DAYS AFTER PLANTING 69 were the greatest on the primary leaves, and the first trifoliolate leaf became colonized while expanding. The pOpulation of R10-86 on the above-ground portion of the hypocotyl and stem increased slightly and later declined. Below-ground seedling parts were colonized throughout the entire sampling period but declined slowly on the roots. Nearly the same number of R10-86 associated with root and hypocotyl tissue could be removed by washing as by homogenizing. Multiplication and spread of R10 and Ra in leaves and buds of field-grown beans during the vegetative and early reproductive stages. Population profiles of Ra and R10 in beans from seedling until early reproductive stages could be characterized as a series of bacterial growth curves displaced over time with each curve representing bacterial multiplication on individual leaves relative to the primary leaf node (Figs. 6 and 7). As each leaf differentiated from the main stem it became colonized by Ra or R10 and a gradient of bacterial pOpulations was established in the leaf canopy of the infected plants with the oldest leaves closest to the primary leaf node supporting the highest bacterial populations. The sequence of symptom development followed a similar gradient; symptom appeared first on the primary leaves and sequentially on the 70 Figure 6. Population trends of isolate R10 in the leaves and buds of field-grown Navy beans (cultivar Sanilac). Sixteen-day-old plants with expanding first trifoliolate leaves were inoculated with a suspension (1 x 108 cells/m1) of R10. Bacterial populations were individually sampled from the primary to the seventh trifoliolate leaves of the main stem by homogenizing 14 primary leaves or 21 trifoliolate leaflets (average area, 20 cm2) in 105 ml phosphate buffer. Seven terminal buds from the main stem and 10-50 axillary buds from both main and lateral stems were homogenized with a mortar and pestle with 10 ml of phosphate buffer. Samples were plated on RAM (100 ug/ml rifampin) + cycloheximide (50 ug/ml). Data are means of three replications. Symptoms were initially detected as leaf chlorosis; flower buds were first noted as clusters of swollen buds at the growing tip; bloom represents the stage where all lower canopy flowers were open and some upper canopy flowers were closed; flat green pods indicate pods with no visible seed filling; l-12 indicates the number of trifoliolate leaves expanded along the growing tip of the main stem. 71 w IrlelTlTTTl11’lllTl g o—oTERMINAL O—OAXILLAY \ R s 8 .. “.12 d .— 8 . n “51 21- - § 0 lllJLlll ___.-_*,li-_.441| 8 IIIIIIlIrlIrTIIEIIFIIIfiIIII1ITIW7 0—-0 PRIMARY 0—0 151’ 0—0 2ND&3RD H 4'". 0—0 5TH b—fi 6TH H 71’" ‘1 I lOG NOS. ucrERIA/LEAELEflzo CM21 0 I 5 s-vaproMs rn=nowsR mm ““0051 ‘ FGP-FLAT GREEN P008 4 _ 1-12=NO. or LEAVES EXPANDED - q 1 4 Fl 10512 I [in1114;:1;];111311111111111111242 4 810121416182022242628303234’43 DAYS AFTER INOCULATION 72 Figure 7. Population trends of isolate Ra in the leaves and buds of field-grown Navy beans (cultivar Sanilac). Sixteen-day-old plants with expanding first trifoliolate leaves were inoculated with a suspension (1 x 108 cells/ml) of Ra. Bacterial pOpulations were individually sampled from the prima- ry leaves to the eighth trifoliolate leaves of the main stem by homogenizing 14 primary leaves or 21 trifoliolate leaflets (average area 20 cm2) in 105nm. of phosphate buffer. Seven terminal buds from the main stem and 10-50 axillary buds from both main and lateral stems were homogenized with a mortar and pestle with 10 m1 of phosphate buffer. Samples were plated on RAM (100 ug/ml rifampin) + cycloheximide (50 ug/ml). Data are means of three replications. Symptoms were initially detected as leaf chlorosis; flower buds were first noted as clusters of swollen buds at the growing tip; bloom represents the stage where all lower canOpy flowers were open and some upper canopy flowers were closed; flat green pods indicates pods with no visible seed filling; 1-12 indicates the number of trifoliolate leaves expanded along the growing tip of the main stem. LOG nos. aACtERIA/auD m 0 0 d N U Q LOG nos. BACTERIA/LEAFLETIZO CM2) 73 TlTTTlTYfTIIITlTrjjTTTT O—OTERMINAL o——o AXILLARY IIILLJ_llllll UTrTIIIITrIIrTTTTIIIITjIIjIIITTII o—OPRIMARY o——OIST o—o 2ND&3RD H4111 I—OSIH b—obTH H7111 h—EOTH S3 SYMPTOMS "8 FLOWER BIDS ISBLOOM FGPsFLAT GREEN POD‘ I-128NO. OF LEAVES EXPANDED 1012 141618 20 22 24 26 28 3o 32 ’ DAYS AnER mocuumon 1 .1..1.1.1.1...