INFORMATION TO USERS This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted. The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction. 1. The sign or "target" for pages apparently lacking from the document 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 thru an image and duplicating adjacent pages to insure you complete continuity. 2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame. 3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete. 4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced. 5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received. Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48108 75-27,254 DE JONG, Jan, 1942GENETICS OF RESISTANCE TO CORYNEBACTERIUM MICHIGANENSE IN LYCOPERSICON. Michigan State University, Ph.D., 1975 Agriculture, plant culture Xerox University Microfilms, Anri Arbor, Michigan 48108 GENETICS OF RESISTANCE TO CORYNEBACTERIUM MICHIGANENSE IN LY COPERSICON By Jan de Jong A THESIS Submitted to Michigan State University partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Horticulture 1975 ABSTRACT GENETICS OF RESISTANCE TO CORYNEBACTERIUM MICHIGANENSE IN LYCOPERSICON By Jan de Jong The tomato disease, bacterial canker, is caused by Corynebacterium michiganense (E. F. Sm.) H. L. Jens. Inheritance of resistance to the pathogen was studied from (1) a half diallel cross with 4 resistant and 2 susceptible cultivars, and from (2) F 2 and backcross-generations derived from selected crosses from the above. Various methods of inoculation of seedlings with this bacterium were studied. Stem inoculation was proven most reliable in readily differentiating resistant and susceptible plants. Seedling resistance was shown to be related to resistance of the mature plant. Analysis of the half diallel showed that differ­ ences existed in general and specific combining ability for resistance in the resistant and susceptible cultivars. Resistance to isolate H could best be explained by 4 genes. A combination of one recessive gene (a) and 3 dominant genes (B, C, D) controlled .resistance in Lyco­ pers icon esculentum cv. L. pimpinellifolium cvs. Bulgaria 12. Resistance in the A 129 and A 134 was controlled Jan de Jong by the same genes, but the presence of the dominant allele 2 D enhanced the resistance of these cultivars and their F^ hybrids. Resistance in L. hirsutum PI 251305 was controlled by the genes AAbbccdd and the recessive gene x. A modifier gene F was found to be present. Susceptible cultivars of L. esculentum that 2 carry the dominant allele C transmitted a higher level of resistance to its progeny than cultivars with either the C or c allele. ACKNOWLEDGMENT The author wishes to express his sincere apprecia­ tion to Dr. S. Honma for his guidance as research advisor and for his critical appraisal of this manuscript. I also wish to express my gratitude to Dr. M. W. Adams for advise and valuable discussions during this study. Thanks are extended to the faculty for its time, advise and assistance, particularly those who served on the guidance committee: Dr. H. H. Murakishi and Dr. H. C. Price, and to Dr. H. G. Vest for his help in the linguistic improve­ ment of this dissertation. The author thanks the students and the staff of the Department of Horticulture who made this period interest­ ing and rewarding. TABLE OF CONTENTS PAGE LIST OF TABLES............................................ LIST OF FIGURES iv .................................... vii INTRODUCTION.............................................. 1 REVIEW OF LITERATURE. 4 ......................... MATERIAL AND METHODS ................................. Parental.............................................. Pathogen....................................... M e t hods ....... 13 13 14 14 SCREENING TECHNIQUES.................... 18 ENVIRONMENTAL INFLUENCES................................. 26 Temperature and light............................... Inoculum concentration ........................ 26 28 CRITERIA FOR RESISTANCE.................................. 3l Wilting ....... Ca nkers ..... .......*............................... Vascular discoloration....................... Stunting.............................................. 32 34 RESISTANCE OF MATURE PLANTS........................ 34 35 38 Mature plant vs seedling resistance................ 41 DIALLEL ANALYSIS......................................... 43 Reciprocal differences.............. 44 INHERITANCE OF RESISTANCE................................ 49 SUMMARY AND DISCUSSION................................... 76 BIBLIOGRAPHY.............................................. 80 LIST OF TABLES TABLE 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. PAGE Effect of root inoculation of 5 isolates of C. michiganense on 12 plants each of 6 tomato cultivars.. ................. 19 Percent of wilted seedlings of 3 tomato cul­ tivars inoculated by 6 methods with 2 isolates of C. michiganense............. 