INFORMATION TO USERS This reproduction was made from a copy o f a docum ent sent to us for microfilming. While the most advanced technology has been used to photograph and reproduce this docum ent, the quality of the reproduction is heavily dependent upon the quality o f the material subm itted. The following explanation o f techniques is provided to help clarify markings or notations which may appear on this reproduction. 1.The sign or “ target” for pages apparently lacking from the docum ent photographed is “ Missing Page(s)” . If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure complete continuity. 2. When an image on the film is obliterated with a round black m ark, it is an indication o f either blurred copy because of movement during exposure, duplicate copy, or copyrighted materials that should not have been filmed. For blurred pages, a good image o f the page can be found in the adjacent frame. If copyrighted materials were deleted, a target note will appear listing the pages in the adjacent frame. — 3. When a map, drawing or chart, etc., is part o f the material being photographed, a definite m ethod o f “ sectioning” the material has been followed. It is customary to begin filming at the upper left hand com er o f a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again-beginning below the first row and continuing on until complete. 4. For illustrations that cannot be satisfactorily reproduced by xerographic means, photographic prints can be purchased at additional cost and inserted into your xerographic copy. These prints are available upon request from the Dissertations Customer Services D epartm ent. 5. Some pages in any docum ent may have indistinct print. In all cases the best available copy has been filmed. University Microfilms International 300 N. Zeeb Road Ann Arbor, Ml 48106 8315490 Olson, Brian Douglas EPIDEMIOLOGY AND CONTROL OF BACTERIAL CANKER ON MONTMORENCY SOUR CHERRY CAUSED BY PSEUDOMONAS SYRINGAE PV. MORSPRUNORUM Ph.D. Michigan State University University Microfilms International 300 N. Zeeb Road. Ann Aitoor, MI 48106 1983 PLEASE NOTE: In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this docum ent have been identified here with a check mark V . 1. Glossy photographs or p a g e s______ 2. Colored illustrations, paper or print_____ 3. Photographs with dark background ^ 4. Illustrations are poor copy______ 5. Pages with black marks, not original copy______ 6. Print shows through as there is text on both sides of page______ 7. Indistinct, broken or small print on several pages 8. Print exceeds margin requirem ents_____ 9. Tightly bound copy with print lost in spine______ 10. Computer printout pages with indistinct print______ 11. Page(s)___________ lacking when material received, and not available from school or author. 12. P age(s)___________ seem to be missing in numbering only a s text follows. 13. Two pages num bered____________. Text follows. 14. Curling and wrinkled p ag es______ 15. Other_____________________________________________________________________ University Microfilms International EPIDEMIOLOGY AND CONTROL OF BACTERIAL CANKER ON MONTMORENCY SOUR CHERRY CAUSED BY PSEUDOMONAS SYRINGAE PV. MORSPRUNORUM By Brian Douglas Olson 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 1983 ABSTRACT EPIDEMIOLOGY AND CONTROL OF BACTERIAL CANKER ON MONTMORENCY SOUR CHERRY CAUSED BY PSEUDOMONAS SYRINGAE PV. MORSPRUNORUM By Brian Douglas Olson Rifampicin-resistant strains of Pseudomonas syringae pv. morsprunorum and P^. syringae pv. syringae (PsmR and PssR, respectively) were used in epidemiological and chemical control studies on bacterial canker of Prunus cerasus L. cv. Montmorency (sour cherry) in East Lansing, Michigan. Rainwater was collected under a cherry tree with sequential rain samplers designed to study the dissemination of waterborne pathogens from trees. collected rain from a 5,026-cm for each 0.5-mm rainfall. 2 The apparatus area and saved 5- to 8-ml subsamples Two dispersal patterns of PsmR were observed from sampled rain periods. In the first pattern, populations of PsmR in rainwater increased and leveled off; while in the second, populations increased to the highest level in the first 2.5-mm rain., declined, and then leveled off. The first and second patterns were preceded by 5 days with and without rain, respectively. In 1980 and 1981, populations of PsmR in rainwater were higher in spring and autumn than in summer, whereas populations of PsmR associated with 3 4 2 leaves were relatively constant (10 -10 colony-forming units/cm leaf). Ground cover plants in an orchard were inoculated with PsmR and PssR in November 1979 and 1980. Both strains were recovered from ground cover through the winter and in early spring but not in late spring and summer. In spring of 1980 and 1981, PsmR and PssR were recovered from buds of fruit spurs following inoculation of leaf scars the preceding autumn. Recovery of PsmR and PssR from buds following leaf scar inoculations of fruit spurs in August 1980 were lower than for buds on spurs inoculated in October. Copper treatments were evaluated for reducing populations of PsmR on Montmorency sour cherry trees in spring and early summer 1980 and 1981. Populations of PsmR were reduced more by 636 and 949 mg/L tribasic copper sulfate (TBS) than by 200 mg/L Citcop 4E, but several applications were needed to reduce the populations to a low level. Citcop 4E at 200 mg/L was more phytotoxic to cherry foliage than 636 mg/L TBS. Phytotoxicity was related to the number of applications and was not reduced by adding hydrated lime to copper treatments. The decline of copper residues from leaves was related by multiple regression analyses with rainfall and initial level of copper on leaves. To my father Miles Beardsley Olson ii ACKNOWLEDGMENTS I would like to thank Dr. Alan L. Jones for his patience and help in reviewing this manuscript and for his financial support of my Ph.D. project. I would also like to thank my other committee members, Drs. A. W.„Saettler, D. W. Fulbright, R. L. Perry, and R. L. Andersen, for their assistance. In addition, I wish to thank Dr. E. J. Klos and the Department of Botany and Plant Pathology for their financial support when it was needed. I thank Beth Martin, Steve Tonasager, and Dave Maltby of the Analytical Toxicology Section of the Animal Health Diagnostic Laboratory for making the elemental analyses and Scott Rickettson, Michael Allen, and Jerry Quinn for their technical assistance. Finally, I would like to express to my wife, Val, appreciation for her patience and moral support. TABLE OF CONTENTS Page GENERAL INTRODUCTION AND ....................................... Literature Cited ............................................ 1 11 CHAPTER 1 - A Sequential Sampler for Monitoring Water-Disseminated Pathogens from Trees ...................... 18 A b s t r a c t ....................................................... 18 Introduction ................................................ 18 Materials and Methods 19 ..................................... Instrument design and operation ............... Evaluation of instrument ... 19 ............................ 22 Results and Discussion . . . . . . ........................ 23 Literature Cited ............................................ 27 F i g u r e s ....................................................... 28 CHAPTER 2 - Dissemination in Rainwater and Overwintering of Pseudomonas syringae p v . morsprunorum and . syringae p v . syringae in a Montmorency Sour Cherry Orchard ............... 32 A b s t r a c t ....................................................... 32 Introduction ................................................ 33 Materials and Methods 34 ..................................... Rifampicin-resistant strains ........................ 34 Overwintering ......................................... 35 Monitoring bacterial populations 37 iv .................... Environmental monitoring ............................ 38 Data a n a l y s e s ............................................ 38 R e s u l t s ....................................................... 39 Overwintering on ground cover ........................ 39 Overwintering in leaf s c a r s ............................. 39 Populations associated with leaves Populations in rainwater ................. 39 ............................ 40 Dispersal patterns of bacteria in rainwater ......... 41 D i s c u s s i o n ..................................................... 42 Literature Cited ........................................... 45 Tables and Figures . . . . . 47 ............................... CHAPTER 3 - Reduction of Pseudomonas syringae pv. morsprunorum on Montmorency Sour Cherry with Copper and Dynamics of the Copper Residues ................................................ 52 A b s t r a c t ....................................................... 52 Introduction ................................................ 52 Materials and Methods 53 ..................................... Field i n o c u l a t i o n s ..................................... 53 Spray t r i a l s ............................................54 Monitoring bacterial populations .................... Monitoring copper deposits on leaves ............... 55 57 General data a n a l y s e s ................................... 59 R e s u l t s ....................................................... 59 Copper phytotoxicity ................................. Recovery of bacteria from buffered copper solutions Effect of copper on bacterial populations v 59 . 60 ........... 60 Dynamics of copper deposition ........................ 61 D i s c u s s i o n ..................................................... 63 Literature Cited ............................................ 66 Tables and F i g u r e s ............................................ 69 APPENDIX A - Susceptibility of Prunus Rootstock Cultivars to Pseudomonas syringae pv. syringae and syringae pv. m o r s p r u n o r u m .................................................77 A b s t r a c t ....................................................... 77 Introduction ................................................ 77 Materials and Methods ..................................... 78 Results and Discussion ..................................... 79 Literature Cited ............................................ 82 T a b l e ......................................................... 84 APPENDIX B - Residual Plots for Regression Equations of Citcop 4E and Tribasic Copper Sulfate ........................ vi 85 LIST OF TABLES Page CHAPTER 2 Table 2. L Recovery of rifampicin-resistant Pseudomonas syringae p v . morsprunorum (PsmR) and £. syringae pv. syringae (PssR) from plants in the orchard ground cover following inoculation in the autumn of 1979 and 1980 ........................................ Table 2. I 47 Percent recovery of rifampicin-resistant Pseudomonas syringae p v . morsprunorum (PsmR) and P^. syringae pv. syringae (PssR) from buds on fruit spurs of Montmorency sour cherry in spring following leaf scar inoculations the Table 2. ) previous autumn ............. 48 Populations of rifampicin-resistant Pseudomonas syringae pv. morsprunorum on individual leaves taken from terminal shoots of Montmorency sour cherry trees in 1 9 8 1 .......................................... 49 CHAPTER 3 Table 3. L Effect of copper treatments on populations of rifampicin-resistant Pseudomonas syringae vii pv. morsprunorum (PsmR) on Montmorency sour cherry leaves and the analysis of variance for a set of planned paired comparisons in 1980 Table 3. ..................................... Effect of copper treatments on populations of rifampicin-resistant Pseudomonas syringae pv. morsprunorum (PsmR) on Montmorency sour cherry leaves and the analysis of variance for a set of planned paired comparisons in 1981 Table 3. ..................................... Recovery of rifampicin-resistant Pseudomonas syringae p v . morsprunorum (PsmR) from distilled water and phosphate buffer each containing tribasic copper sulfate (TBS) + hydrated lime (lime) . . . . Table 3. Defoliation of Montmorency sour cherry trees from various copper treatments and the analysis of variance for a set of planned paired comparisons in 1981 ......... Table 3. The relationship of the number of copper sprays applied to Montmorency sour cherry leaves and residues of copper on the leaves to phytotoxicity ................... APPENDIX A Table 4.1 Canker lengths on Prunus rootstock cultivars following wound inoculations with Pseudomonas syringae pv. morsprunorum and ]?. syringae pv. syringae ..................................... LIST OF FIGURES Page CHAPTER 1 Figure 1. 1 Sequential rain sampler. A, funnel; B, cover; C, platform with three leveling bolts; and D, c o l l a r .............. . ............ 28 Figure 1. 2 Mechanism for the sequential rain sampler. A, tipping bucket; B, bucket bearing; C, drain; D, plexiglass box and siphon; E, collection tube rack and tubes; F, turntable; G, pivoting arm; H, advancing block; I, electronic switch; J, stationary stop piece; K, drain tube; L, level; M, plexiglass arm; and N, counterweights Figure X .3 . . . . . . . . . . . . . . Bacterial frequency (%) in collection tubes where sufficient quantities of bacteria in suspension were added to the sampler for two bucket tips, followed by sufficient quantities of phosphate buffer for six bucket tips. Prior to sampling, the sequential rain samplers were thoroughly rinsed with phosphate buffer. x A, Empirical 29 data were the means of three replicate experiments from three samplers. B, Predicted values were from the sampler retention equation ............................ Figure 1.4 30 Concentrations of rifampicin-resistant Pseudomonas syringae p v . morsprunorum in samples of rainwater collected under a Montmorency sour cherry tree with a sequential sampler during a rain on 8 October 1980. Arrows indicate the beginning and end of a period of light mist . 31 CHAPTER 2 Figure 2.1 Recovery of rifampicin-resistant Pseudomonas syringae pv. morsprunorum (PsmR) from leaves and runoff rainwater from a Montmorency sour cherry tree in relation to mean daily air temperatures and rainfall. expressed as Populations were (colony-forming units of PsmR + 1 per milliliter rainwater) or per cm2 leaf. Figure 2.2 A, 1980; B, 1 9 8 1 .................... 50 Dispersal patterns for rifampicin-resistant Pseudomonas syringae pv. morsprunorum (PsmR) in runoff rainwater collected serially under a Montmorency sour cherry tree during rains on 14 (type 2) and 15 (type 1) June 1980 . . . xi 51 CHAPTER 3 Figure 3. 1 Populations of rifampicin-resistant Pseudomonas syringae p v . morsprunorum (PsmR) on Montmorency sour cherry leaves sprayed with copper treatments. A, In 1980, treatments were 986 mg/L captafol, 949 mg/L tribasic copper sulfate (TBS), and 100 mg/L and 200 mg/L Citcop 4E. B, In 1981, treatments were 636 mg/L TBS and check treatments (nonsprayed and 491 mg/L captafol combined) Figure 3. 2 .......................... The relationship between rainfall and the loss of copper from Montmorency sour cherry leaves. Trees were sprayed on 18 September 1981 with 6 g/L tribasic copper sulfate (TBS) + 12 g/L hydrated lime and on 18, 28 September, 8, and 19 October with 2.4 g/L TBS with or without 4.8 g/L hydrated lime. Arrows indicate the dates sprays were applied. not sprayed Figure 3. 