¥....3°1”..fsf‘pi’tg 74 oldest to the youngest trifoliolate leaves. Symptom expression on each leaf corresponded to the period of stationary phase bacterial growth and required a minimal bacterial population as previously described. The same sequence of symptom expression described for the main stem was also associated with leaves of lateral stems. As pods developed they were colonized similar to expanding leaves. By bloom, symptoms in the R10 and Ra inoculated plants were detected only up to the fourth and sixth trifoliolate leaves, respectively, even though bacteria were detected by plate counts or leaf prints up to the top trifoliolate leaf. Thus, the bacteria were spreading more rapidly upward and outward into the leaf canopy than the symptoms indicated. Generally, isolate Ra colonized the plants more rapidly, produced higher stationary phase populations and produced more symptoms than isolate R10. Isolates R10 and Ra were consistently detected in terminal and axillary buds throughout the monitoring period (Figs. 6 and 7). Bacteria in either bud type displayed a considerable variation in population between samples which ranged from near zero to almost 103 bacteria per bud. Newly expanded leaves less than 1 cm long consistently were colonized by R10 and Ra. Flower buds were consistently colonized with Ra and R10 75 levels comparable to the vegetative buds. Up to l x 104 bacteria were detected in both unopened and opened flowers from plants used in this study and those from 1976, however, the populations were generally 101-2x102 bacteria per flower. Field observations of disease development. Overhead visual ratings of plants inoculated with R10 and Ra (Figs. 6 and 7) were taken periodically in 1977. These observations simulated what a grower inspecting his field or a researcher inspecting experimental plots might observe. Typical field symptoms were not detected by overhead observation taken throughout the entire vegetative stage. Only when the outer leaf canopy was cut away could the lower leaves with typical symptoms be seen. The rapidly expanding outer foliage continued to camouflage blight symptoms in the leaf canopy until ten days after bloom when symptoms developed on the outer trifoliate leaves. This pattern of symptom appearance was identical to that normally seen in commercial fields. Multiplication and systemic spread of R10 and Ra in stems of fieldegrown Navy beans. Ten days after inoculation isolates R10 and Ra initially were detected internally in stems of Navy beans which were used for the study shown in Figures 6 and 7. A steady increase in the number of plants systemically colonized was 76 observed. Thirteen, 19, 26, and 31 days after inocula— tion 20, 64, 71, and 85% of the stems were infected with R10; at 12, 21, 28, and 31 days after inoculation 19, 43, 62, and 80% of the stems were infected with Ra. Studies in 1976 indicated that 75% of Seafarer plants were systemically infected with R10. Tuscola plants were infected to a similar degree. Isolates R10 and Ra, which were introduced into the cotyledon node to study systemic colonization, moved rapidly upward in the main stem and established a pop- ulation gradient inversely related to the distance from the cotyledon node. The population profile of R10 and Ra in the main stem during the vegetative and repro- ductive stages was characterized by a series of growth curves with each representing bacterial multiplication in a section of stem relative to the cotyledon node (Figs. 8 and 9). Similar patterns of bacterial multiplication and movement was detected in lateral stems and leaf petioles. The average doubling time of R10 and Ra in stem tissue up to the fourth trifoliate node was 22.8 hours (range 14.6-37.6 hours) and 23.8 hours (range 12.3-31.0 hours), respectively. Stem lesion formation required minimal populations of R10 and Ra of approximately 107-108 bacteria per 0.35 g stem tissue; symptoms appeared first at the injection 77 Figure 8. Population trends of isolate R10 in Navy bean stems and roots (cultivar Seafarer). Twenty-day-old plants with third trifoliate leaves just expanding were inoculated by jab- bing the cotyledon scar with a syringe containing 1 x 108 cells R10 per ml. Bacterial populations were sampled from the roots up to the eighth trifoliate leaf node of the main stem by homogenizing stem portions or roots from five plants with phosphate buffer and plating on RAM + cycloheximide. All data are based on an average node + internode fresh weight of 0.35 g and are means of two replications. ROOT, included both the tap and the fibrous roots; COT, included the cotyledon node and internodal region from the soil line to the primary leaf node; PRI, included the primary node and the internodal region up to the first trifoliolate leaf node; lST, included the first trifoliolate leaf node and the internodal region up to the second trifoliolate leaf node; 2ND-3RD, included the second and third trifoliolate leaf nodes and the internodal region up to the fourth trifoliolate leaf node; 4TH-5TH, included the fourth and fifth trifoliolate leaf node and the internodal region up to the sixth trifoliolate leaf node; 6TH-7TH, included the sixth and seventh trifoliolate leaf node and the internodal region up to the eighth trifoliolate leaf node. Symptoms were noted as the first appearance of a redding of the node or internode; flower buds were first noted as a cluster of swollen buds at the growing tip; bloom represents the stage where all lower canopy flowers were open and some upper canopy flowers were closed; flat green pods indicate pods with no visible filling; 3-9 indicates the number of trifoliolate leaves expanded from the main stem. 78 ZO:.<._=UOZ. umhu< m>