21 Evaluation of 3 inoculation methods with 2 isolates of C. michiganense on 4 cultivars of tomato. .......................................... 25 The cumulative mean score of disease resistance of 7 plants, for Fi hybrids and parents grown in the greenhouse and in growth chambers...... 27 Number of days from inoculation until 50% of the plants showed wilting......... 29 The reaction, 25 days after petiole-inoculation, of 3 cultivars of tomato to 2 isolates of C. michiganense..................................... 33 The mean fruit weight of.4 cultivars inoculated with isolate F or cm2 1 ........... 40 The yield of 5 cultivars of tomato inoculated at anthesis with isolate F or cm21 and control......... 40 Disease rating of 5 cultivars of tomato steminoculated with isolate H in the seedling stage and with isolate F or cm 21 at anthesis in the field......... .............................. ........ 42 F-, hybrid and parent mean values for resistance to isolate H .................... 45 Variance analysis for individual observations of disease resistance for the 15 F, hybrids and 6 parental lines of the half diallel hybridiza­ tion scheme .................. 46 PAGE TABLE 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Combining ability analysis for resistance to C. michiganense isolate H . *..... 46 General combining ability for resistance to C. michiganense isolate H .......................... 47 Specific combining ability for resistance to C. michiganense isolate H . . . ...................... 47 Chi-square test for 4-gene model for resistance to isolate H in the F , , BC to P-t, and BC to P ? of the cross Bulgaria 12 (P^) x MSU 72-279 (P2 ) ..... 53 Chi-square test for 4-gene model for resistance to isolate H in the F 2 / BC to P^, and BC to P 2 of the cross Bulgaria 12 (P^ x Earliana (P2 ) ...... 58 Chi-square test for 4-gene model for resistance to isolate H, 51 days after inoculation, i n .the F 2 » BC to P , , and BC to P 2 of the cross MSU 72-279 TPi) x A 134 (P2) ...... 61 Chi-square test fox; 4-gene model for resistance to isolate H in the F 2 , BC to Pi, and BC to P 2 of the cross Earliana (P^) x 134 (P2 ) ............. 63 Segregation ratios and Chi-square test for resistance to isolate H, 54 days after inocu­ lation, of progenies of Earliana (Pi) x A 134 (P2) ............ 65 ... Chi-square test for 4-gene model for resistance to isolate H, 34 days after inoculation, in the F 2 / BC to P^, and BC to P 2 of the cross MSU 72-279 (Px) x A 129 (P2 ) ................. 67 Chi-square test for 3 gene model for resistance to isolate H, 50 days after re-inoculation, in the F 2 , BC to P^, and BC to P 2 of the cross Earliana (Pi) x L. hirsutum (P2 ) .... 70 Chi-square to isolate the F 2 , BC M S U 72*279 72 test for 1 gene model for resistance H, 50 days after re-inoculation, in to Pi, and BC to P 2 of the cross x L. hirsutum........................... Segregation for resistance in the F 2 # BC to P ^ and BC to P 2 of the cross Bulgaria 12 (Pi) x L . hirsutum (P2 ) after petiole-inoculation with isolate cm 2 1 ..... v 74 TABLE 24. Segregation for resistance in the F 2 , BC to P-, , and BC to P 2 , of the cross A 134 (P^) x L. hirsutum (P2 ) after petiole-inoculation with isolate cm 2 1 ................................... . LIST OF FIGURES FIGURE 1. Disease ratings based on symptoms of plants inoculated with C . michiganense; .............. A. B. C. D. E. F. Plant showing one w i l t e d .leaf; Rating #2 Plant showing progressive wilting of leaves; Rating #1 Plant with wilted leaves and dead growing point; Rating #0 Plant showing several wilted leaves, growing point alive; Rating #1 Plant showing unilateral wilting of a leaf; Rating #2 Plant with stem canker; Rating #1 vii „ s ' INTRODUCTION Bacterial canker caused by Corynebacterium michiganense (E.F.Sm.) H. L. Jens., was first found in a green­ house of tomatoes Rapids, Mich. (Lycopersicon esculentum Mill.) near Grand (35). The disease is still being reported in the tomato growing areas of the world, with losses up to 80% of the crop (41). Initially control measures were aimed at cultural practices, attempting to prevent the spread of the bacteria in the field through quarantine and treatment of in­ fected seed. Symptoms of the disease in the greenhouse are best described by Bryan (6). wilting of the leaves. Most characteristic is the unilateral The cotyledonary leaves of infected seedlings first show wilt, followed by browning and shrivel­ ing. Later the true leaves wilt and shrivel, but the petioles of the shriveled leaves remain attached to the stem. The first wilting appears during the hottest part of the day but initially the plant recovers at night. of the stem are discolored The vascular bundles (brown) and in the advanced stage the stem cracks open and a canker develops. The seedlings show stunting and sometimes develop curved tips. With further development of the disease the whole pith of the stem becomes brown and loses its structure 1 (6). 