3 Check trees were ................................. Nomogram relating the amount of copper lost from Montmorency sour cherry leaves to initial deposits of copper on the leaves and to the amount of rainfall. A, Citcop 4E; B, Tribasic copper sulfate ................. APPENDIX B Figure 5.1 Plot of residuals (observed minus predicted values) against predicted amounts of copper loss indicating that the errors are independent, have zero mean with a constant variance, and follow a normal distribution. A, Citcop 4E; B, Tribasic copper sulfate ............. GENERAL INTRODUCTION Bacterial canker on Prunus sp. is caused by Pseudomonas syringae pv. syringae (van Hall) Young et al. and £. syringae pv. morsprunorum (Wormald) Young et al. The disease is found in most regions of the world where Prunus sp. are grown, including California (51), Georgia (21), Michigan (36), Oregon (6), and Utah (2), USA; Nova Scotia (32) and Ontario (1) Canada; England (52); France (49); Germany (44); New Zealand (22); and Poland (39). Pseudomonas syringae pv. syringae was first described by van Hall in 1902 as c using a leaf spotting on lilac (Syringae vulgaris L.) (48). Since 1902, P. syringae pv. syringae has been identified as a pathogen on more than 43 hosts, including such important crops as apple, cherry, Citrus sp., common bean, peach, and pear (7). In 1932, Wormald (52) identified syringae pv. morsprunorum as the cause of bacterial canker on plum. Pseudomonas syringae pv. morsprunorum has been identified as a pathogen on 10 hosts; i.e., almond, apricot, sweet and sour cherry, common plum, flowering plum, Myrobalan plum, peach, Prunus sibirica, and purple leaved plum (7). Pseudomonas syringae pv. morsprunorum and P^. syringae pv. syringae are obligate aerobes, 0.7-1.2 by 1.5-3.0 um in size with polar multitrichous flagella, oxidase negative, and they produce a diffusible fluorescent pigment (20). Since Wormald described _P. syringae pv. morsprunorum, there has been a controversy whether £. syringae pv. morsprunorum is 1 2 different enough from syringae pv. syringae to warrant a separate species name (33). Many studies have focused on the taxonomic controversy of £. syringae pv. morsprunorum and P^. syringae pv. syringae, fluorescent psuedomonads, and plant pathogenic bacteria in general (31, 33). In a taxonomic study, Garrett et al. (31) found £. syringae pv. syringae and £. syringae pv. morsprunorum gave positive and negative responses, respectively, to the following tests: liquefaction of gelatin; hydrolysis of casein, aesculin, and arbutin; L-leucine as sole carbon and nitrogen source; use of lactic acid; and metabolism of arbutin. Pseudomonas syringae pv. morsprunorum and £. syringae pv. syringae gave positive and negative responses, respectively, to the following tests: acid production on purple lactose agar, use of tartaric acid, and use of L-tyrosine as sole carbon and nitrogen source (31). Pseudomonas syringae p v . syringae and syringae p v . morsprunorum produced yellow and white pigment, respectively, in sucrose broth (31). Fruit inoculations with P^. syringae pv. syringae produced black sunken lesions on pear and cherry fruitlets and lemons while syringae pv. morsprunorum produced no lesions on pear fruitlets and only small brown superfical lesions on cherry fruitlets and lemon (31). Garrett and Crosse (29) were able to distinguish strains of syringae pv. morsprunorum and P^. syringae pv. syringae, indigenous to England, by phage typing. Presley and Crosse (42) also used phage typing to separate two races of P^. syringae pv. morsprunorum, each pathogenic on different sweet cherry cultivars (Roundel and Napoleon). Ice nucleation activity has been used to detect taxonomic differences between P. syringae pv. morsprunorum and P. syringae 3 pv. syringae. Hirano et al. (35) tested many fluorescent plant pathogenic pseudomonads for ice nucleation activity. Of 10 £. syringae pv. syringae isolates from Prunus species, six had ice nucleation activity at -5.5 C; while of 10 syringae pv. morsprunorum isolates, none had ice nucleation activity at -5,5 or -10.0 C (34). But, Latorre and Jones (37) found 48.6% of 105 isolates of P. syringae pv. syringae had ice nucleation activity between -4 and -6 C, while at the same temperatures only 24% of 96 £. syringae pv. morsprunorum isolates had ice nucleation activity. Production of syringomycin has been used to separate I?, syringae pv. morsprunorum from P^. syringae pv. syringae. Seemuller and Arnold (46) found 34 of 40 isolates of £. syringae pv. syringae produced syringomycin in vitro and none of 36 P^. syringae pv. morsprunorum isolates produced syringomycin. found 74.5% of 132 Similarly, Latorre and Jones (37) syringae pv. syringae isolates produced syringomycin and only three of 127 of £. syringae pv. morsprunorum isolates produced syringomycin. By using the tests mentioned, P. syringae pv. morsprunorum can be separated from _P. syringae pv. syringae. Often there were many strains of P. syringae pv. morsprunorum and ]?. syringae pv. syringae that did not respond typically to the tests mentioned. Therefore, some researchers prefer to identify species of P^. syringae as physiotype 1 (1?. syringae pv. syringae) or physiotype 2 (P. syringae pv. morsprunorum) from known hosts (17). Crosse (17) suggested plant pathogenic bacteria should be classified according to characters of the bacteria that favored its natural selection in the field. Young et al. (53) proposed a change 4 in the nomenclature and classification of all plant pathogenic bacteria in 1978 that would lower many species to pathovars of one specie and the primary distinction between pathovars would be host range. This proposal was approved by the International Society for Plant Pathology in 1980 (24). Species names such as £. morsprunorum and P^. syringae were changed to .P. syringae pv. morsprunorum and ]?. syringae pv. syringae, pathovars of 1?. syringae. Pseudomonas syringae pv. morsprunorum was distinguished from P^. syringae pv. syringae based on differences in host range; meanwhile both pathovars were still known pathogens of apricot, sweet and sour cherry, peach, and common plum (7). The new rules on classification stated that biochemical tests can aid in distingishing pathovars but host range must be the deciding factor for classification. Disease symptoms of bacterial canker on Prunus species are most prevalent in spring. On sweet and sour cherry bacterial canker causes lesions on leaves and fruit, and cankers on limbs (36, 37). Lesions on leaves are angular, delimited by veins on the leaf, and often coalesce during severe infections (37). Chlorotic halo and water- soaked areas of leaf tissue commonly surround leaf lesions. During periods of hot dry weather, lesions may dry up and fall out of the leaf causing a tattered or shot-hole appearance. Lesions that develop on sweet and sour cherry fruit and pedicels begin as water-soaked areas that develop into dark, sunken regions (37). In Michigan, cankers caused by P^. syringae morsprunorum on trunks and scaffold limbs are rare on sour cherry (37) but cankers on sweet cherry are commonly caused by P^. syringae pv. syringae (36). Cankers on limbs 5 usually develop at the base of infected fruit spurs, and can girdle and destroy entire limbs of trees (9). The disease cycle of bacterial canker on sweet cherry has been studied extensively in Oregon (7) and England (16) where bacterial canker is caused by P^. ayringae pv. syringae and £. syringae pv. morsprunorum, respectively. In England, P^. syringae pv. morsprunorum overwinters in cankers that are active through the winter months in mild temperatures (9). In spring, P^. syringae pv. morsprunorum are disseminated from cankers to swollen buds, flowers, and leaves by splashing rainwater (16). leaves are susceptible to leaf spotting (19). Only young cherry Schimdle and Zeller (44) determined from in vitro studies with sour cherry (Prunus cerasua) and an inoculum mixture of £. syringae pv. morsprunorum and P^. syringae pv. syringae that the optimal temperature range for leaf spotting was 15 to 25 C with 100% relative humidity. Cold temperatures may predispose leaf tissue to leaf spotting because of the ice nucleation activity of some £. syringae pv. syringae strains. syringae pv. morsprunorum and Sour cherry trees inoculated with V_. syringae pv. syringae and exposed to temperatures ranging from -2.0 to -0.5 C with 100% relative humidity developed more leaf spotting symptoms than trees exposed to 0 C (54). Similarly, terminal shoots of peach inoculated with P. syringae pv. syringae developed typical shoot lesions when exposed to -5 C for 15 days (50). In spring, severe epidemics of bacterial canker are associated with cool moist weather conditions (10, 16). Crosse (10) reported in England that a severe outbreak of bacterial canker leaf spotting in May 1953 was associated with three consecutive days of rain 6 accompanied by heavy winds. Crosse (10) suggested that windblown rain containing JP. syringae pv. morsprunorum inoculum was impacted into » leaves via the stomata. In summer, cankers are not active on sweet cherry trees and the bacteria in the cankers often die (9). morsprunorum and But, Pseudomonas syringae pv. syringae pv. syringae do survive as epiphytes associated with leaves and numerous studies have focused on this subject. Crosse (14) measured populations of JP. syringae pv. morsprunorum on cherry leaves by determining the quantities of syringae pv. morsprunorum in leaf washings. Populations of P^. syringae pv. morsprunorum on leaves of Napoleon, a bacterial cankersusceptible cultivar, were 1.2 to 3.3 times greater than those on Roundel, a bacterial canker-resistant cultivar (15). These differences correlated with leaf scar infections in autumn and were not associated with leaf spot infections in spring (15). In Poland, Burkowicz (5) measured inoculum potential of P^. syringae pv. morsprunorum on sweet cherry (Czarna Pozna and Hedelfinger) leaves using Crosse's leaf washing procedure (14). For three years, populations of P. syringae pv. morsprunorum varied from 10 3 to 10 7 colony-forming units/leaf and populations were always highest in autumn (5). During the summer months in Michigan, Latorre and Jones (37) found that P_. syringae pv. morsprunorum survived as an epiphyte associated with leaves of sour cherry. They also found JP. syringae pv. morsprunorum was predominantly associated with leaf spotting on sour cherry in spring, but _P. syringae pv. syringae and £. syringae pv. morsprunorum were found on symptomless leaves and the former was commonly found on ground cover plants in the orchard (37, 38). In 7 Georgia, pathogenic strains of syringae pv. syringae were also isolated from apparently healthy peach twigs through the growing season except from June to September (21). Because of the wide host range of I>. syringae pv. syringae, it appears to be ubiquitous in many orchards. The significance of the simultaneous presence of syringae pv. morsprunorum and P^. syringae pv. syringae in the orchard is not known. The epiphytic stage of the life cycle permits _P. syringae pv. morsprunorum and 1?. syringae pv. syringae to survive through the summer and infect woody plant tissue in autumn for overwintering. In England, £. syringae pv. morsprunorum infected sweet cherry leaf scars and when inoculum concentrations of J?. syringae pv. morsprunorum were increased, the number of leaf scar infections increased (12). In autumn, leaf scars were most susceptible to infection in mid-October (12). Leaf scar infections on sweet cherry with a cherry strain of _P. syringae pv. morsprunorum were 50% less when leaf scars were first inoculated with a £. syringae pv. morsprunorum strain from prune (19). During leaf scar infection, the bacteria were absorbed into the leaf scar tissue, eventually migrating to adjacent buds (11, 34). In England during mild winters, stem cankers developed from P^. syringae pv. morsprunorum-infected leaf scars. In Oregon, buds on sweet cherry trees are infected directly by P^. syringae pv. syringae from November to January rather than through leaf scars (6). Throughout the growing season, rainwater is important for the dissemination of _P. syringae pv. morsprunorum and £. syringae pv. syringae from cankers to leaves, from leaves to leaves, and from leaves to leaf scars (16). Fregoin and Crosse (26) have collected P. syringae pv. morsprunorum isolates from rainwater but they did not determine the relationship between inoculum levels and disease incidence. Because many factors are involved for a severe epidemic of bacterial canker to occur, the disease is sporatic, but can unexpectedly cause extensive damage to cherry trees. To reduce infection of cherry trees by rain-disseminated P_. syringae pv. morsprunorum and P^. syringae pv. syringae, chemical control practices have been evaluated. These studies have focused on reducing infection of leaves, blossoms, and fruit in spring, and leaf scars in autumn. In 1945, Montgomery and Moore (41) reported the incidence of bacterial canker on Bigarreau de Schrecken sweet cherry trees in England was less after 4 years of single-treatment applications of Bordeaux mixture in spring and autumn than on nonsprayed check trees. Bordeaux mixture applied after petal fall did cause severe copper phytotoxicity damage to leaves (41). In greenhouse studies, Dye (23) reported that 250 ug/ml streptomycin sulphate reduced wound-inoculated infections by P^. syringae pv. syringae on peach seedlings by 84* compared to nonsprayed check seedlings. In the field Crosse (13) found three sprays of streptomycin hydrochloride applied at full bloom and at 75 and 100% petal fall reduced leaf spotting by 94%, while a single application of Bordeaux mixture at white bud had no effect on leaf spot infection. Copper sprays applied to sweet cherry trees in autumn reduced the level of leaf scar infection by P^. syringae pv. morsprunorum that autumn as well as canker development the following spring (13). Connecticut (43) and California (4), repeated applications of streptomycin and Bordeaux mixture before, during, and after bloom In 9 significantly reduced the incidence of blossom blast on pear caused by £. syringae pv. syringae. In Oregon, chemical control practices are recomended for bacterial canker of sweet cherry (3). In Michigan, no control practices are recommended for bacterial canker on sweet and sour cherry. Changes in cultural practices may also reduce the incidence of bacterial canker. In Oregon, it was reported (8) and recommended (3) that scions budded high on the rootstocks had lower incidence of bacterial canker than those budded at ground level. Development of bacterial-resistant Prunus sp. cultivars is another means to control bacterial canker. Sour.cherry scion cultivar Schattenmorello is highly resistant to leaf infection by P^. syringae pv. syringae (45) and supports low epiphytic populations of P. syringae pv. morsprunorum compared with Nefris sour cherry (47). A drawback with resistant cultivars is that £. syringae pv. morsprunorum and P^. syringae pv. syringae have many strains with different host ranges and levels of virulence; therefore, resistance to bacterial canker will probably be overcome by new strains of the pathogen. For example, in England Garrett and Crosse (29) identified two strains of P^. syringae pv. morsprunorum; one was a pathogen of prune and the other was a pathogen of cherry (race 1). Napoleon and Roundel sweet cherry cultivars were susceptible and resistant, respectively, to the cherry strain of P. syringae pv. morsprunorum (15). Later in 1974, race 2 of the cherry JP. syringae pv. morsprunorum strain was identified using phage typing and was pathogenic on Roundel, the bacterial canker-tolerant sweet cherry cultivar (27). Garrett (28) reported F12/1 was tolerant to race 1 of P^. syringae pv. morsprunorum 10 but was susceptible to race 2. Also, Allen and Dirks (1) screened many sweet cherry cultivars for resistance to P^. syringae pv. morsprunorum and P. syringae pv. syringae with wound inoculations and found different pathogen strains were pathogenic on different cultivars. These examples of variability of P^. syringae pv. morsprunorum and P_. syringae pv. syringae indicate that a breeding program for bacterial canker-resistant cherry cultivars should be a continuous program. Evidence of an indigenous system of gene transfer has been found for £. syringae pv. morsprunorum (25), indicating the possibility £. syringae pv. morsprunorum may overcome host resistance by genetic recombination. The objectives of this study were to: (i) develop a rain sampling device to monitor rain-disseminated pathogens from trees; (ii) investigate the dissemination of £. syringae pv. morsprunorum in rainwater and the overwintering of _P. syringae pv. morsprunorum and syringae pv. syringae in leaf scars of fruit spurs and on ground cover in a Montmorency sour cherry orchard; (iii) investigate the effectiveness of fixed copper compounds for reducing populations of syringae pv. morsprunorum from early bud break through early summer on Montmorency sour cherry and to examine the retention of copper on the foliage; and (iv) to test the susceptibility of eight cherry rootstock cultivars to P^. syringae pv. morsprunorum and JP. syringae pv. syringae by wound inoculations. LITERATURE CITED Allen, W. cherry in R., and Dirks, V. A. 1978. Bacterial canker of sweet the Niagara peninsula of Ontario: Pseudomonas species involved and cultivar susceptibities. Can. J. Plant Sci. 58:363-369. Anderson, J. L . , Wadley, B. N . , and Schaelling, J. P. 1969. Pseudomonas syringae infection of sweet cherry fruit in Utah. Plant Dis. Rep. 53:301-303. Anonymous. 