2 In the field, the primary symptoms are the browning of the leaves and internal discoloration. The cankers, common in inoculated seedlings in the greenhouse, are not always a typical symptom in the field. The symptoms on the fruit do not always appear, as they are highly dependent upon rain or overhead irrigation. Fruit symptoms are known as bird's eye spot; brown centers with a characteristic white halo, ranging in size from 3 to 6 mm. Resistance in Lycopersicon pimpinellifolium Mill, was reported in 1944 (1). Elenkov (8) incorporated the resis­ tance of L. pimpinellifolium into L. esculentum cultivar Bulgaria 12. Fruits of Bulgaria 12 were too small for accep­ tance in the USA but the cultivar has been utilized by breed­ ers as a source of resistance to C . michiganense. Another source of resistance, L. hirsutum Humb. and Bonpl. PI 251305 (17) has not been used in breeding. Isolates of C. michiganense differ in virulence (43,49). Seedlings of the resistant cultivars succumb to the disease when inoculated with the virulent isolate cm 21, how­ ever, it is not known how the mature plant reacts to this isolate. Therefore, there is a need to compare mature plant resistance against seedling resistance, using the virulent isolate. Resistance was reported to be inherited as a dominant character (8). This was confirmed by Laterrot (25) who noticed that hybrids of susceptible tomato cultivars with Bulgaria 12 are as resistant as Bulgaria (12). 3 Thyr (49) suggested polygenic control of resistance. Apart from these observations there are no published data on the inheritance of resistance to C . michiganense. In­ formation on the genetics of the resistance of tomato to C. michiganense could improve the efficiency of breeding programs aimed at incorporating resistance into L. esculentum. The purposes of this study are; (1) The development of a fast and reliable screening method for detecting resistance to C. michiganense. (2) A determination of an acceptable level of resistance in L. esculentum. (3) Investigation of the genetic basis of resistance. REVIEW OF LITERATURE v. Smith, who originally called the disease the "Grand Rapids disease", named the causal bacterium Bacterium michiganense. The name of the disease was changed to bacterial canker in 1920 (37) . The organism has since been referred to as Pseudomonas michiganense, Aplanobacter michiganense, Phytomonas michiganense and now as Corynebacterium michigan­ ense (19) . The bacterium is disseminated by water, tools and man, and enters host plants primarily through wounds (28). Though it has never been proved that wounding is the only mode of entrance, most inoculation methods employ wounding to secure fast and uniform infection. Bacteria have been re­ covered from stomatal chambers after the leaves were sprayed with a bacterial suspension in water. Trichomes, both broken and entire, are thought to be an important point of entry of the bacterium, but subsequent spread through the plant is slow (2, 21). For example, 30 days after inoculating the leaf the bacterium had only reached the petiole (21). It has been suggested that minute rifts in the cuticle of the fruit i or broken hairs provide an opportunity for bacteria to pene­ trate the epidermis of the fruit. C.michiganense is a xylem invader and Thyr (44) has 5 reported that as low as 5 cells of the bacterium introduced directly into the xylem can cause infection. The initial infection in the plant and the subsequent movement of the bacteria takes place in the spiral vessel elements of the primary xylem (31). The bacteria move downward first, follow­ ed by upward movement in the xylem. Subsequently, when the bacteria break through the cell w a l l s , they move laterally, forming pockets of bacteria in the surrounding tissue. bacteria do not move longitudinally in the phloem The (31). Since C. michiganense is primarily a xylem invader, the most successful inoculation techniques are those which introduce the pathogen directly into the xylem (45). Inoculation of the root, stem, petiole and leaf have been re­ ported (1, 2, 17, 20, 21). Root inoculation involves wounding the root system and exposing of the wounded surface to bacteria. The roots are pruned while they are in the inoculum (20) , or the plant is dipped in inoculum after the roots have been cut back (38, 42). Other workers placed a bacterial suspension into the soil around young seedlings and then cut the roots at 4 points about an inch from the plant (31) . have given variable results Root inoculations (22, 31). Stem inoculations are made in several ways. The commonly used method involves piercing the stem with an in­ fected needle. Ark (1) using this method obtained only 2 infected out of 50 inoculated plants. Strider (42), pierced