1976. Oregon Plant Disease Control Handbook. Extension Service, Oregon State University, Corvalis 97331. Bethell, R. S., Ogawa, J. M . , English, W. H . , Hansen, R. R . , Manji, B. T. and Schick, F. J. 1977. to control pear blossom blast. Burkowicz, A. 1981. Copper-streptomycin sprays Calif. Agric. 31:7-9. Population dynamics of epiphytic Pseudomonas morsprunorum on the leaf surfaces of two sweet cherry cultivars. Fruit Sci. Reports 1:37-48. Cameron, H. R. 1962. Mode of infection of sweet cherry by Pseudomonas syringae. Phytopathology 52:917-921. Cameron, H. R. Diseases of deciduous fruit trees incited 1962. by Pseudomonas syringae van Hall. with additional data. A review of the literature Oregon Agric. E x p t . Stn. Tech. Bull. 66. 64 pp. Cameron, H. R. 1971. Effect of root or trunk stock on susceptibility of orchard trees to Pseudomonas syringae. Plant Dis. Rep. 55:421-423. Crosse, J. E. 1954. of cherry and plum. Bacterial canker, leaf spot, and shoot wilt Rep. East Mailing Res. Stn. 1953:202-207. 12 10. Crosse, J. E. 1956. An epidemic leaf spot and spur wilt of cherry caused by Pseudomonas mors-prunorum. Rep. East Mailing Res. Stn. 1955:122-125. 11. Crosse, J. E. 1956. Bacterial canker scar infection of cherry. 12. Crosse, J. E. 1957. of stone-fruits. II. Leaf J. Hortic. Sci. 31:212-224. Bacterial canker of stone-fruits. III. Inoculum concentration and time of inoculations in relation to leaf-scar infections 13. Crosse, J. E. Crosse, J. E. Ann. Appl. Biol. 1957. Streptomycin in the control canker of cherry. 14. of cherry. 45:19-35. of bacterial Ann. Appl. Biol. 45:226-228. 1959. Bacterial canker of stone-fruits. IV. Investigation of a method for measuring the inoculum potential of cherry trees. 15. Crosse, J. E. Ann. Appl. Biol. 47:306-317. 1963. Bacterial canker of stone-fruits. V. A comparison of leaf-surface populations of Pseudomonas mors­ prunorum in autumn on two cherry varieties. Ann. Appl. Biol. 52:97-104. 16. Crosse, J. E. 1966. Epidemiological relations of the pseudomonad pathogens of deciduous fruit trees. Annu. Rev. Phytopathol. 4:291-310. 17. Crosse, J. E. 1968. The importance and problems of determining relationships among plant-pathogenic bacteria. Phytopathology 58:1203-1206. 18. Crosse, J. E., and Garrett, C. M. E. stone-fruits. VII. 1966. Bacterial canker of Infection experiments with Pseudomonas mors-prunorum and P. syringae. Ann. Appl. Biol. 58:31-41. 13 19. Crosse, J. E., and Garrett, C. M. E. 1970. Pathogenicity of Pseudomonas morsprunorum in relation to host specificity. J. Gen. Microbiol. 62:315-327. 20. Doudoroff, M . ,* and Palleroni, N. J. Migula 1894, Pages 217-243 in: Determinative Bacteriology, Gibbons eds., 21. 1974. In Genus I. Pseudomonas Bergy's Manual of 8th Ed. R. E. Buchanan and N. E. Williams & Wilkins Co., Baltimore, Md, USA. Dowler, W. M . , and Weaver, D. J. 1975. Isolation and characterization of fluorescent pseudomonads from apparently healthy peach trees. 22. Dye, D. W. 1954. Phytopathology 65:233-236. Blast of stone-fruit in New Zealand. New Zealand J. Sci. Technol. 35:451-461. 23. Dye, D. W, and Dye, M. H. 1954. Effectiveness of therapeutants including antibiotics in preventing development of blast of stone fruit (Pseudomonas syringae van Hall). New Zealand J. Sci. Technol. 35:21-26. 24. Dye, D. W., Bradbury, J. F., Goto, M . , Hayward, A. C., Lelliott, R. A., and Schroth, M. N. 1980. International standards for naming pathovars of phytopathogenic bacteria and list of pathovar names and pathotype strains. 25. Errington, J., and Vivian, A. Rev. Plant Pathol. 59:153-168. 1981. An indigenous system of gene transfer in the plant pathogen Pseudomonas morsprunorum. Gen. Microbiol. 26. 124:439-442. Freigoun, S. 0., and Crosse, J. E. rain water from fruit trees. 1971:127-129. J. 1972. A trap for collecting Rep. East Mailing Res. Stn. 14 27. Freigoun, S. 0., and Crosse, J. E. 1975. Host relations and distribution of a physiological and pathological variant of Pseudomonas morsprunorum. 28. Garrett, C. M. E. mors-prunorum. 1978. Ann. Appl. Biol. 81:317-330. Pathogenic races of Pseudomonas Pages 889-890 in: Proc. IVth Int. Conf. Plant Path. Bact. Angers. 29. Garrett, C. M. E., and Crosse, J. E. 1963. lysogeny in the plant pathogens Pseudomonas syringae. 30. Garrett, Observations on mors-prunorum and P s . J. Appl. Bact. 26:27-34. C. M. E., and Crosse, J. E. 1975. Interactions between Pseudomonas morsprunorum and other pseudomonads in leaf-scar infection of cherry. 31. Physiol. Plant Pathol. 5:89-94. Garrett, C. M. E., Panagopoulos, C. G., and Crosse, J. E. 1966. Comparison of plant pathogenic pseudomonads from fruit trees. J. Appl. Bact. 29:342-356. 32. Gourley, C. 0. Scotia. 33. 1965. Bacterioses of stone fruits in Nova Can. Plant Dis. Surv. 45:101-102. Greib, T . s Fliege, H. F . : and Stille. B. 1975. Untersuchungen zur Differenzierung von Pseudomonas morsprunorum Wormald und Pseudomonas syringae van Hall mit Hilfe physiologischer und biochemischer Verfahren. 34. Hignett, R. C. 1974.Absorption of Pseudomonas morsprunorum through leaf scars. 35. Hirano, S. S., Zbl. Bakt. Abt, II. 130:409-423. J. Gen. Microbiol. 80:501-506. Maher, E. A., Kelman, A., and Upper, C. D. 1978. Ice nucleation activity of fluorescent plant pathogenic pseudomonads. Pages 717-724 in: Path. Bact. Angers. Proc. IVth Int. Conf. Plant 15 36. Jones, A. L. Michigan. 37. 1971. Bacterial canker of sweet cherry in Plant Dis. Rep. 55:961-965. Latorre, B. A. and Jones, A. L. 1979. Pseudomonas morsprunorum, the cause of bacterial canker on sour cherry in Michigan, and its epiphytic association with P^. syringae. Phytopathology 69:335-339. 38. Latorre, B. A. and Jones, A. L. 1979. Evaluation of weeds and plant refuse as potential sources of inoculum of Pseudomonas syringae in bacterial canker of cherry. Phytopathology 69:1122- 1125. 39. Lyskanowska, M. K. 1976. (Prunus avium) in Poland. economic importance. 40. Lyskanowska, M. K. I. Symptoms, disease development and Plant Dis. Rep. 60:465-469. 1976. (Prunus avium) in Poland. Pseudomonas mors-prunorum. 41. Bacterial canker of sweet cherry Bacterial canker of sweet cherry II. Taxonomy of the causal organism, Plant Dis. Rep. 60:822-826. Montgomery, H. B. S., and Moore, M. H. 1945. bacterial canker and leaf-spot in sweet cherry. The control of J. Pomology 21:155-163. 42. Persley, G. J., and Crosse, J. E. 1978. A bacteriophage specific to race 2 of the cherry strain of Pseudomonas morsprunorum. 43. Ann. Appl. Biol. 89:219-222. Sands, D. C. and McIntyre, J. L. 1977. control pear blast caused by Pseudomonas Rep. 61:311-312. Possible methods to syringae. Plant Dis. 16 44. Schimdle, A. and Zeller, W. 1976. Der Ginfluss von Temperatur und Luftfeuchte auf die Infektion von Pseudomonas spp. bei Blattern von Sauerkirschen (Prunus cerasus). Phytopath. Z. 87:274-283. 45. Schimdle, A., and Zeller, W. 1979. Untersuchungen zur Resistenz von Sauerkirschsorten gegen den Bakterienbrand Pseudomonas syringae van Hall. Nachrichtenbl. Deut. Pflanzenschutzd. 31:177— 178. 46. Seemuller, E. and Arnold, M. 1978. Pathogenicity, syringomycin production and other characteristics of pseudomonad strains isolated from deciduous fruit trees. Pages 703-710 in: Proc. IVth Int. Conf. Plant Path. Bact. Angers. 47. Sobiczewski, P. 1978. Epiphytic populations of pathogenic pseudomonads on sour cherry leaves. Pages 753-762 in: Proc. IVth Int. Conf. Plant Path. Bact. Angers. 48. van Hall, C. J. J. 1902. bakterielle plantenziekten. Bijdragen tot de kennis der Ph.D. thesis, University of Amsterdam. 49. Vigouroux, A. 1970. Une nouvelle bacteriose du Ptcher: desription, etiologie, developpement du parasite. Ann. Phytopathol. 2:155-197. 50. Vigouroux, A. 1974. Obtention de symptSmes de bacteriose du Pecher (Pseudomonas morsprunorum f . sp. persicae) sur rameaux de Ptcher detaches et conserves en survie effet du froid. Ann. Phytopathol. 6:95-98. 51. Wilson, E. E. 1939. Factors affecting development of the bacterial canker of stone fruits. Hilgardia 12:259-298. 52. Wormald, H. Britain. trees. 53. IV. 1932. Bacterial disease of stone fruit trees in The organism causing bacterial canker in plum Trans. Brit. Mycol. Soc., 17:157-169. Young, J. M., Dye, D. W., Bradbury, J. F., Panagopoulus, C. G., and Robbs, C. F. 1978. A proposed nomenclature and classification for plant pathogenic bacteria. New Zealand J. Agric. Res. 21:153-177. 54. Zeller, W . , and Schimdle, A. 1979. Der Einfluss von Frost auf die Infektion von Pseudomonas syringae van Hall bei BlSttern von Sauerkirsche (Prunus cerasus). Pflanzenschutzd. 31:97-99. Nachrichtenbl. D e u t . CHAPTER 1 A Sequential Sampler for Monitoring Water-Disseminated Pathogens from Trees ABSTRACT An apparatus for studying the dissemination of waterborne pathogens from trees is described and illustrated. It is used for monitoring a population of rifampicin-resistant Pseudomonas syringae pv. morsprunorum in rainwater collected under a Montmorency sour cherry tree. The apparatus collects rain from a 5,026-cm saves 5- to 8-ml subsamples for each 0.5-mm rainfall. 2 area and Simplicity of design and sequential sampling based solely on a mechanical mechanism are advantages of this apparatus for certain phytopathological studies. INTRODUCTION Many bacterial pathogens of fruit trees are waterborne during part of their life cycle; e.g., £. syringae pv. morsprunorum and Pseudomonas syringae pv. syringae, the cause of bacterial canker on cherry (2, 4), Splashing and windblown rain are considered important in the spread of these bacteria (2, 4). Bacterial canker is sporatic in occurrence; therefore, to comprehensively study the role of rain in removing and disseminating £. syringae pv. morsprunorum from leaves, we needed an apparatus that sequentially sampled rain dripping from 18 19 infected Prunus cerasus L. cv. Montmorency sour cherry trees. Sequential rain samplers are used for air pollution studies (1, 7, 8), but due to their large size, limited sampling capacity, and high cost, they are not practical for most epidemiological studies or for widespread monitoring of chemicals. This paper describes the design and use of a sequential rain sampler for monitoring runoff rainwater for a bacterial pathogen. MATERIALS AND METHODS Instrument design and operation. The self-advancing sequential rain sampler resembles a fraction collector and tipping bucket rain gauge combined (Figures 1.1 and 1.2). A large funnel collects the rain that fills and tips a divided bucket. On tipping, a subsample of water is siphoned into a 9-ml collection tube and the rack holding the tubes is automatically advanced. The funnel is made of 8-mil polyethylene and has an 80-cm diam opening with a rim made from 1.9-cm diam rigid plastic tubing (Figure 1.1). The rim is supported 92 cm above the ground with 51-cm lengths of 1.9-cm diam steel pipe attached to the sampler four cover. A collar placed on the sampler cover is constructed from an inverted 20-cm diam plastic funnel, with a 10-cm diam orifice. The bottom of the polyethylene funnel is fitted into the collar and is held in place by inserting a 10-cm diam plastic funnel inside the polyethylene. bottom of the 10-cm diam plastic funnel directs water into The the tipping bucket and is held in place with a metal pin. The sampler cover is made of a 25-cm diam x 31.7-cm high metal can (Figure 1.1). The sampler is mounted on two 1.9-cm thick plywood 20 boards spaced 10 cm apart with three 0.95xl6-cm leveling bolts. The upper and lower boards are 55- and 77-cm equilateral triangles, respectively. The lower board is fastened to the ground with three 0.95x30-cm steel spikes. The tipping bucket and sampling mechanism are constructed of 3.17- and 6.35-mm thick plexiglass (Figure 1.2). The tipping bucket has two compartments, each with a capacity of 280 ml. Tipping bucket volumes are adjusted by placing counterweights on a metal rod suspended below the tipping bucket. The axis for the tipping bucket is supported on both ends with removable bearing supports of Teflon (E.I. duPont de Nemours & Co., Wilmington, DE 19898). The movement of the tipping bucket also advances the rack of collection tubes. A plexiglass arm, reinforced with a 2.17-mm steel shaft, is attached near the top center of the tipping bucket. A Teflon sliding bearing at the end of the steel shaft is inserted into a pivoting arm connected to the advancing mechanism located at the base of the sampler. Each compartment of the tipping bucket empties into a separate drain. The floor of each drain slopes down 6.15 mm from the side walls to a center outlet. A 2.5-cm long x 2-cm wide x 2.5-cm high plexiglass box with an open top is placed on the floor of each drain directly above a row of collection tubes. Each box has a siphon made of 5-mm diam glass tubing for transferring water from the box to a collection tube each time the tipping bucket empties. A turntable and rack of collection tubes are located underneath the floor of the drains. The turntable consists of three 21.6-cm diam plates made from 3.17-mm thick plexiglass. The top and bottom plates are notched with 30 teeth pointing clockwise and counterclockwise, respectively, and are cemented to the center plate. The turntable revolves on a 10.16-cm diam Lazy Susan bearing (Triangle Manufacturing Co., Oshkosh, WI 54901). A removable rack for holding 13xl00-mn test tubes is rotated by the turntable. The rack is made of three 20.5-cm diam plexiglass plates, two of which are 3.15-mm thick and drilled with holes in two concentric circles each to hold 30 tubes. Using this same pattern, the bottom plate, 6.35-nm thick, is drilled with wells 3.2-mm deep for stabilizing the respective tubes. Collection tubes in the outside circle are filled by the right siphon, those in the inside circle by the left siphon. Two consecutive bucket tips fill a collection tube in each row of tubes and also advance the turntable one notch. The back and forth movement of the tipping bucket pivots the advancing arm and slides a Teflon advancing block back and forth in a partially enclosed box. An angled piece of flexible metal, protruding from the inside surface of the advancing block, catches a clockwise tooth on the turntable. When the left bucket tips, the advancing block moves from left to right, rotating the turntable counterclockwise one notch. As the advancing block moves from left to right, a second angled piece of flexible metal is exposed and catches a counterclockwise tooth, stopping the advancement of the turntable. The metal stop is pulled away from the turntable as the advancing block moves from right to left. Clockwise movement of the turntable is prevented by a stationary stop piece. To record when rain samples are collected, an electronic switch (MICROswitch, Freeport, IL 61032) is attached to the pivoting arm just below the plexiglass arm coming from the tipping bucket. Each tipping 22 of the bucket causes the switch to complete an electronic circuit that sends a pulse to an event recorder (WEATHERtronics, Inc., West Sacramento, CA 95691). Evaluation of instrument. To measure the volume of water needed to tip each bucket, known quantities of water were poured into the sampler until the bucket tipped. This was repeated 20 times per bucket for each of the three samplers. The volumes of water were _3 multiplied by 1.99 x 10 , the millileters of rain necessary to collect 1 ml of water in the sampler, to obtain the quantity of rain needed to tip each bucket. Redistribution of bacteria within each sampler was tested by pouring a bacterial suspension into the samplers, then removing bacteria that adhered to each sampler with simulated rain. A rifampicin-resistant strain of £. syringae pv. morsprunorum (PsmR) (6) grown on King's medium B (5) for 2 days at 22 C was suspended in 0.01-M phosphate buffer adjusted to pH 7.2. The sampler was washed thoroughly with phosphate buffer before pouring in enough bacterial suspension for two bucket tips. This was immediately followed by pouring in enough phosphate buffer for six bucket tips. Bacterial suspensions and phosphate buffer were poured around the perimeter of the funnel. From each of the eight collection tubes, duplicate 0.1-ml subsamples were pipetted with a Finnpipette (Finnpipette, Helsinki, Finland) onto a modified King's medium B amended with 50 jig/ml rifampicin (Calbiochem-Behring Corp., La Jolla, CA 92037) and 25 jig/ml cycloheximide (Sigma Chemical Co., St. Louis, MO 63178). Colonies were counted after 5-days incubation at 22 C and the colony-forming units (cfu)/ml sample were computed. The concentration of bacteria in 23 each collection tube was expressed as a percentage of the concentration of bacteria in the initial suspension (bacterial frequency). This experiment was replicated three times per sampler. In the field a sequential rain sampler was positioned under the drip line of a Montmorency sour cherry tree in East Lansing, Michigan that was spray-inoculated with PsmR at sunset on 17, 29 April, 8, and 29 May 1980. Inoculum was prepared from 2-day-old cultures of PsmR grown on Ring's medium B incubated at 22 C. The cultures were g suspended in phosphate buffer to give a final concentration of 10 cfu/ml. The Montmorency sour cherry leaves were lightly misted with 1.5 L of inoculum applied with a handgun sprayer operated at 2 28 kg/cm . Concentrations of PsmR in rainwater were determined for each collection tube by plating duplicate 0.1-ml subsamples onto a modified King's medium B amended with rifampicin and cycloheximide. Colonies were counted after 5-days incubation at 22 C and the cfu per milliliter of rainwater were computed. Rainwater was plated within 12 hr after the end of each rain period. RESULTS AND DISCUSSION The amount of rain needed to tip each bucket varied slightly among samplers and ranged from 0.533 _+ 0.23 m m to 0.452 _+ 0.023 mm. When suspensions of bacteria were added to samplers that were still wet from being washed with phosphate buffer, there was a mean reduction of 16 and 9% in the concentration of bacteria in collection tubes 1 and 2, respectively, compared to the initial bacterial suspension (Figure 1.3A). When sterile phosphate buffer was added to the sampler (Figure 1.3A, samples 3 to 8), the concentration of 24 bacteria in successive collection tubes declined until no bacteria / were detected. The results fit a distinct pattern that was reconstructed with the following equation: M n - 0.9[0.9(I ) + 0.1(1 .)] + 0.1 (M ,) n n—i n —z (Eq. 1) where M ■ concentration of bacteria in collection tube number n, I = concentration of bacteria in the solution entering the sampler, and n ” collection tube number. This equation assumes that the funnel and sampling mechanism each retain 10% of the bacteria from the previous sample. Therefore, the quantities 0.1 (ln and 0.1 represent the amount of bacteria remaining on the funnel and sampling mechanism, respectively. To reconstruct the experiment using equation 1, we assumed 1 . * 0 and M „ * 0 for sample 1 and M n-1 n-2 r n-2 0 for sample 2, r * since the sampler was cleaned with phosphate buffer prior to the experiment. When the equation was solved with known concentrations of bacteria entering the sampler, the predicted values were similar to the empirical data points (Figure 1.3B). A correlation coefficient of r ■ 0.96 was obtained when the empirical data were compared with the predicted data. Equation 1 was used to estimate the concentration of bacteria in rain entering the sampler from the concentration of bacteria detected in the collection tubes. If the sampler was dry before a rain period, the concentration of bacteria in the first collection tube would be equal to the concentration of bacteria in water entering the sampler. The concentration of bacteria in the second collection tube would be corrected for retention of bacteria on the funnel and the concentration of bacteria in the third to nth collection tubes would be corrected for retention of bacteria on the funnel and sampling mechanism. The prediction equations were: 1st tube 1^ ■ Mj (Eq. 2) 2nd tube Ij - [M2 - O . K l j M / O . 9 (Eq. 3) 3rd - nth tubes I - [M -0.09(1 ,) -0.1(M J J / 0 . 8 1 n n n—l n— / (Eq. 4) where, I * concentration of bacteria in the rain entering the sampler M = measured concentration of bacteria in the collection tube, and n collection tube number. Retention of bacteria on the funnel was accounted for by 0.1(1,) and 0.09(1 „) in equations 3 and 4, 1 n-2 respectively, and retention of bacteria on the sampling mechanism was accounted for by 0.1(Mn in equation 4. During a rain period on 3 October 1980, the sampler collected rainwater from a Montmorency sour cherry tree previously inoculated with PsmR. Actual concentrations of FsmR detected in the collection tubes and corrected concentrations were plotted over time (Figure 1.4). High concentrations of bacteria were detected in the first two samples collected after the onset of the rain. As the intensity of the rain increased, concentrations of bacteria in the samples decreased. But when the intensity of the rain decreased, concentrations of bacteria in subsequent samples increased with the maximum concentration being detected at 1630 hours. This may be explained if we consider that during a rain period the leaves are saturated with water and the bacteria in the leaves move out of the 26 leaves through the stomata at a constant rate (cfu per unit of time). When the rain intensity increases, the concentration per unit volume of water decreases and vice versa. After 1630 hours, there was a decline in the concentration of bacteria in the rain samples. Corrected concentrations of PsmR were similar to measured concentrations except in the last two samples. Corrected values were lower because many of the bacteria recovered were retained from the previous sample at 1630 hours. Crosse (3) measured the inoculum potential of J?. syringae pv. morsprunorum on sweet cherry leaves in vitro by washing detached leaves in water. The sequential sampler allows researchers to associate in vivo inoculum concentrations of pathogens in foliarrunoff rainwater with temporal environmental events. One sampler functioned for two growing seasons and two samplers functioned for one growing season without problems. There were no clocks to wind or batteries to replace except on the event recorder. The samplers were easily transported, disassembled, and cleaned in the field. The sequential sampler may also be useful for monitoring fungal spores and pesticide runoff in rainwater from trees. i 27 LITERATURE CITED 1. Coscio, M. R . , Pratt, G. C., and Krupa, S. V. 1982. refrigerated sequential precipitation sampler. An automatic Atmos. Envir. 16:1939-1944. 2. Crosse, J. E. 1956. An epidemic leaf spot and spur wilt of cherry caused by Pseudomonas mors-prunorum. Res. Stn. 3. Crosse, J. Rep. East Hailing 1955:121-125. E. 1959. Bacterial canker of stone-fruits. IV. Investigation of a method for measuring the inoculum potential of cherry trees. 4. Crosse, J. E. Ann. Appl. Biol. 47:306-317. 1966. Epidemiological relations of the pseudomonad pathogens of deciduous fruit trees. Annu. Rev. Phytopathol. 4:291-310. 5. King, E. 0., Ward, M. K., and Raney, D. E. 1954. Two simple media for the demonstration of pyocyanin and fluorescein. J. Lab. Med. 44:301-307. o. Latorre, B. A., and Jones, A. L. 1979. Evaluation of weeds and plant refuse as potential sources of inoculum of Pseudomonas syringae in bacterial canker of cherry. Phytopathology 69:1122- 1125. 7. Raynor, G. Si, and McNeil, J. P. precipitation sampler. 8. 1979. Atmos. Envir. An automatic sequential 13:149-155. Ronneau, C., Cara, J., Navarre, J. L., and Priest, P. automatic sequential rain sampler. 9:171-176. 1978. An Water, Air and Soil Pollution 28 Figure 1.1. Sequential rain sampler, A, funnel; B, cover; C, platform with three leveling bolts; and D, collar. 29 Figure 1.2. Mechanism for the sequential rain sampler. A, tipping bucket; B, bucket bearing; C, drain; D, plexiglass box and siphon; E, collection tube rack and tubes; F, turntable; G, pivoting arm; H, advancing block; I, electronic switch; J, stationary stop piece; K, drain tube; L, level; M, plexiglass arm; and N, counterweights. 30 SSlOO -A 80 4) >% o c S'60 it .240 •B ^ ^4 \ \ \ \ \\ ° Bucket I A Bucket II \ \ \ t tx I 20 o cd n --- T.“I— I I II I if1t »i I I f V + 2 3 4 5 6 7 8 Sample Number — I— I— I— I t 1 1 2 3 4 5 6 7 8 Sample Number Figure 1.3. Bacterial frequency (%) in collection tubes where sufficient quantities of bacteria in suspension were added to the sampler for two bucket tips, followed by sufficient quantities of phosphate buffer for six bucket tips. Prior to sampling, the sequential rain samplers were thoroughly rinsed with phosphate buffer. A, Empirical data were the means of three replicate experiments from three samplers. B, Predicted values were from the sampler retention equation. 31 £ 2. u. -o-M easured -** Corrected 900 1100 1300 1500 Hour of Day 1700 1900 Figure 1.4. Concentrations of rifampicin-resistant Pseudomonas syringae pv. morsprunorum in samples of rainwater collected under a Montmorency sour cherry tree with a sequential sampler during a rain on 8 October 1980. Arrows indicate the beginning and end of a period of light mist. CHAPTER 2 Dissemination in Rainwater and Overwintering of PseudomonaB syringae pv. morsprunorum and P. syringae pv. syringae in A Montmorency Sour Cherry Orchard ABSTRACT Rifampicin-resistant strains were used to study the survival and dissemination of Pseudomonas syringae pv. morsprunorum and .P. syringae pv. syringae (PsmR and PssR, respectively).in a Montmorency sour cherry orchard. Orchard ground cover plants were inoculated with PsmR and PssR in November 1979 and 1980. Both strains were recovered from plants in the ground cover in winter and in early spring, but not in late spring and summer. In spring, PsmR and PssR were recovered from buds on fruiting spurs following the inoculation of leaf scars the preceding autumn. Recovery of PsmR and PssR from buds following leaf scar inoculations in August 1980 was lower than from buds inoculated in October 1980, but recovery of PsmR from buds inoculated in September and October 1979 were similar. In two growing seasons, populations of PsmR isolated from leaves were relatively constant 3 4 2 (10 -10 colony-forming units/cm leaf), whereas populations of PsmR in rainwater were higher in spring and autumn than in summer. Two patterns of dispersal for PsmR were observed when rainwater was collected during rain periods with a sequential sampling apparatus. In the first pattern, populations of PsmR in rainwater increased 32 33 before leveling off; while in the second, initial populations in rainwater were higher but with time they declined and leveled off. The first and second patterns of dispersal were preceded by 5 days with and without rain, respectively. INTRODUCTION Bacterial canker on Prunus cerasus L. cv. Montmorency sour cherry in Michigan is caused by Pseudomonas syringae pv. morsprunorum (11). The disease is sporatic, but a severe epidemic occurred in 1976 (11). Leaf and fruit infections are the most common stages of bacterial canker noted on Montmorency sour cherry in Michigan. The epiphytic nature of £. syringae pv. morsprunorum on stone fruits is well established (2, 11, 14), but research on the dissemination of P. syringae pv. morsprunorum in rainwater is limited mostly to field observations (6). Waterborne syringae pv. morsprunorum are considered the source of inoculum that infects leaves and flowers in spring (6) and leaf scars in autumn (5). On sweet cherry, P_. syringae pv. morsprunorum infects fruit spurs through leaf scars from late September to mid-October, and the incidence of infection increases with increased inoculum concentrations (5). In England, P. syringae pv. morsprunorum overwinters in the vascular system of fruit spurs and in spring incites stem cankers and blossom blight (4, 8). In Oregon, P. syringae pv. syringae infects buds of sweet cherry directly rather than indirectly through leaf scars (7). Latorre and Jones (12) presented evidence that P. syringae pv. syringae overwinters on grasses and herbaceous plants under sour 34 cherry trees. The importance of this source of inoculum compared to other sources of inoculum in spring is not known. The objectives of this study were to investigate the dissemination of P. syringae pv. morsprunorum in rainwater and the overwintering of P. syringae pv. morsprunorum and 7. syringae pv. syringae in leaf scars of fruit spurs arid on ground cover in a Montmorency sour cherry orchard. MATERIALS AND METHODS Rifampicin-resistant strains. The rifampicin-resistant strain of syringae pv. morsprunorum (PsmR) we used was selected by Latorre and Jones (12). Rifampicin-resistant strains of P^. syringae pv. g syringae (PssR) were selected by spreading 10 colony-forming units (cfu) of P. syringae pv. syringae onto King's medium B (10) containing 50 pg/ml rifi-mpicin (Calbiochem-Behring Corp., La Jolla, CA 93037). Eleven isolates from sweet cherry leaves were screened for resistant strains. Colonies that developed after 4-days incubation at 22 C were considered rifampicin-resistant. Fathogenicity of parental and PssR strains were tested by inoculating tobacco leaves and Montmorency sour cherry fruit and leaves with inoculum concentrations of 10^ cfu/ml 0.01-M phosphate buffer, pH 7.2. A strain of PssR similar in pathogenicity to the parental isolate was used in this study. Inoculum of PsmR and PssR for each experiment was prepared from 2-day-old cultures grown on King's medium B incubated at 22 C. Petri plates were flooded with 10-ml phosphate buffer and the bacteria were suspended by stirring with a sterile glass rod. concentrations of 10 8 Inoculum cfu/ml phosphate buffer were determined 35 turbidimetrically and lower concentrations of inoculum were obtained by serial dilution in phosphate buffer. Overwintering. Survival of PsmR and PssR on ground cover plants was studied in a 5-yr-old Montmorency sour cherry orchard in 1979 and 1980 in East Lansing, Michigan. cover was Poa pratensis L. The predominant plant in the ground Other plants in decreasing frequency were Medicago sativa L . , Festuca elatior var. arundinacea (Shreb.), Panicum dichotomiflorum Michx., Stellaria media (L.) Cyrill., Medicago lupulina L . , Taraxacum officinale Weber., Rumex acetosella L . , Plantago lanceolata L . , Plantago major L . , and Lychnis alba M i l l . g Suspensions of PsmR and PssR (10 cfu/ml phosphate buffer) were sprayed onto these plants with a backpack sprayer charged with CO^ gas 2 and operated at 2.8 kg/cm . each isolate. Four separate plots were inoculated with A plot consisted of a row of three cherry trees with a strip of ground cover in the tree row killed with herbicide. A 9-m long x 30-cm wide area on each side of the herbicide strip was sprayinoculated with 1 L of inoculum. From each plot, two bulk samples of plant material, were collected, one on each side of the herbicide strip. Plants were sampled at 1-m intervals along the strips. Samples were weighed and a 20-g subsample from each sample was homogenized for2 min in a blender (Waring Products Inc., New Hartford, CT 06057) with 300-ml phosphate buffer. The homogenate was serially diluted in phosphate buffer and duplicate 0.1-ml subsamples were plated on a modified King's medium B (KBrc) amended with 50 jig/ml rifampicin and 25 Jig/ml cycloheximide (Sigma Chemical Co., St. Louis, M0 63178). Colonies were counted 36 after 5-days incubation at 22 C and the cfu per gram of plant material were calculated. In summer, leaves were collected from the three trees in each plot to determine if PsmR or PssR had moved up into the trees from the ground cover. Each tree was divided into quadrants and 10 leaves were randomly collected from each quadrant. The 40 leaves were measured with an area meter (Model LI-3000, LAMBDA Instrument Comp., Lincoln, NB 68504) before homogenizing them for 2 min in a blender with 300-ml phosphate buffer. The homogenate was serially diluted in phosphate buffer and duplicate 0.1-ml subsamples were plated on KBrc. Colonies were counted after 5-days incubation at 22 C and the cfu of PsmR per cm 2 of leaf were determined. Leaf scars on fruiting spurs were inoculated with PsmR and PssR on 21 September, 5 October 1979; 19 August, 8 October 1980 in East Lansing. 