the stem at the base of the cotyledonary leaves of 2 and 4 6 week old seedlings with a root canal knife dipped in inoculum, and obtained infection from 96% to 100%. Seedlings have also been successfully infected by cutting the tip of the stem with a contaminated knife (1). An often used and successful method is cutting a petiole with a contaminated knife (31, 45). Hassan et al (17) screened for resistance using symptoms of the cotyledons and found a relationship between susceptibility, as determined by the reaction to root and/or stem inoculations, and the number of spots produced per unit area on the cotyledons. A well defined bacterial concentra­ tion and a standarized pressure of application was required for the cotyledonary method of, screening, since increased con­ centration and pressure of application resulted in a higher rating of susceptibility. Suspensions of bacteria have been sprayed on leaves with varying degrees of success, Ark (1) kept seedlings in mist chambers prior to spraying with a heavy suspension of C . michiganense and failed to obtain infection. However, atomization of a bacterial suspension from a distance of 35 cm on seedlings kept at high humidity, resulted in in­ fection of the leaves (21, .2). Layne (23) sprayed leaves of seedlings kept in high humidity with and without injuring the trichomes and obtained blister like lesions on both, however, the number of blisters tripled with injury of the trichomes. No systemic infection was reported. Smith (36) noticed bac­ teria in the stomata after spraying bacteria on the plant, without obtaining systemic infection. Scoring resistance is based on one or more symptoms 7 of the disease. Thyr (47) who does not advocate using the length of the canker as a measure of resistance to C. michigan­ ense, favored the length of the vascular discoloration, ex­ pressed as the percentage of stem length beyond the point of inoculation. Initial wilting as a measure of resistance is not advisable as L. pimpinellifolium, which is a source of re­ sistance to bacterial canker, has a tendency to show initial wilting followed by a complete recovery, Thyr (49) reported vascular discoloration as being superior to stunting in the determination of susceptibility since the lag between vascu­ lar discoloration and stunting was not the same for all cul­ tivar and isolate combinations. Other workers have used the degree of advanced wilting as a measure of resistance (22, 42). In screening, the experimental conditions have to be carefully monitored including plant age, inoculum concen­ tration and temperature (2, 14, 20, 24, 45). Optimal concentration of the inoculum depends on the method of inoculation. more cells/cm With foliar application, 10® or are necessary for infection (2). Piercing the stem with a rootcanal file dipped in a bacterial suspension requires a concentration of 10? to 10^ cells/cm^ (14), if in­ troduced directly into the xylem a minimum of 5 bacterial cells will infect seedlings and possibly only one cell is enough to infect the plant (44) . Part of the difference in the various concentrations needed for infection may be explained by the effectiveness of the various inoculation methods in intro­ ducing the bacteria directly into the xylem. Bacteria, 8 c especially in concentrations lower then 10 very rapidly in aqueous suspensions 3 cells/cm , die (40, 44). Symptoms take longer to appear in older plants (20, 42). Inoculation of 5 week old tomato plants (22 cm tall), produced the largest difference in disease rating be­ tween susceptible and resistant plants not the same for all varieties. (14) , however, this was L. hirsutum, (P.I. 251305) was most resistant at 4 w e e k s , L. pimpinellifolium, 340905) (P.I. at 5 weeks and L. esculentum cv Bulgaria 12 at 6 weeks. A day temperature of 24° C and a night temperature of 18° C is optimal for maximal destinction between resistant and susceptible plants, however, the temperature requirement differed with tomato accessions (14). It was generally con­ firmed that the disease progressed most rapidly under condi­ tions most favorable to the host, but that slightly less than optimal conditions gave maximum seperation between susceptible and resistant plants (4, 14). C. michiganense is potentially a devastating orga­ nism to the tomato. (3, 11). plants. In addition, certain weeds are reported to be hostSolanum douglasii, the perennial nightshade (18), Solanum mammosum L. Nutt. The bacterium may overwinter in the soil (36) , and recently Solanum triflorum (48) have been found to host pathogen. Hassan et al (17) induced symptoms of the cotyledons on a range of Solanaceous plants, but did not reisolate the bacteria from those plants. When their roots were inoculated only the Lycopersicon and Solanum species died or wilted but the 9 Capsicum species, were not affected. Volcani et al (51) d e ­ scribed a new leaf and fruit