11 September, and Three Montmorency sour cherry trees were inoculated with each isolate. A leaf was removed from each spur and 0.05 ml of inoculum was placed on the exposed leaf scar. Inoculum concentrations were 10^. 10^, and 10^ cfu/ml phosphate buffer in 1979 2 a and 10 , 10 , and 10 a cfu/ml phosphate buffer in 1980. concentration was applied to 18 leaf scars per tree. kept in an ice bath during the inoculation period. Each Inoculum was Inoculated spurs were collected during bloom on 15 May 1980 and 14 May 1981, and individually frozen at -20 C. Isolations were made by dicing the buds of each spur with a sterile razor blade and placing the pieces in 10ml phosphate buffer. was plated on KBrc. After 1 hr, the suspension was mixed and 0.1 ml The presence of rifampicin-resistant strains was 37 recorded after 5-days incubation at 22 C and the percentage of infected leaf scars was calculated for each replication. Monitoring bacterial populations. Populations of PsmR associated with Montmorency sour cherry leaves were monitored in East Lansing. A single 12-yr-old Montmorency sour cherry tree was spray-inoculated on 17, 29 April, 8, 29 May 1980; 15, 22, 29 April, and 7 June 1981 with g 1.5-L phosphate buffer containing 10 2 sprayer operated at 28 kg/cm . sunset. cfu PsmR/ml using a handgun All spray inoculations were applied at Leaves were collected eight and 18 times from 4 June to 23 September 1980 and from 23 May to 10 October 1981, respectively (Figure 2.1). The tree was divided into quadrants and 10 leaves were collected at random from each quadrant. The 40 leaves were measured with an area meter before they were homogenized for 2 min in a blender with 300-ml phosphate buffer. The homogenate was serially diluted in phosphate buffer and duplicate 0.1-ml subsamples were plated on KBrc. Colonies were counted after 5-days incubation at 22 C and the cfu of PsmR per cm 2 of leaf were calculated. Populations of PsmR from individual leaves on terminal shoots were determined in 1981. Two 13-yr-old Montmorency sour cherry trees were spray-inoculated for the first time in 1981 as described for the single tree. Populations associated with leaves were determined on 26 May, 2, 17, 24 June, and 2 July. Leaves on six to 10 terminal shoots were harvested, measured with an area meter, and homogenized individually for 2 min in a blender with 50-ml phosphate buffer. The homogenate was serially diluted in phosphate buffer and duplicate 0.1ml subsamples plated on KBrc. Colonies were counted after 5-days 38 incubation at 22 C and the cfu of PsmR per cm 2 of leaf were calculated. In 1980 and 1981, one and three sequential rain samplers (see Chapter 1), respectively, were placed under the drip line of the inoculated tree used for monitoring bacterial populations on leaves. From each 5- to 8-ml sample of rainwater, duplicate 0.1-ml subsamples were plated on KBrc within 12 hr after each rain ended. Colonies were counted after 5-days incubation at 22 C and the cfu of PsmR per milliliter of rainwater were determined. Environmental monitoring. Air temperature was recorded with a 7-day recording hygrothermograph (Bendix Co. Inc., Baltimore, MD 21204) placed in a weather shelter 2 m above the ground in the orchard. Calibration of the hygrothermograph was checked weekly with a sling psychrometer. Rainfall was measured with a tipping bucket rain gauge (WEATHERtronics, Inc., West Sacramento, CA 95691) and recorded with an event recorder or microcomputer (9). Average populations of PsmR in rainwater for each rain period and populations of PsmR associated with leaves were separately correlated with daily average air temperatures for the preceding 10 days. Data analyses. All populations of PsmR in rainwater and associated with leaves were transformed by adding 1 to each value before calculating the l o g ^ . All data were analyzed with the Stat 4 statistical program (1) and Cyber 750 computer (Control Data Corp., Minneapolis, MN 55440). 39 RESULTS Overwintering on ground cover. Following inoculation of the orchard ground cover on 15 November 1979, populations of PssR and PsmR were high (4.21 to 4.47 l°g^Q cfu/g ground cover) on 18 January (Table 2.1). Fewer bacteria were recovered on 11 March and 17 April, and no bacteria were recovered on 16 May, 23 June, and 24 September 1980. Following a second inoculation on 2 November 1980, populations of PssR were high while populations of PsmR were quite low on 13 November. Both strains were recovered from the ground cover on 16 December 1980, only PssR was recovered on 8 March, and neither PssR or PsmR were recovered on 28 May 1981. Neither PsmR or PssR were recovered from Montmorency sour cherry leaves collected on 23 June 1980 and 8 June 1981. Overwintering in leaf scars. In the spring of 1980 and 1981, PsmR and PssR were recovered from buds of spurs inoculated in the autumn of 1979 and 1980, respectively (Table 2.2). In both years, recovery of PsmR and PssR from buds increased as inoculum concentrations were increased. Recovery of PssR from spurs inoculated 5 October 1979 was higher than from spurs inoculated 21 September 1979, but the recovery of PsmR did not differ significantly between inoculation dates. In 1980, the recovery of PsmR and PssR was significantly higher for spurs inoculated 8 October than for those inoculated 19 August. Populations associated with leaves. Through the 1980 and 1981 growing seasons, populations of PsmR associated with symptomless 2 leaves were about 4 logjo cfu/cm leaf (Figure 2.1). In 1980 and 2 1981, populations above 5 1°8 jq cfu/cm leaf were recorded only once 40 (24 June 1980); populations below 3 1°8 jq cfu/cm twice (27 June and 13 August 1981). 2 leaf were recorded Fluctuations in air temperature did not appear to significantly alter populations of PsmR associated with leaves. Populations of PsmR associated with individual leaves of terminal shoots were determined in 1981. On 2, 17, 24 June, and 2 July newly emerged leaves had significantly (P^ = 0.05) lower populations of PsmR than older leaves (Table 2.3). Usually, no PsmR were isolated from the youngest leaves, but populations of 2 to nearly 3 1°8 j q cfu/cm 2 leaf were recovered from the older basal leaves. Populations in rainwater. In 1980, populations of PsmR in rainwater collected from 29 May to mid-June were 1.3 to 3.7 l°8ig cfu/ml rainwater and from 1 July to mid-August were 0.0 to 0.7 1° S jq cfu/ml rainwater (Figure 2.1A). Populations of PsmR increased to 3.1 l o g ^ cfu/ml rainwater on 31 August, but on the following day populations were much lower (0.5 1°8^q cfu/ml rainwater). Populations of PsmR increased from 0.5 1°8 j q cfu/ml rainwater on 1 September to 3.0 1°8 j q cfu/ml rainwater on 13 and 17 October. Prior to leaf fall on 10 November, populations of PsmR dropped to 1.9 1°8 j q cfu/ml rainwater on 3 and 8 November. In 1981, the highest populations of PsmR in rainwater, 3.5 i°8^o cfu/ml rainwater, were recorded on 10 May, 2 days before full bloom (Figure 2.IB). From 24 May to 13 June, populations of PsmR were 1.6 to 2.9 log j q cfu/ml rainwater. On 8 June, populations of PsmR may have been higher than normal because the tree was inoculated on 7 June. From 22 June to 22 September, populations of PsmR in rainwater 41 were 0.0 to 1.9 1°S} q cfu/ml; on 25 September they increased to 2.6 logjg cfu/ml and in autumn, populations peaked at 3.0 21 October. cfu/ml on Populations of PsmR decreased to 0.3 L°S j q cfu/ml rainwater on 5 November, 3 days before leaf fall. Correlation coefficients between populations of PsmR in rainwater and average daily temperatures for the preceding 10 days were -0.5436 and -0.5096 (P ■ 0.01) for 1980 and 1981, respectively, indicating that as air temperature increased, populations of PsmR in rainwater decreased and vice versa (Figure 2.1). Dispersal patterns of bacteria in rainwater. Populations of PsmR in serially collected rain fractions revealed two types of dispersal patterns (Figure 2.2). In a type 1 pattern, populations of PsmR increased during the early stages of a rain period, then leveled off. Rains resulting in a type 1 dispersal pattern were preceded within 5 days by rain. In a type 2 pattern, populations of PsmR were high in the first fractions, declined and leveled off in the later fractions. Rains resulting in type 2 patterns were not preceded within 5 days by rain. When five or more fractions were collected and PsmR were isolated, the dispersal pattern was classified. Of 52 rain periods in 1980 and 1981, 24 were not classified because no bacteria were isolated (six periods) or less than five fractions were collected (18 periods). Of 21 rain periods classified as type 1, 20 were preceded within 5 days by rain. None of the seven rain periods classified as type 2 were preceded within 5 days by rain. In 1981, dispersal patterns did not differ between the three samplers. 42 DISCUSSION Both PsmR and PssR overwintered in Montmorency sour cherry buds of fruit spurs following leaf scar inoculations in autumn. As reported in England (5), infection through leaf scars with PsmR was greater in late autumn than in early autumn. In this study, the incidence of infection through leaf scars by £. syringae pv. syringae and P^. syringae pv. morsprunorum were similar, while Crosse and Garrett (7) found the incidence of infection with a strain of JP. syringae pv. syringae from Oregon, USA was less than with a strain of P. syringae pv. morsprunorum from England. These differences in efficiency of leaf scar infection between P^. syringae pv. syringae and I>. syringae pv. morsprunorum may be caused by differences in the. way infection was determined. Crosse and Garrett (7) determined infection by visual observation of symptoms, while we isolated the pathogen. Through the winter and early spring, PsmR and PssR were recovered from plants in the ground cover but not in late spring or summer, and were not disseminated up into the Montmorency sour cherry tree. These results indicate PsmR and PssR failed to become established on new ground cover growth in spring, probably because the strains were not pathogenic to any of the available plant species. Latorre and Jones (12) suggested P^. syringae pv. morsprunorum did not overwinter in the ground cover because its host range was more limited than that of P^. syringae pv. syringae. In this study, populations of PsmR associated with Montmorency leaves were relatively constant through the growing season; while in Poland (2, 14), populations of syringae pv. morsprunorum fluctuated during the growing season on Nefris, North Star, and Schattenmorele 43 sour cherry as well as on sweet cherry. These differences may be caused by sampling technique; in other studies (2, 14), populations were determined from leaf washings and expressed as cfu per leaf, while in this study total populations asssociated with leaves were expressed as cfu per cm 2 of leaf. Recovery of PsmR from newly emerged Montmorency sour cherry leaves was infrequent, while PsmR was always recovered from older fully expanded leaves. Therefore, PsmR was either washed by rainwater from older infected leaves to newly emerged leaves or newly emerged leaves were infected but the bacteria were not detectable with our isolation procedure. Populations of PsmR in rainwater collected under the drip line of a Montmorency sour cherry tree correlated negatively with air temperature. Cool air temperatures and high populations of PsmR in rainwater correspond with infection of leaves and leaf scars in spring and autumn, respectively. Schimdle and Zeller (13) demonstrated that at 100% relative humidity, air temperatures between 15 and 20 C were optimal for infection and disease development of P_. syringae pv. morsprunorum on sour cherry. In this study, populations of PsmR associated with leaves were not altered by air temperature; but air temperature may have affected the survival of PsmR in rainwater or release of PsmR from the leaves. The two dispersal patterns of PsmR in rainwater were dependent on the release of PsmR from Montmorency sour cherry leaves. When rains were more than 5 days apart, PsmR probably built up on or near the leaf surface and were released in high numbers with the onset of rain. Dew periods may have contributed to this buildup by providing conditions suitable for egress of bacteria through stomata onto leaves and for growth of bacteria on the leaf surface. 45 LITERATURE CITED 1. Anonymous. 1977. Guide VIII. Michigan State University STAT System: User's Computer Laboratory, Michigan State University, East Lansing 48824. 2. Burkowicz, A. 1981. Population dynamics of epiphytic Pseudomonas morsprunorum on the leaf surfaces of two sweet cherry cultivars. 3. Fruit Sci. Reports 8:37-48. Cameron, H. R. 1962. Mode of infection of sweet cherry by Pseudomonas syringae. Phytopathology 52:917-921. 4. Crosse, J. E. 1956. Bacterial canker of Leaf scar infection of cherry, 5. Crosse, J. E. 1957. stone-fruits. II. J. Hortic. Sci. 31:212-224. Bacterial canker of stone-fruits. III. Inoculum concentration and time of inoculation in relation to leaf-scar infection of cherry. 6. Crosse, J. E. 1966. Ann. Appl. Biol. 45:19-35. Epidemiological relations of the pseudomonad pathogens of deciduous fruit trees. Annu. Rev. Phytopathol. 4:291-310. 7. Crosse, J. E . , and Garrett, C. M. E. stone-fruits. VII. Hignett, R. C. 1974. Ann. Appl. Biol. 58:31-41. Absorption of Pseudomonas morsprunorum through cherry leaf scars. 9. Bacterial canker of Infection experiments with Pseudomonas mors-prunorum and I*, sryingae. 8. 1966. J. Gen. Microbiol. 80:501-506. Jones, A. L., and Fisher, P. D. 1980. Instrumentation for in-field disease prediction and fungicide timing. 2:215-218. Frotec. Ecol. 46 10. King, E. 0., Ward, M. K., and Raney, D. E. 1954. Two simple media for the demonstration of pyocyanin and fluorescin. J. Lab. Clin. Med. 44:301-307. 11. Latorre, B. A., and Jones, A. L. 1979. Pseudomonas morsprunorum, the cause of bacterial canker of sour cherry in Michigan, and its epiphytic association with _P. syringae. Phytopathology 69:335-339. 12. Latorre, B. A., and Jones, A. L. 1979. Evaluation of weeds and plant refuse as potential sources of inoculum of Pseudomonas syringae in bacterial canker of cherry. Phytopathology 69:1122- 1125. 13. Schmidle, A., and Zeller, W. 1976. Der EinfluB von Temperatur und Luftfeuchte auf die Infektion von Pseudomonas spp. bei BlMttern von Sauerkirsche (Prunus cerasus). Phytopathol. Z. 87:274-283. 