spot disease of pepper caused by C. michiganense. Bacterial canker is seedborne, although only 1% or less of the seeds in infested fruits carry the pathogen (20). Bacteria have been observed in the seedcoat cells but not in the endosperm or the embryo (29). The current recommended seed treatment is either fermentation at 18° C for 96 hours, or soaking of the dry seeds in a 0.6 solution of acetic acid for 24 hours modified (41). Recently, this method was slightly (50). Direct seeding of the tomato has helped to cut losses drastically in California (16). Experimental data on the effect of bacterial canker on the yield of tomato is p r e ­ sented by Emmatty & John (9). Early inoculation, up to 2 weeks after transplanting, resulted in a high mortality rate and consequently lowered the yield per plant. Inoculation as late as 4 to 6 weeks after transplanting caused signifi­ cant reduction in yield of the susceptible cultivar H 6 , as compared to H2990, a resistant tomato cultivar. Varying degrees of resistance to C. michiganense in L. esculentum have been noted since 1937 (28). However none of the commercial varieties tested were immune. The resistance in some L. pimpinellifolium lines is superior to that in L. esculentum in that after showing slight wilting following inoculation, the plants recover completely (8, -1) . 10 Elenkov (8), without presenting supporting data, reported resistance to be inherited as a dominant character. A level of resistance equivalent to that of the resistant cultivar 8/12 is expressed in the hybrids of 8/12 with the susceptible tomato cultivars Monalbo and Porphyre (25). Polygenic control of the reaction to C ., michiganense has also been suggested (49). Investigations into genetics of pathogenic bacteria - hostplant relationships are limited, probably due to pathogen variability. Pseudomonas phaseolicola (Burk.) Dows., the cause of halo blight in beans is believed to have differentiated into 2 races. Resistance to race one, as found in the Red Mexican U.I.3 cultivar of Phaseolus vulgaris L. is inherited as a monogenic dominant character (52). Schuster (34), using a host of different genetic constitution and possibly a different isolate of the bacterium reported 2 re­ cessive factors controlling resistance. Resistance to Xanthomonas campestris (Pammel) Dows., the cause of black rot in cabbage was found to be determined by one major gene f whose expression in the heterozygous con­ dition is influenced by 2 modifier genes. Ratios in crosses between resistant and susceptible plants therefore depend not only on the constitution of the major gene f but also on the differences at either the a or b locus . Discovery of other modifier genes was not excluded (55). The resistance to bacterial wilt of maize, caused by Phytomonas stewartii (E.F. Sm.) Bergey et al appears to 11 be controlled by 3 dominant, independently inherited g e n e s , and supplementary factors are thought to be involved (54). Bacterial variability has been demonstrated in Phytomonas stewartii (53). Successive passage through sus­ ceptible maize decreased virulence while passages through resistant maize increased virulence. Initially, this could be explained by selection of the existing variation in m i x ­ tures of virulent and avirulent bacteria. Lincoln (26) observed the same phenomenon, however, in single cell cul­ tures , with the exception that cultures passing through the resistant line retained their virulence without appreciable change. He notes that the work with single cell cultures did not differ from earlier results with colony or mass inoc­ ulations, Mutations at an estimated rate of between 1 to 20,000 and 1 to 800,000 supposedly furnished a source of varia bility great enough for this variation in virulence. Clonal variation in C. michiganense is well known. Ark (1) described color mutants and variants that were patho­ genic to tomato in varying degrees. A gradual loss of viru­ lence has been reported when the pathogen was grown on arti­ ficial media (37, 20) . Color and colony form has been linked to degree of virulence. Yellow and white forms that were r e ­ peatedly obtained from the pink were markedly more virulent than the parent culture (12). The high rate of mutation and the known recombination capability of bacteria through con­ jugation, transformation and transduction can explain the numerous strains and the continuous shift in virulence. 