14. Sobiczewski, P. 1978. Epiphytic populations of pathogenic pseudomonads on sour cherry leaves. Int. Conf. Plant Path. Bact. Angers. Pages 753-762 in: Proc. IVth 47 Table 2.1. Recovery of rifampicin-resistant Pseudomonas syringae pv. morsprunorum (PsmR) and £. syringae pv. syringae (PssR) from plants in the orchard ground cover following inoculation in the autumns of 1979 • and 1980. Recovery on KBrcx (log^Q[colony-forming units + 1/g ground cover]) Sampling dates______________________PsmR___________________ PssR (Inoculated 15 Nov 1979) y 18 Jan 1980 4.21 + 0.27Z 4.47 + 0.34 11 Mar 1980 3.96 + 0.54 4.27 + 0.55 17 Apr 1980 1.38 + 0.17 2.52 + 1.49 16 May 1980 0.00 + 0.00 0.00 + 0.00 23 Jun 1980 0;00 + 0.00 0.00 + 0.00 24 Sep 1980 0.00 + 0.00 0.00 + 0.00 13 Nov 1980 0.22 + 0.38 3.64 + 0.25 16 Dec 1980 1.62 + 0.76 3.74 + 0.29 8 Mar 1981 0.00 + 0.00 1.29 + 0.85 28 May 1981 0.00 + 0.00 0.00 + 0.00 (Inoculated 2 Nov 1980)y KBrc ■ A modified King's medium B amended with 50 pg/ml rifampicin and 15 Mg/ml cycloheximide. v # 8 JFor each pathovar a plot was inoculated with 2 L of 10 colonyforming units/ml phosphate buffer and each plot contained two 30-cm wide by 9-m long areas separated by a row of three Montmorency sour cherry trees. Mean of four plots followed by the standard error of the mean. 48 Table 2.2. Percent recovery of rifampicin-resistant Pseudomonas syringae pv. morsprunorum (PsmR) and P^. syringae pv. syringae (PssR) from buds on fruit spurs of Montmorency sour cherry in spring following leaf scar inoculations the previous autumn. Inoculum concentrations (cfu/ml buffer)^ PsmR PssR Inoculation date 102 106 47± 19 — 0 7 14 7+11 48+28 - 12 24 54 30+22 - - 6+10 15+10 34+26 108 Means* 56' 61 60 78 - 25 5 Oct 1979 - 5 - 104 106 21 Sep 1979 Means 102 104 15+13 58+16 70+19 108 Means - 19 Aug 1980 2 24 62 - 29+25 2 9 55 - 22+26 11 Sep 1980 2 34 64 - 34+_26 2 25 65 - 3H28 8 Oct 1980 2 61 80 - 48+35 2 16 71 - 30+32 _ — Means 2+3 40+19 69+11 2+3 16+15 64+15 _ ^cfu ■ Colony-forming units. Each value is the mean of three replications each consisting of 12 to 18 leaf scar inoculations. Buds were collected on 15 May 1980 and 14 May 1981, and infection was established by isolating on Ring's medium B amended with 50 jig/ml rifampicin and 25 jig/ml cycloheximide. 2 Values are the mean and standard error of the mean. 49 Table 2.3. Populations of rifampicin-resistant Pseudomonas syringae pv. morsprunorum on individual leaves taken from terminal shoots of Montmorency sour cherry trees in 1981. Sampling date Leaf* 26 May 17 June 24 June 2 July 2 2.03(6)yaz 2.24(6) a 2.16(10) a 2.86(10) a 2.93(10) a 4 1.19(6) a 1.30(6) a 1.67(10) a 2.54(10) a 2.82(10) a 6 1.62(5) a 0.78(6) a 1.00(10) ab 2.01(10) ab 2.18(10) ab 0.00(3) b 1.51(10) a 1.42(10) b 1.93(10) ab b 0.37(10) c 1.24(9) b 0.17(8) 2.04(5) ab 8 - 10 - 12 X 2 June _ 0.00(9) — c Leaves are numbered starting from the base of the shoot. yMean of the log.n (colony-forming units + 1 per cm^ leaf) followed by the number (in parentheses) of leaves sampled. Bacteria were recovered on a modified King's medium B amended with 50 jig/ml rifampicin and 25 >ig/ml cycloheximide. 2 Values in a column followed by the same letter do not differ significantly (P ■ 0.05) according to Duncan's multiple range test. 50 o 14 Rain wafer Harvest Bloom 29 26 Jun May 30 24 Aug Jul Sep Nov Oct 67 92 (=ul4 L eaves .Rain water. Bloom Harvest 24 May Jul Auo Sep Oct Nov Figure 2.1. Recovery of rifampicin-resistant Pseudomonas syringae pv. morsprunorum (PsmR) from leaves and runoff rainwater from a Montmorency sour cherry tree in relation to mean daily air temperatures and rainfall. Populations were expressed as log (colony-forming units of PsmR + 1 per milliliter rainwater) or per cm leaf. A, 1980; B, 1981. 51 3- 2 - Type (/> Type 2 2 4 6 8 10 12 14 16 Collection tube number Figure 2.2. Dispersal patterns for rifampicin-resistant Pseudomonas syringae pv. morsprunorum (PsmR) in runoff rainwater collected serially under a Montmorency sour cherry tree during rains on 14 (type 2) and 15 (type 1) June 1980. CHAPTER 3 Reduceion of Pseudomonas syringae pv. morsprunorum on Montmorency Sour Cherry with Copper and Dynamics of the Copper Residues ABSTRACT A rifampicin-resistant strain of _P. syringae pv. morsprunorum (PsmR) was used to study the effectiveness of tribasic copper sulfate (TBS) and Citcop 4E for reducing populations of PsmR on Montmorency sour cherry trees in spring and early summer. Populations of PsmR were reduced more by 636 and 949 mg/L TBS than by 200 mg/L Citcop 4E, but several applications were needed to reduce the populations to a low level. Citcop 4E at 200 mg/L was more phytotoxic to cherry foliage than TBS at 636 mg/L. Phytotoxicity was related to the number of applications and was not reduced by adding hydrated lime to the copper treatments. The decline of copper residues from leaves was related by multiple regression analyses with rainfall and initial level of copper on leaves. INTRODUCTION Bacterial canker, caused by Pseudomonas syringae pv. morsprunorum, was first observed on Prunus cerasus L. cv. Montmorency sour cherry in Michigan in 1976 (13) and no effective control measures have been tested. In Oregon (2), application of copper compounds have been recommended in autumn and just before bud break to control 52 53 infection of dormant buds on sweet cherry. In England (15), Bordeaux mixture applied to sweet cherry trees at white bud stage and again at petal fall reduced the leaf spot phase of the disease, but the petal fall spray was highly phytotoxic to the leaves. A single spray of Bordeaux mixture at white bud had no significant effect on the leaf spot phase (8, 15), but sprays of streptomycin at full bloom and at 75 and 100% petal fall reduced leaf spot symptoms by 93 to 96% (6). In California (4) and Connecticut (17), blossom blast of pears, caused by P. syringae pv. syringae, was reduced by a series of streptomycin or fixed copper treatments before, during, and after bloom. These studies suggest that repeated spray treatments in spring may control leaf, blossom, and fruit infections by JP. syringae pv. morsprunorum in Michigan. The objectives of this study were to investigate the effectiveness of fixed copper compounds for reducing populations of JP. syringae pv. morsprunorum from early bud break through early summer on Montmorency sour cherry and to examine the retention of copper on the foliage. Copper compounds were evaluated rather than streptomycin because copper compounds have activity against some important fungal pathogens of cherry, and registration of these compounds by regulatory agencies was more likely than registration of streptomycin. MATERIALS AND METHODS Field inoculations. The rifampicin-resistant strain of P. syringae pv. morsprunorum (PsmR) selected by Latorre and Jonhs (14) was used in all experiments. Inoculum grown on King's medium B (12) for 2 days at 22 C was suspended in 10 ml of 0.01-M phosphate buffer 54 (pH 7.2) by stirring with a sterile glass rod. concentrations of 10 Inoculum 8 colony-forming units (cfu)/ml phosphate buffer were obtained by adjusting the turbidity of suspensions to 0.04 absorbance with a spectrophotometer (Spectronic 20, Bausch and Lomb, Rochester, NY 14625) set at 625 nm. Montmorency sour cherry trees in East Lansing, Michigan were spray-inoculated on 17, 29 April, 8, 29 May 1980; 15, 22, 29 April, and 7 June 1981. All trees were 2 inoculated at sunset with a handgun sprayer (28 kg/cm ) until leaves were lightly wet. The amount of inoculum applied to each tree was 0.2 to 1.5 L depending on tree size. Spray trials. On inoculated trees, tribasic copper sulfate 53W (TBS) and Citcop 4E (48% copper salts of fatty and rosin acids; Cities Service Co., Atlanta, GA 30302) were evaluated along with streptomycin sulfate 21.2W (Pfizer Inc., New York, NY 10017). Concentrations of spray solutions containing copper were based on the amount of copper salt per liter. Hydrated lime [CaCOH^] was evaluated for reducing phytotoxicity from copper sprays. Because of captafol's (Difolatan 4F, Chevron Chemical Co., Richmond, CA 94804) persistence when applied to apple trees (16), the possibility it may improve the retention of copper on cherry leaves was evaluated. All treatments (Tables 3.1 and 3.2) were applied to runoff with a handgun sprayer 2 operated at 28 kg/cm . Application dates were 21 April, 2, 27 May, June, 1, 14, 29 July 1980; 17, 27 April, and 25 June 1981. 12 13, 21, 27 May, 3, 11, 17, Treatments were applied to single-tree plots arranged in a randomized complete block design according to tree size. Each treatment was replicated five times. 55 In 1981, defoliation was rated by visually estimating the percentage of fallen leaves per replication on 6 and 8 July, 11 and 13 days, respectively, after the final copper treatment on 25 June. Concentrations of copper on leaves were measured on 8 July, 3 days after the onset of defoliation. Five terminal shoots, each with 12 intact leaves, were selected at random from each replication. Starting from the base of each shoot, leaves at nodes 1, 3, 6, 8, and 11 were harvested and pooled by node number. The leaves were measured with an area meter (Model LI-3000, LAMBDA Instruments Corp., Lincoln, MB 68504) and copper concentrations were measured with a plasma emission spectrometer described later. Leaf age was estimated using a degree-day leaf emergence model developed .for Montmorency sour cherry (10). Temperatures were recorded with a hygrothermograph (Bendix Co. Inc., Baltimore, MD 21204) placed in a weather shelter 1.5 m above the orchard floor. The number of copper sprays applied to each leaf was estimated based on the approximate date of leaf emergence. In East Lansing, I studied the loss of copper from leaves on Montmorency sour cherry trees treated once with 6 g/L TBS + 12 g/L hydrated lime and from trees treated four times with 2.4 g/L TBS with and without 4.8 g/L hydrated lime. Treatments were applied to 30-yr- old trees and each treatment was replicated four times in a randomized complete block design. Treatments were applied as described earlier. Each replicate contained two or three trees. The high rate of TBS was applied on 18 September 1981 and the lower rate was applied on 18, 28 September; 8 and 19 October 1981. Monitoring bacterial populations. Recovery of bacteria from plant material sprayed with copper compounds can be difficult when 56 using standard isolation procedures. In 1978, Young (20) reported that the bactericidal activity of copper was inactivated with phosphate buffer. To verify this report, we determined the survival of PsmR in distilled water and phosphate buffer with and without TBS + hydrated lime. Treatment solutions of phosphate buffer (13.6 mg/ml KH^PO^ + 14.2 mg/ml Na^HPO^) and distilled water were amended with 18.7 mg/ml TBS + 39.6 mg/ml hydrated lime and then diluted 10- and 100-fold with phosphate buffer and distilled water, respectively. stock solution containing approximately 10 water was divided into 100-ml aliquots. 4 A cfu PsmR/ml distilled Aliquots of stock solution were mixed with 10-ml aliquots of treatment solutions to give the treatments listed in Table 3.3. After 10 min, duplicate 0.1-ml subsamples from each replication were plated onto King's medium B. Colonies were counted after 5-days incubation at 22 C. Each treatment was replicated five times. Populations of PsmR were monitored on each inoculated single-tree replicate. Each tree was divided into quadrants and 10 fully expanded leaves were randomly chosen from each quadrant. The leaves were measured with an area meter before each 40-leaf sample was homogenized for 2 min with 300-ml phosphate buffer in a blender (Waring Products Inc., New Hartford, CT 06057). The homogenate was serially diluted in phosphate buffer and duplicate 0,1-ml subsamples were pipetted onto a modified King's medium B amended with 50 /ig/ml rifampicin (CalbiochemBehring Corp., La Jolla, CA 92037) and 25 ^ig/ml cycloheximide (Sigma Chemical Co., St. Louis, MO 63178). Colonies were counted after 5-days incubation at 22 C and the cfu of PsmR per cm determined. 2 of leaf were 57 Monitoring copper deposits on leaves. To determine the rate of copper loss from Montmorency leaves, leaf samples were collected after treatments were applied. Samples were taken on 2, 10 June, 1, 7, 14, 29 July, 6, 12 August 1980; 2 June, 8 July, and 10 September 1981. In autumn of 1981, samples were collected on 18, 23, 28 September, 3, 8, 13, 19, 24, and 29 October. To determine concentration of copper on leaves, each replicate was divided into quadrants and 10 leaves were randomly collected per quadrant. The quantity of copper per cm for each replication by cutting 1.5-cm all 40 leaves with a cork borer. 2 2 of leaf area was determined leaf discs from the center of The discs were dried at 72 C for at least 18 hr, weighed, and ground to a fine powder with a mortar and pestle. For each replication, 100 mg of leaf powder was digested for 18 hr at 72 C with 2 ml of "Baker Instra-Analyzed" nitric acid (J. T. Baker Chemical Co., Philipsburg, NJ 08865) in a 10-ml screw-cap Teflon vial. Using a volumetric flask, each digested replication was increased to 10 ml with double-glass-distilled water filtered through a Milii-Q water purification system (Millipore Corp., Bedford, MA 01730). Each diluted sample was poured into a 16xl25-mm culture tube with a screw cap and held at room temperature for 24 hr until precipitates formed. Each sample was centrifuged at 150 £ for 30 min and the supernatant was decanted into a clean culture tube. Digested samples were analyzed for copper using a plasma emission spectrometer (Jarrel-Ash Model 955 Plasma Atom Comp.; Fisher Scientific, Waltham, MA 02254) with the inductively coupled plasma operated at 1.1-kw forward power, 1.6-mm flame height, 0.84 to 0.98 2 kg/cm nebulizer pressure, and a sample flow rate of 1.4 ml/min. The 58 spectrometer was standardized with a 10 pg/ml stock solution of copper in 20% "Baker Instra-Analyzed" nitric acid prepared by serially diluting a 1,000 pg/ml copper atomic spectral standard solution (J. T. Baker- Chemical Co., Philipsburg, NJ 08865). The results from the spectrometer were used to determine the micrograms of copper per cm of leaf. All vials, analyses were washed 2 glassware, and utensils used in the copper in 24% nitric acid (commercial grade) for 30min before rinsing with distilled and double-glass-distilled Milli-Q filtered water. Blank samples were processed repeatedly to monitor for copper in reagents and on glassware. The relationship between copper loss from the leaves and rainfall was established using stepwise regression (9). The amount of rainfall and the difference in the level of copper on leaves between two sampling periods were computed. Rainfall was measured with a tipping bucket rain gauge (WEATHERtronics Inc., West Sacramento, CA 95691) and recorded with an event recorder or a microcomputer-based instrument (11). Citcop 4E and TBS treatments were analyzed separately using a second degree polynomial of the general form: CL - b Q + bjC + b 2R + bjRC + b 4C2 + bjR2 + bg(RC)2 + e 2 •f-f where CL = the difference in the level of copper (pg Cu per cm leaf) on leaves between two sampling periods, R « rainfall in milliliter between sampling periods, C ■ the level of copper (pg per 2 cm leaf) on leaves for the first of the two sampling periods, b's ■ partial regression coefficients, and e is a normally distributed 59 2 random variable with mean zero and variance tf . the final equation were significant at P ■ 0.01. Factors included in Only the regression models having the best combination of high coefficients of determination and residuals supporting the assumptions that errors were independent and normally distributed (see Appendix 2) were retained (9). The data were plotted using a computer program designed to plot randomly selected three-dimensional data points (19). General data analyses. Data collected sequentially during the growing season were analyzed as a split-plot with time (18). Treatment differences in most experiments were analyzed using planned paired comparisons. All data were analyzed with the Stat 4 statistical program (3) and Cyber 750 computer (Control Data Corp., Minneapolis, MN 55440). Unless stated otherwise, differences were significant at the 1? * 0.01 level. RESULTS Copper phytotoxicity. In 1980, leaf chlorosis and defoliation were not observed in any copper treatment. In 1981, copper-treated trees had significantly greater defoliation than nonsprayed, streptomycin-, and captafol-sprayed trees (Table 3.4). Trees sprayed with Citcop 4E had significantly more defoliation than trees sprayed with TBS. Hydrated lime and captafol did not reduce the level of defoliation in Leaves at to show injury the TBS and Citcop 4E treatments. the base of terminal and lateral shoots from copper sprays. more leaves showing injury were the first Trees sprayed with Citcop 4E had than trees sprayed with TBS (Table3.5). 