12 Low temperature storage of bacteria, keeping down the biolog­ ical activity, could result in stable strains. Strider (38) kept cultures of C . michiganense on nutrient agar at 2°C for 30 months without loss of virulence. Infected tomato stem pieces were frozen for 5 months after which test inoculations showed that the virulence was not lessened (20). Difference in virulence between isolates of C. michiganense was described by Strider (43) but no variation in virulence was found among single colony subcultures within isolates. The effect on virulence of successive passages through resistant or susceptible hosts has not been studied in C. michiganense. Isolates of C. michiganense that differ in patho­ genicity do not affect Lycopersicon accessions in the same manner. Isolate 829-S was more virulent on L. esculentum than isolate cm 4, as measured with vascular discoloration and stunting, however, isolate cm 4 was more pathogenic than 829-S on L. pimpinelli folium and L. hirsutum. This isolate accession interaction was insignificant when only L. esculentum was studied (49). MATERIALS AND METHODS Parental Six tomato cultivars, selected on the basis of their reaction to C. michiganense, were crossed in a half diallel crossing scheme. The cultivars were: A 129 (P.I. 344102, L. pimpinellifolium, resistant) A 134 (Utah 737 , L. pimpinellifolium, resistant) Bulgaria 12 ( , L. esculentum ,resistant) Earliana ( , L. esculentum ,susceptible) MSU 72-279 ( , L. esculentum , susceptible) P.I. 251305 ( , L. hirsutum , resistant) Cultivars A 129 and A 134 were obtained from Dr. Bill Thyr, USDA, Reno, Nevada. With the exception of L. hirsutum, all plants were selfed one generation prior to hybridization. A single plant from each cultivar was used for the 6-parent half diallel crossing, and was vegetatively propagated for the backcrosses. The F 2 generation consisted of the bulked seed from 6 Fj plants grown in the field. Six reciprocal combi­ nations were included to determine maternal effects. unilateral incompatibility Due to (27) of L. hirsutum all crosses with this genotype were made with L . hirsutum as,the male parent. 13 14 In addition to the cultivars mentioned above, the cultivars or F-^ hybrids; Saturn, Sl^ x Farthest North, P.I. 340905, G 14565 and Rapids were used for inoculation ex­ periments. The resistance to C. michiganense of mature plants of the cultivars Bulgaria 12, MH-1, Earliana, Saturn and G 14565 was measured in a yield trial. Pathogen Originally, 5 isolates of C. michiganense were available: cm 3, cm 15 and cm 21 from Dr. Bill Thyr, isolate F from Dr. James Farley O.S.U., Wooster, Ohio and isolate H from H. J. Heinz C o . , Bowling Green, Ohio. ported Isolate H was re­ (Emmatty, D.A. personal correspondence) to be identi­ cal to cm 15, but in this study it was more virulent. Iso­ lates H and cm 21 were used for the genetic study. The isolates were stored in the refrigerator at 4°C and once a year they were passed through a susceptible cul­ tivar. The inoculum was prepared from a nutrient broth cul­ ture which was continuously agitated for 4 days at room tem­ perature. Prior to inoculation the bacterial concentration was determined by making a cell count with a hymacytometer. Methods Seeds were planted in vermiculite and seedlings were transplanted 2 weeks later into flats. Tests were con­ ducted throughout the year in the greenhouse at 19°C or higher, or in growth chambers at a constant 19°C. Due to the 15 slow growth of L. hirsutum it was necessary to sow these seeds 3 days earlier than the seed of other cultivars. Three inoculation methods were used in this study. A. Stem-inoculation. Two to 3 weeks after transplanting the seedling tops were clipped off 1 cm above the cotyledons and a drop of inoculum was applied directly on the clipped stem. To compensate for the small cotyledons of the cul­ tivars of L. pimpinellifolium the tops were clipped above the first true leaf while for the tops of L. hirsutum were clipped above the second true leaf. When inocula­ ting with the less virulent isolates, plants were reinoc­ ulated 25 days after the first inoculation by clipping the tops of the plants above the first leaf and applying B. a drop of inoculum on the clipped stem. Petiole-inoculation. petiole of the first The true leaf of 4 week old seedlings was severed 3 mm from the stem with scissors which had been dipped into the inoculum. C. Root-inoculation. Roots of 2 week old seedlings were cut 1 cm below the hypocotyls, then the de-rooted plant was dipped in inoculum for one minute and planted into flats. Randomized block and split plot designs were used for this study. Cultivars were the main treatment and inoc­ ulation methods were the subtreatments. Unless mentioned elsewhere 3 replications were used with 7 plants of each entry per replication. For the diallel test, each replication consisted of 21 Fi and parental entries in 3 flats. Seven seedlings of 16 each entry were planted in a row with the seedlings spaced at intervals of 4.7 cm. To correct for variation between flats within replicates one row of the cultivar Earliana was planted in every flat. Four replicates were grown in the greenhouse and 2 in seperate growth chambers. The results were analysed according to Griffing (15) model 1, method 2. For the inheritance study, 7 plants of the P^, P 2 and F 1# 28 plants of the F 2 , 10 plants of the BC to P ^ , and 11 plants of the BC to P 2 were grown in one flat as one rep­ licate. Six to 12 replicates were used. The data were statistically interpreted by means of variance analysis. "T" tests, or when appropriate, Duncans multiple range tests were used to compare the m e a n s . Greenhouse plants were rated for disease develop­ ment several times during the experiment using the following scale: O = The growing point has succumbed to the disease. 1 = Extensive wilting, large cankers or stunted growth. 2 = Plant approaches normal size, but shows some wilting symptoms. 3 = Healthy seedling with no apparent symptoms of the disease. Where vascular discoloration was measured it was expressed as a percentage of the total stem length beyond the point of inoculation for petiole inoculated plants. If the seedling was decapitated for stem-inoculation the vascular discoloration was measured as a percentage of the total stem length of the tallest sprout. The yield trial for determining the resistance of mature plants consisted of 3 replications of 5 cultivars and 17 2 isolates in a split plot design with isolates as the main treatment and cultivars as the subtreatments. Guard rows were planted around the main treatment p l o t s , which were sep­ arated by a 4 meter wide alley. Five plants of each entry were harvested weekly. The number and weight of the mature fruit, recorded on a weekly basis, were used as a measure of resistance. SCREENING TECHNIQUES Preliminary tests were made to determine the most rapid and reliable method of inoculation under greenhouse conditions, and to learn which isolates of the pathogen pro­ duced the most uniform results. Resistant and susceptible tomato cultivars were inoculated with 5 isolates of C. michiganense by cutting the roots of 2 week old seedlings below the hypocotyls, and dip­ ping the stems in a water suspended inoculum. Although cul­ tivars A 134, A 129 and Bulgaria 12 have been reported to be resistant to C. michiganense (45, 47), results reported here (Table 1) do not agree with Thyr's observations. The viru­ lent isolate, cm 21, killed all cultivars although in resis­ tant lines the isolate required a longer incubation period. The resistant cultivars A 129 and A 134 did not succumb to isolate H as did the susceptible cultivars, therefore iso­ late H was selected for further studies . Since the preliminary study did not readily dif­ ferentiate between resistant and susceptible plants, a study was made to evaluate various inoculation methods. Isolates H and cm 21, resistant cultivars A 129, Bulgaria 12, and the susceptible cultivar Rapids were used. Inoculum was prepared from nutrient agar plates or nutrient broth shake cultures. 18 19 Table 1. Effect of root inoculation of 5 isolates of C, michiganense on 12 plants each of 6 tomato cultivars. Number of diseased plants Cultivar A 129 Bulgaria 12 Rapids S13 x FN PI 340905 A 134 X = not planted Days after inoculation cm3 cm15 12 16 19 24 12 16 19 24 12 16 19 24 12 16 19 24 12 16 19 24 12 16 19 24 F cm21 3 1 5 6 11 1 5 5 8 3 6 11 7 9 12 3 9 12 12 8 11 11 2 10 12 12 2 11 12 12 9 12 12 H 1 2 X X X X 1 1 2 X X X X 3 2 3 X X X X 4 8 8 3 11 11 11 1 1 4 4 7 2 9 9 11 8 10 12 12 3 7 10 2 7 9 11 1 2 2 5 20 The concentration of inoculum was adjusted to 4 x 10® bacteria / cm®. A randomized block design with 2 replications of 10 to 14 plants of each entry was used. The inoculation methods were as follows; I . Two week old tomato seedlings which had their roots cut 1 cm below the hypocotyl, were dipped into the inoculum for a minute and planted in flats. II. The petiole of the first true leaf of 4 week old seedlings was severed 3 mm from the stem, with scissors dipped in inoculum. III. A suspension of bacteria was atomized from a distance of 10 cm onto 3 week old seedlings, until they were dripping wet. IV. Carborundum powder was dusted on the cotyledons of 3 week old seedlings and gently rubbed with foam pads which had been dipped in inoculum. V. Inoculum grown for 5 days on nutrient agar was sus­ pended in water. Seeds of the cultivar Rapids were left to soak in the suspension for 72 hours prior to planting in flats. VI. Control. As method III, plants were atomized with distilled water. The results of the various inoculation methods are presented in Table 2. Method I resulted in the breakdown of resistance of A 129 and B 12, since plants from both cul­ tivars succumbed to isolates H and cm 21 after 19 d a y s . Method II showed the greatest difference in disease rating between resistant and susceptible cultivars with isolate H. 21 Table 2. Percent of wilted seedlings of 3 tomato cultivars inoculated by 6 methods with 2 isolates of C. michiganense. Cultivar Inoculation method Days after inoculation A 129 Bulgaria 12 Isolate cm 21 H Isolate cm 21 H Isolate H cm 21 % % % % Rapids % % Root dip 8 13 19 0 0 61 0 0 18 0 4 32 0 