60 On 8 July, the amount of copper on leaves sprayed with TBS or Citcop 4E was significantly (P^ * 0.05) higher on leaves that emerged on 11 May than emerged on 26 June (Table 3.5). The approximate threshold concentration of copper that resulted in phytotoxicity was 8.3 and 1.4 jig Cu ++ /cm 2 of leaf for TBS and Citcop 4E, respectively. These leaves received seven and four applications of TBS and Citcop 4E, respectively. Recovery of bacteria from buffered copper solutions. High counts of PsmR (278 to 372 cfu/ml) were recovered from all phosphate buffer solutions containing TBS (Table 3.3). Only a few PsmR (0 to 3 cfu/ml) were recovered from distilled water amended with 0.017 mg/ml TBS + 0.036 mg/ml hydrated lime. Recovery of PsmR from the phosphate buffer + TBS treatments and from phosphate buffer alone were similar and significantly greater than recovery of PsmR from distilled water alone. Effect of copper on bacterial populations. In 1980, trees treated with copper had significantly lower populations of PsmR than trees treated with captafol (Table 3.1). Among the copper treatments, trees treated with TBS had significantly fewer PsmR than trees treated with Citcop 4E, and trees treated with 200 mg/L Citcop 4E had significantly fewer PsmR than trees treated with 100 mg/L. Following the last inoculation on 29 May, populations of PsmR declined steadily as four applications of copper were applied (Figure 3.1A). By 31 July, populations of PsmR on trees treated with TBS and Citcop 4E (200 mg/ml) were 98.9 and 95.0% less, respectively, than those on captafol-treated trees. The decline in populations of PsmR from 4 June to 10 June in the TBS treatment occurred despite the 61 development of leaf symptoms of bacterial canker starting 10 June. However, leaf symptoms were not severe enough to detect differences between treatments. In 1981, trees treated with copper and streptomycin had significantly fewer PsmR than nontreated and captafol-treated trees (Table 3.2). Streptomycin-treated trees had populations of PsmR similar to trees treated with TBS or Citcop 4E. Trees treated with TBS had significantly fewer PsmR than trees treated with Citcop 4E. Except on 20 June, populations of PsmR were lower on TBS-treated trees than on nontreated or captafol-treated trees (Figure 3.IB). On 29 June and 10 September, populations of PsmR on the TBS-treated trees were 94 and >99% less than their respective check treatments. Populations of PsmR on trees treated with 200 mg/ml Citcop 4E were below populations on nonsprayed trees except on 14 July (data not shown). Populations of PsmR were not affected by adding hydrated lime or captafol to the copper treatments in 1980 and 1981. No symptoms of bacterial canker developed in 1981. Dynamics of copper deposition. In autumn or 1981, the amount of copper loss for trees treated with 2.4 g/L TBS with and without hydrated lime were 28 and 27%, respectively; therefore, data from these two treatments were combined. Leaves from trees sprayed with 6 g/L TBS had 9.5 and 6.0 jig more copper per cm 2 leaf on 18 and 23 September, respectively, than trees sprayed with 2.4 g/L TBS (Figure 3.2). On 28 September, the concentrations of copper on leaves retreated with 2.4 g/L TBS increased from 3 to 10 jig/cm leaf, and was 3 jig higher than the concentration remaining in the 6 g/L TBS treatment. Copper concentrations dropped dramatically between 18 and 62 23 September and between 28 September and 3 October because of 40- and 89-mm rain on 21 and 30 September, respectively. On 3 October, copper 2 concentrations for the 2.4 and 6.0 g/L TBS treatments were 3 jig/cm leaf and concentrations in the latter treatment remained relatively constant. On 8 October, the 2.4 g/L TBS treatment was sprayed a third 2 time and the concentration of copper increased to 10.2 jig/cm leaf, remaining at this concentration through 13 October due to lack of rain. But after 15 mm of rain, the concentration of copper dropped 4.8 jig/cm2 leaf by 19 October. to On this date, the 2.4 g/L TBS treatments were sprayed a fourth time, and concentrations of copper increased to 9.5 ;ig/cm 2 2 leaf but dropped to 6.1 .pg/cm 24 October, following a 8-mm rain. leaf on On 27 October, 5-mm rain had no effect on the concentration of copper. Copper residue data for Citcop 4E in 1980 and 1981 were analyzed using stepwise regression to determine if the loss of copper (CL) from leaves was a function of rainfall (R) and initial concentration of copper (C) on leaves. The data included 225 points and the resulting model w a s : CL - - 7.6827 x 10~3 + 1.6186 x 10"1C + 1.9529 x 10_2RC - 3.080 x 10_4R 2 The R o value was 0.823. copper loss (pg Cu ++ A nomogram of the original data shows how per cm 2 leaf) increased as rainfall and concentration of the initial deposits of copper on leaves increased (Figure 3.3A). 63 A similar analysis of residue data for TCS in 1980 and 1981 included 194 points and the resulting model was: CL - - 1.6584 + 2.8058 + 5.5985 x 10-3RC x 10_1C + 1.2271 x l O ^ R - 1.167 x 10_3R 2 2 The R value was 0.9073. As rainfall amounts and the initial concentrations of copper on leaves increased, greater quantities of copper were lost from leaves (Figure 3.3B). There were no differences in copper loss between copper treatments with and without captafol or hydrated lime. DISCUSSION In the early 1950's, organic fungicides began to replace copper compounds for the control of cherry leaf spot and brown rot on Montmorency sour cherry in Michigan. eliminated from most cherry By 1970, copper compounds were disease control programs. Bacterial canker developed for the first time on Montmorency sour cherry in Michigan in 1976 (13) and has continued to be a significant but sporatic problem. In this study, repeated application of copper compounds reduced populations of JP. syringae pv. morsprunorum, the cause of bacterial canker in Michigan (13). This suggests that, prior to 1970, the frequent use of copper to control other diseases also controlled bacterial canker. Now that copper compounds are not used on cherry, the pathogen is no longer being suppressed. 64 Tribasic copper sulfate has several advantages over Citcop 4G for use on Montmorency sour cherry. Tribasic copper sulfate was more effective in reducing populations of PsmR than Citcop 4G, was less phytotoxic than Citcop 4E, and resulted in higher concentrations of copper on leaves than Citcop 4E. Unlike Bordeaux mixture, the addition of hydrated lime to TBS and Citcop 4E treatments did not reduce the phytotoxic effects of copper. This results indicate the following strategy should be used for reducing populations of £. syringae pv. morsprunorum on Montmorency sour cherry. Tribasic copper sulfate should be applied every 7 to 10 days starting at green tip bud stage to prevent buildup of populations of ]?. syringae pv. morsprunorum on emerging leaves and flower parts. Populations of PsmR were probably higher in the experimental orchard than populations of JP. syringae pv. morsprunorum in commercial orchards because we made several artifical inoculations. Therefore, populations of P^. syringae pv. morsprunorum in commercial orchards should be reduced to levels below those recorded in this study. As reported by other workers (6, 15), repeated applications were more effective than single applications for bacterial canker disease control. However, applications should be discontinued after shuck-fall bud stage to avoid phytotoxicity. In England (5), leaf scars of sweet cherry infected with _P. syringae pv. morsprunorum in autumn developed cankers the following spring, and application of Bordeaux mixtures to sweet cherry trees in autumn decreased the development of cankers by 75 to 85% (6). In Michigan, Montmorency sour cherry was susceptible to leaf scar infection by £. syringae pv. morsprunorum and _P. syringae pv. syringae 65 in September and October (see Chapter 2). Therefore, repeated application of coppers in autumn should help control leaf scar infections, particularly during periods of frequent rainfall. Early defoliation from autumn treatments of Bordeaux mixture was a problem on sweet cherry (1); but in this study, defoliation by TBS applied in autumn was not observed on Montmorency sour cherry. This suggests autumn treatments with TBS may be safer than with Bordeaux mixture. In 1980, PsmR was recovered from Montmorency leaves even after repeated application of coppers in spring and summer. In laboratory studies, survival of PsmR in distilled water was dramatically reduced when amended with TBS + hydrated lime at concentrations similar to those applied to Montmorency sour cherry trees in 1980. This indicates PsmR may survive in or on leaf tissues in areas protected from copper being deposited on the leaf surface. It may also be possible that the mucoid substances referred to by Crosse (7) protect the bacteria from copper residues. 66 LITERATURE CITED 1. Allen, W. R., and Dirks, V. A. 1979. The use of rapeseed oil to reduce premature defoliation in sweet cherry sprayed with Bordeaux mixture for control of bacterial canker. Can. J. Plant Sci. 59:487-489. 2. Anonymous. 1976. Oregon Plant Disease Control Handbook. Extension Service, Oregon State University, Corvalis 97331. 3. Anonymous. 1977. Michigan State University STAT System: Guide VIII. User's Computer Laboratory, Michigan State University, East Lansing 48824. 4. Bethell, R. S., Ogawa, J. M., English, W. H . , Hansen, R. R . , Manji, B. T . , and Schick, F. J. 1977. sprays control pear blossom blast. 5. Crosse, J. E. 1957. Copper-streptomycin Calif. Agric. 31:7-9. Bacterial canker of stone-fruits. III. Inoculum concentration and time of inoculations in relation to leaf scar infection of cherry. 6. Crosse, J. E. 1957, canker of cherry. 7. Crosse, J. E. Ann. Appl. Biol. 45:19-35. Streptomycin in the control of bacterial Ann. Appl. Biol. 45:226-228. 1959. Bacterial canker of stone-fruits. IV. Investigation of a method for measuring the inoculum potential of cherry trees. 8. Crosse, J. E. Ann. Appl. Biol. 47:306-317. 1975. (Pseudomonas spp.). 9. Draper, Bacterial canker of stone fruits Rep. East Mailing Res. Stn. 1974:105-106. N. R . , and Smith, H. John Wiley and Sons, 1966. Applied Regression New York. 407 pp. Analysis. 67 10. Eisensmith, S. P., Jones, A. L., and Flore, J. A. 1980. Predicting leaf emergence of Montmorency sour cherry from degree-day accumulations. 11. J. Am. Soc. Hortic. Sci. 105:75-78. Jones, A. L., and Fisher, P. D. 1980. Instrumentation for in-field disease prediction and fungicide timing. Protec. Ecol. 2:215-218. 12. King, E. 0., Ward, M. K., and Raney, D. E. 1954. Two simple media for the demonstration of pyocyanin and fluorescein. J. Lab. Med. 44:301-307. 13. Latorre, B. A., and Jones, A. L. 1979. Pseudomonas morsprunorum, the cause of bacterial canker on sour cherry in Michigan, and its epiphytic association with P^. syringae. Phytopathology 69:335-339. 14. Latorre, B. A., and Jones, A. L. 1979. Evaluation of weeds and plant refuse as potential sources of inoculum of Pseudomonas syringae in bacterial canker of cherry. Phytopathology 69:1122-1125. 15. Montgomery, H. B. S., and Moore, M. H. 1945. bacterial canker and leaf-spot in sweet cherry. The control of J. Pomology 21:155-163. 16. Ross, R. G., and Gaul, S. 0. 1980. Persistence of captafol applied with foliar nutrients during fruit bud development of apple. 17. Can. J. Plant Pathol. 2:205-208. Sands, D. C., and McIntyre, J. L. 1977. Possible methods to control pear blast, caused by Pseudomonas syringae. Rep. 61:311-312. Plant Dis. 68 18. Steel, R. G. D., and Torrie, J. H. Procedures of Statistics. 2nd Ed. 1980. Principles and McGraw-Hill, New York. 633 pp. 19. Wittick, R. I. 1974. GEOSYS. A computer system for the description and analysis of spatial data. Computer Institute for the Social Sciences Research Technical Report 74-53. Department of Geography, Michigan State University, East Lansing 48824. 35 pp. 20. Young, J. M. 1978. Survival of bacteria on Prunus leaves. Pages 779-786 in: Proc. IVth Int. Conf. Plant Path. Bact. Angers. 69 Table 3.1. Effect of copper treatments on populations of rifampicinresistant Pseudomonas syringae pv. morsprunorum (PsmR) on Montmorency sour cherry leaves and the analysis of variance for a set of planned paired comparisons in 1980.______________________________________________ V Treatments and rates per liter PsmR recovered on KBrc 2 (log^[cfu-fl/cm leaf]) Captafol, 986 mg Tribasic copper sulfate, 949 mg + lime, 3.59 g Citcop 4E, 100 mg Citcop 4E, 100 mg + lime, 1.19 g Citcop 4E, 100 mg + captafol, 986 mg Citcop 4E, 200 mg Citcop 4E, 200 mg + lime, 1.19 g Citcop 4E, 200 mg + captafol, 986 mg Reduction in w bacteria (%) 3.05 + 1.26 0 + 1.54 + 1.23 + 1.05 7 0.98 + 1.39 + 1.22 7 1.29 94 56 40 48 90 91 74 1.84 2.69 2.83 2.77 2.06 2.02 2.46 ms .z E 1 1 1 15.6 13.6 20.0 10. 6** 9. 4** 13. 8** 1 1 1 1 4 28 1.1 1.4 0.1 2.5 7.2 1.5 Planned paired comparisons between treatments^ df Captafol vs copper Tribasic copper sulfate vs Citcop 4E Citcop 4E, 100 mg vs Citcop 4E, 200 mg (I) Citcop 4E alone vs Citcop 4e + lime and Citcop 4E + captafol (II) Citcop 4E + lime vs Citcop 4E + captafol (III) Interaction between comparisons I and II Interaction between comparisons I and III Blocks Error 0. 8 1. 0 0. 0 1. 7 5. 2* y KBrc ■ A modified K i n g ’s medium B amended with 50 jig/ml rifampicin and 25 jig/ml cycloheximide. Each value is the mean of five replications measured on eight sampling dates followed by the standard error of the mean. Percent reduction in bacteria is the difference in recovery of PsmR between the captafol (alone) treatment and each copper treatment divided by the captafol treatment and multiplied by 100. Lime * hydrated lime, C a t o H ^ . ^Planned paired comparisons between treatments were determined for PsmR recovered on KBrc. Z* = significant (P_ ■» 0.05), ** “ significant (]? = 0.01). 70 Table 3.2. Effect of copper treatments on populations of rifampicinresistant Pseudomonas syringae pv. morsprunorum (PsmR) on Montmorency sour cherry leaves and the analysis of variance for a set of planned paired comparisons in 1981.______________________________________________ V Treatments and rates per liter PsmR recovered on KBrc 2 (log^[cfu-t-l/cm leaf]) Check (no spray) Captafol, 491 mg Tribasic copper sulfate, 636 mg Tribasic copper sulfate, 636 mg + lime, 3.59 g Tribasic copper sulfate, 636 mg + captafol, 491 mg Citcop 4E, 200 mg Citcop 4E, 200 mg + lime, 3.59 g Citcop 4E, 200 mg + captafol, 491 mg Streptomycin sulfate, 121 mg . Reduction in w bacteria (%) 2.76 + 0.72 2.79 + 0.93 2.16 + 1.04 0 0 76 1.91 +_ 1.20 86 2.05 2.40 2.52 2.27 2.03 81 58 44 69 82 + + + + + • • y Planned paired comparisons between treatments Check vs captafol alone Ch>ck and captafol vs copper and streptomycin Copper vs streptomycin Tribasic copper sulfate vs Citcop 4E (I) Copper alone vs copper + lime and copper + captafol (II) Copper + lime vs copper + captafol (III) Interaction between comparisons I and II Interaction between comparisons I and III Blocks Error 1.02 0.74 0.91 1.12 1.08 ms FZ 1 1 1 1 0.0 37.2 2.1 13.1 0.0 28.3** 1.6 10.0** 1 1 1 1 4 32 0.8 0.2 0.8 2.5 12.5 1.3 0.2 0.2 0.6 1.9 9.5** df vKBrc - A modified King's medium B amended with 30 p g / m l rifampicin and 25 jig/ml cycloheximide. Each value is the mean of five replications measured on 15 sampling dates followed by the standard error of the mean. w Percent reduction in bacteria is the difference in recovery of PsmR between check treatments (mean of check and captafol treatments) and each copper treatment divided by check treatments and multiplied by 100. XLime * hydrated lime, C a C O H ^ . y m Planned paired comparisons between treatments were determined for PsmR recovered on K B r c . Z** « significant (£ * 0.01). 71 Table 3.3. Recovery of rifampicin-resistant Pseudomonas syringae pv. morsprunorum (PsmR) from distilled water and phosphate buffer each containing tribasic copper sulfate (TBS) + hydrated lime (lime). Concentration of amendments Suspending solution TBS (mg/ml) Lime (mg/ml) PsmR recovered on KBrcx (colony-forming units/ml) Distilled water 0.000 0.000 187 + 83 Distilled water 0.017 0.036 3 Distilled water 0.170 0.360 0 + 0 Distilled water 1.700 3.600 0 + 0 Phosphate buffer^ 0.000 0.000 294 + 80 Phosphate buffer 0.017 0.036 372 + 72 Phosphate buffer 0.170 0.360 278 + 56 Phosphate buffer 1.700 3.600 371 + 71 + 1 x KBrc * A modified King's medium B amended with 50 pg/ml rifampicin and 25 jig/ml cycloheximide. ^Phosphate buffer at 0.01 M (pH ■ 7.2). 2 Mean of five replications followed by the standard error of the mean. Using a set of planned paired comparisons between treatments, distilled water without TBS + hydrated lime vs all phosphate buffer treatments were significantly (JP ■ 0.01) different (ms ■ 80712, F “ 13.5, error df ” 16, and error ms ” 5968). 72 Table 3.4. Defoliation of Montmorency sour cherry trees from various copper treatments and the analysis of variance for a set of planned paired comparisons in 1981. Treatments and rates per liter Defoliation (%)X Check (no spray) Captafol, 491 mg Tribasic copper sulfate, 636 mg Tribasic copper sulfate, 636 mg + lime^ 3.59 g Tribasic copper sulfate, 636 mg + captafol, 491 mg Citcop 4E, 200 mg Citcop 4E, 200 mg + lime, 3.59 g Citcop 4E, 200 mg + captafol, 491 mg Streptomycin sulfate, 121 mg Planned paired comparisons between treatments Copper vs no copper Tribasic copper sulfate vs Citcop 4E (I) Copper alone vs copper + lime and copper + captafol (II) Copper + lime vs copper + captafol (III) Interaction between comparisons I and II Interaction between comparisons I and III Check vs captafol alone Captafol vs streptomycin sulfate Block Error 0.1 0.3 6.5 7.1 11.7 43.7 45.4 37.1 0.1 df + + + + + + + + +_ ms 1 1 1 12575 16951 0 1 1 1 1 1 4 32 35 96 419 0 0 243 331 0.3 0.5 6.5 8.7 9.2 30.8 32.6 26.6 0.3 _z F 38** 51** 0 0 0 1 0 0 1 ^ a c h value is the mean of five replications followed by the standard error of the mean. Data were recorded on 6 and 8 July. ^Lime ■ hydrated lime, CaCOH)^. z** - significant (£ ■ 0.01). 73 Table 3.5. The relationship of the number of copper sprays applied to Montmorency sour cherry leaves and residues of copper on the leaves to phytotoxicity. Leaf emergence Sprays (date)8 (no.)* „ ++, 2 , _u jig Cu /cm leaf CheckV z TBSW Citcop 4EX - + 1.4 b - - + 4.6 d 0.9 be - - - 2.6 e 0.6 c - - - 7.2 b 1.5 b 4 0.5 a 5.5 c 11 June 3 0.4 a 26 June 1 0.3 a 22 May 5 2 June Citcop 4E - 0.5 a 0.4 a TBS + 2.3 a 7 Check + 8.3 a 11 May Phytotoxicity^ Q Knowing the position of the leaf on the terminal, the date of leaf emergence was calculated with a Montmorency sour cherry leaf emergence degree-day model. ^Estimated number of sprays applied to each leaf. uEach value is the mean of five replications sampled on 8 July 1981. v Nonsprtyed. w TBS ■ Tribasic copper sulfate at 636 mg/L. xCitcop 4E at 200 mg/L. = chlorotic leaves observed, - ■ no chloretic leaves observed. Values within a column followed by the same letter do not differ significantly (£ * 0.05) according to Duncan's multiple range test. 74 datM 4 3 Captafol 22 Citcop 4E 100 mg .Citcop 4E 2 0 0 Inoculation datM -O TBS 30 Apr 20 23 25 Jul Jun May Aug S ep Spray date* C heck o 9 a Inoculation d a te s 30 Apr May 27 Jun 25 Jul 22 Aug Sep Figure 3.1. Populations of rifampicin-resist ant Pseudomonas syringae pv. morsprunorum (PsmR) on Montmorency sour cherry leaves sprayed with copper treatments. A, In 1980, treatments were 986 mg/L captafol, 949 mg/L tribasic copper sulfate (TBS), and 100 and 200 mg/L Citcop 4E. B, In 1981, treatments were 636 mg/L TBS and check treatments (nonsprayed and 491 mg/L captafol combined). 75 TBS (2.4 g) . Check. Figure 3.2. The relationship between rainfall and the loss of copper from Montmorency sour cherry leaves. Trees were sprayed on 18 September 1981 with 6 g/L tribasic copper sulfate (TBS) + 12 g/L hydrated lime and on 18, 28 September, 8, and 19 October with 2.4 g/L TBS with or without 4.8 g/L hydrated lime. Arrows indicate the dates sprays were applied. Check trees were not sprayed. 76 1.75 2 .5 - IJ 5 0.75 0.5 o> .9 - 0.25 10 0 Cu*"" loss (jig/cm2 of leaf) 20 30 40 Rainfall (mm) 21- 16- (j — ol a 'S Cu' * loss (vg/tm2 of leaf) 20 60 40 Rainfall (mm) 80 100 Figure 3.3. Nomogram relating the amount of copper lost from Montmorency sour cherry leaves to initial deposits of copper on the leaves and to the amount of rainfall. A, Citcop 4E; B, Tribasic copper sulfate. APPENDICES APPENDIX 1 Susceptibility of Prunus Rootstock Cultivars to Pseudomonas syringae pv. morsprunorum and P^. syrinage pv. syringae ABSTRACT Terminal shoots of eight cherry rootstock cultivars were wound-inoculated with Pseudomonas syringae pv. morsprunorum and JP. syringae pv. syringae. The shortest cankers developed on Colt and Mahaleb rootstocks and the longest cankers developed on MxM 2. Cankers on F12/1, MxM 14, MxM 39, MxM 97 rootstocks were intermediate • in length. INTRODUCTION Bacterial canker on Prunus sp. caused by Pseudomonas syringae pv. morsprunorum (5, 11) and I*, syringae pv. syringae (3, 7) can cause severe injury to susceptible cultivars. Resistance to bacterial canker has been measured by length of cankers after inoculation of leaf scars (6) and wound-inoculation of stems (1, 6, 10), and by severity of leaf spotting after natural infection of leaves (13). Canada, two distinct patterns of host susceptibility were recorded with syringae pv. morsprunorum and I?, syringae pv. syringae on wound-inoculated sweet cherry cultivars (1). On sweet cherry in England, cankers were 50% longer after wound inoculation with JP. syrinage pv. syringae from Oregon, USA than with £. syringae pv. 77 In 78 morsprunorum from England (6). Using the same isolates, cankers were 302! longer after inoculation of leaf scars with. P^. syringae pv. morsprunorum than with syringae pv. syringae (6). The purpose of this research was to test the susceptibility of eight cherry rootstock cultivars to P. syringae pv. morsprunorum and £. syringae pv. syringae by wound inoculations. MATERIALS AND METHODS One of strain £. syringae pv. morsprunorum and two strains of Pj. syringae pv. syringae were used to wound-inoculate the following rootstocks: Colt (JP. avium x P. pseudocerasus) , F12/1, Mahaleb (JP. mahaleb L.), MxM 2 (F12/1 x Mahaleb), MxM 14, MxM 39, MxM 60, and MxM 97'. The P^. syringae pv. morsprunorum isolate was a rifampicin- resistant strain (PsmR) initially isolated from Montmorency sour cherry (11). One _P. syringae pv. syringae strain was isolated from chokecherry (£. virginiana L.) leaves (PssC) and the other was a rifampicin-resistant strain (PssR) initially isolated from sweet cherry leaves (see Chapter 2). Inoculum was grown on King's medium B (8) for 2-days incubation at 22 C and suspended in 0.01-M phosphate buffer (pH 7.2) by stirring with a sterile glass rod. Inoculum g concentrations of 10 colony-forming units/ml phosphate buffer were obtained by adjusting the turbidity of the suspension to 0.04 absorbance with a spectrophotometer (Spectronic 20; Bausch & Lomb, Rochester, NY 14625) set at 625nm. Inoculum was kept in an ice bath during the inoculation period. Twenty to 30 1-yr-old trees of each rootstock cultivar- were planted in East Lansing, Michigan in June 1980, and were inoculated in 79 the summer of 1981. The rootstocks were pruned twice in 1981 to maintain four actively growing terminal shoots per tree. Each terminal was inoculated three leaf nodes below the meristem by inserting a 25-gauge needle longitudinally a few millimeters into the terminal and placing a 0.05 ml of inoculum at the point of insertion. Rootstocks were inoculated on 16 June with a mixture of PsmR and PssR, 13 July with PasC, and 31 August with PssC, and the cankers were measured on 11 July, 20 August, and 15 November, respectively. terminals per tree were inoculated with bacteria. Three A fourth terminal on each tree was inoculated with phosphate buffer without bacteria to serve as a control. Differences in length of cankers between rootstock cultivars for each inoculation date were analyzed using Duncan's multiple range test. The error mean square was pooled from all-three inoculation dates. The error mean square for each comparison was recalculated according to the number of replications in each treatment (14). RESULTS AND DISCUSSION Terminals inoculated only with phosphate buffer formed callus tissue around the point of inoculation. raised and was 0- to 1-mm long. The area of callusing was Mean length of cankers varied between inoculation dates; therefore, length of cankers were included for each inoculation date. The greater the length of time cankers were allowed to develop, the larger the cankers were that developed (Figure 4.1). The longest cankers developed after the inoculation on 31 August with PssC and data were taken 76 days later, while the smallest cankers developed 80 after the inoculation on 13 July with PssC and data were taken 38 days later. Cankers from the latter were smaller than cankers that developed after the inoculation on 16 June with PssR + PsmR and data were taken 25 days later. The small cankers, developed after the 13 July inoculation, may be caused by canker inactivity in summer or differences in virulence between PssC and PsmR + PssR. In England (5) and Oregon (3), cankers on sweet cherry were inactive in summer and the bacteria in the cankers frequently died. In this study, the activity of cankers, or rate of canker growth, from inoculations on 13 July (summer) and 31 August (autumn) were similar. Differences in length of cankers formed from inoculations on 16 June and 13 July were probably caused by PssC being less virulent than PsmR + PssR. Although the time of year inoculations were performed may have affected the rate of canker growth, growth rate of cankers was greatest for inoculations made on 16 June (0.41 mm/day) compared to 0.20 and 0.24 mm/day for inoculations on 13 July and 31 August, respectively. Crosse (5) reported canker growth was most active in spring and was associated with actively growing plant material. In this study, rootstocks were deliberately pruned twice in one growing season to maintain actively growing shoots. growth rate of cankers may be associated Therefore, differences in again with differences in virulence between PssC and PsmR + PssR. Aside from the differences in virulence and length of time for canker development, the rootstocks can be divided into three groups based on differences in length of cankers (Table 4.1). Generally, cankers on MxM 2 were longest; on F12/1, Mahaleb, MxM 14, MxM 60, and MxM 97 were intermediate; on and Colt were shortest. MxM 39,MxM 2 81 had the longest cankers on 16 June with PsmR + PssR and on 13 July with PssC. On 31 August, cankers on MxM 2 were 20-cm long, second to the 26-mm cankers on MxM 97. the shortest cankers. On all three inoculation dates, Colt had Cankers on MxM 39 inoculated on 16 June were in the shortest size group, while on 13 July and 31 August the cankers were intermediate in length. on MxM 14. Opposite to this were lengths of cankers These differences coincide with the use of different pathogen strains. Similar differences between strains of £. syringae pv. morsprunorum and P^. syringae pv. syringae and cultivars of sweet cherry were reported by Allen and Dirks (1). Similarly, Garrett (9) reported that F12/1 was moderately susceptible to race 1 of _P. syringae pv. morsprunorum, but was highly susceptible to the newly described race 2 of JP. syringae pv. morsprunorum. Rootstocks resistant to bacterial canker, such as Colt, may be useful to growers if they are budded high as reported and recommended in Oregon (2, 4). If rootstock cultivars resistant to bacterial canker are widely used, new virulent strains of P^. syringae pv. morsprunorum and JP. syringae pv. syringae may appear or develop because of host selection pressure. Therefore, all indigenous strains of the pathogen should be used for the development of bacterial canker resistance in an ongoing Prunus species breeding program. 82 LITERATURE CITED 1. Allen, W. R . , and Dirks, V. A. 1978. Bacterial canker of sweet cherry in the Niagara peninsula of Ontario: involved and cultivar susceptibilities. PseudomonaB species Can. J. Plant Sci. 58:363-369. 2. Anonymous. 1976. Oregon Plant Disease Control Handbook. Extension Service, Oregon State University, Corvalis 97331. 3. Cameron, H. R. 1962. Diseases of deciduous fruit trees incited by Pseudomonas syringae van Hall. with additional data. A review of the literature Oregon Agric. Exp. Stn. Tech. Bull. 66. 64 p p . 4. Cameron, H. R. 1971. Effect of root or trunk stock on susceptibility of orchard trees to Pseudomonas syringae. Plant Dis. Rep. 55:421-423. 5. Crosse, J. E. 1966. Epidemiological relations of the pseudomonad pathogens of deciduous fruit trees. Annu. Rev. Phytopathol. 4:291-310. 6. Crosse, J. E., and Garrett, C. M. E. stone-fruits. Jones, A. L. Michigan. 8. Bacterial canker of VII. Infection experiments with Pseudomonas mors-prunorum and P^. syringae. 7. 1966. 1971. Ann. Appl. Biol. 58:31-41. Bacterial canker of sweet cherry in Plant Dis. Rep. 55:961-965. King, E. 0., Ward, M. K., and Raney, D. E. 1954. Two simple media for the demonstration of pyocyanin and fluorescin. Clin. Med. 44:301-307. J. Lab. 83 9. Garrett, C. M. E. mors-prunorum. 1978. Pathogenic races of Pseudomonas Pages 889-890 in: Proc. IVth Int. Conf. Plant Path. Bact. Angers. 10. Garrett, C. M. E. 1979. Screening Prunus rootstocks for resistance of bacterial canker, caused by Pseudomonas morsprunorum. 11. J. Hortic. Sci. 54:189-193. Latorre, B. A., and Jones, A. L. 1979. Pseudomonas morsprunorum, the cause of bacterial canker on sour cherry in Michigan, and its epiphytic association with £. syrinage. Phytopathology 69:335-339. 12. Latorre, B. A., and Jones, ,A. L. 1979. Evaluation of weeds and plant refuse as potential sources of inoculum of Pseudomonas syringae in bacterial canker of cherry. Phytopathology 69:1122-. 1125. 13. Schmidle, A., and Zeller, W. 1979. Untersuchungen zer Resistenz von Sauerkirschsorten gegen den Bakterienbrand Pseudomonas syringae van Hall. Nachrichtenbl. Deut. Pflanzenschutzd. 31:177-178. 14. Steel, R. G. D., and Torrie, J. H. 1980. Procedures of Statistics. McGraw-Hill, New York. 633 pp. 2nd Ed. Principles and 84 Table 4.1. Canker lengths on Prunus rootstock cultivars following wound inoculations with Pseudomonas syringae pv. morsprunorum P. syringae pv. syringae. Date of inoculation Rootstock Colt Mahaleb 16 June 13 July 4.6X a 3.7 a . 6.0 a 11.4 w 31 August 13.1 a cd F12/1 10.8 b 8.1 MxM 14 12.9 be 4.9 a 14.8 a 8.5 18.7 b b MxM 39 5.4 a b 13.6 a 13.4 a be MxM 60 13.6 be 6.2 ab 19.4 MxM 97 10.9 b 5.1 a 26.2 MxM 2 14.9 Check^ c 12.2 d 20.0 0.5 0.4 0.8 Mean 10.3 7.5 18.1 Rate of growth2 0.41 0.20 0.24 w x c b 8 Actively growing terminal shoots were inoculated with 0.05 ml of 10 colony-forming units/ml phosphate buffer in 1981. On 16 June, terminals were inoculated with a 1:1 mixture of rifampicin-resistant strains of £. syringae pv. syringae recovered from sweet cherry and P^. syringae pv. morsprunorum recovered from sour cherry, and data were taken on 11 July. On 13 July and 31 August, a strain of J?. syringae pv. syringae from chokecherry was used to inoculate terminal shoots, and data were taken on 20 August and 15 November, respectively. Each value is a mean of 19 to 54 cankers and values in a column followed by the same letter do not differ significantly (JP ■ 0.05) according to Duncan's multiple range test. ^Mean length of cankers for all rootstocks inoculated with phosphate buffer alone. 2 The mean length of all cankers on a given inoculation date was divided by the number of days between the inoculation date and date data were taken to give the mm of canker growth per day. APPENDIX 2 RESIDUAL PLOTS FOR REGRESSION EQUATIONS OF CITCOP 4E AND TRIBASIC COPPER SULFATE A 1-0 .T 0 .5 1.0 1.5 Predicted Cu** loss P redicted CuM loss Figure 5.1. Plot of residuals (observed minus predicted values) against predicted amounts of copper loss indicating that the errors are independent, have zero mean with a constant variance and follow a normal distribution. A, Citcop 4E; B, Tribasic copper sulfate. 85