4 32 0 25 64 0 25 79 Petiole first leaf 8 13 19 0 43 81 0 0 5 0 15 85 0 5 14 0 29 86 0 50 100 Atomization 8 76 86 100 100 100 90 Abrasion of cotyledons 8 100 100 100 100 100 100 0 0 0 0 0 0 0 0 0 0 0 0 Seed soak 8 13 19 Control 8 13 19 not planted 0 0 0 not planted 0 0 0 0 0 0 0 0 0 22 Inoculation with isolate cm 21 produced no difference be­ tween resistant and susceptible cultivars. Method III: Seedlings of all cultivars showed 1 m m size white blisters on the cotyledons. The seedling failed to wilt and the disease did not become systemic. Method IV: Injuring the cotyledons prior to applying inoculum resulted in numerous blisters on the cotyledons. The white blisters, associated with infection, later darkened and coalesced, followed by dropping of the leaves. There was no systemic infection. The seedlings of Method V did not develop any symptoms. Probably the bacteria were inviable by the time the root and plumule emerged: cells/cm bacteria in dilution of 10^ lose viability between 24 and 168 hours (44), or the pathogen was unable to penetrate the seedling. No symptoms were observed on the control plants. Qualification of the term "resistance11 is necessi­ tated by the different responses of the resistant varieties to isolates and inoculation methods. Regardless of inocula­ tion technique, A 129 was not resistant to isolate cm 21 or to isolate H when inoculated on the roots, or when receiving a foliar spray. A 129 was resistant to isolate H when petiole inoculated. Therefore, petiole inoculation was investigated further. Pine et al (31) noted that bacteria moved from 10 to 45 mm in one day and suggested that bacteria may be drawn into the vessels following petiole inoculation. introduces non-genetic variation. This If the bacteria could be 23 allowed to spread in the plant and reach the growing point of all plants at the same time, the non-genetic variation could be reduced. By cutting the growing point 10 mm above an axillary bud, the bacteria would reach the bud in a day, or 6 days before the bud breaks. Petiole inoculation was compared to stem inocula­ tion on 4 cultivars to determine which method of inoculation would minimize the experimental and environmental variation. If cultivars are assumed to be pure lines all variation within a cultivar would be non-genetic. The number of days between wilting of the first and the last plant is consid­ ered to be a valid estimate of the non genetic variation. The following methods were used: I. II. Control, no inoculation. The growing points of 5 week old seedlings were clipped .5 to 1 cm above the first true leaf, and a drop of inoculum was applied to the decapitated stem. III. The first true leaf was removed 3 mm from the stem, with scissors dipped into the inoculum. IV. The petiole of the 2nd or 3rd emerging leaf was re­ moved as in Method III. Isolates cm 21 and F were used in a split plot design, with 3 replicates of 7 plants of each genotype. One replicate of the control seedlings of the cultivar MH-1 died prior to inoculation. Therefore, only 2 replications of MH-1 were used for the analysis of variance. The results for isolate cm 21 are presented in Table 3. The mean number of days between the appearance of symptoms on the first and the last plant was 3.8, 7.2, and 9.3 days respectively for Methods II, III and IV. ing was observed in the control. No wilt­ Method II showed minimal environmental variation and was therefore used for the genetic studies. Inoculation with isolate F caused wilting 23 days after inoculation in 3% of the seedlings of Bulgaria 12, 16% in Earliana, 40% in MH-1 and none in A 134. The scoring of wilting symptoms per se with this isolate would introduce a variation as plants within a susceptible variety could be rated either susceptible or resistant. Therefore, all the seedlings were cut tranversely at 2 mm intervals from the growing point toward the root to determine the percentage of the vascular discoloration (Table 3). A significant difference between cultivars was noted, with Method II producing the best differentiation between resistant and susceptible cultivars (Table 3). However, the coefficient of variation of individual measurements was 51%, which made this method of screening undesirable for individual plants. Table 3. Evaluation of 3 inoculation methods with 2 isolates of C. michiganense on 4 cultivars of tomato. Cultivar Days between wilting of first and last seedling inoculated with isolate cm 21 II III IV Petiole Petiole 2nd Stem 1st leaf or 3rd leaf Days Percent discoloration in seedlings inoculated with isolate F II III IV Petiole Petiole 2nd Stem 1st leaf or 3rd leaf Days Days % % % MH-1 5.7 7.3 10.0 94 73 68 Earliana 3.3 4.0 10.7 74 54 47 Bulgaria 12 3.3 6.7 7.7 5 7 6 A 134 2.7 10.7 8.7 0 0 0 Mean 3.8 7.2 9.3 43 34 31 LSD (.05) = 3.1 days LSD (.05) = 8% ENVIRONMENTAL INFLUENCES Temperature and light. A large variation in the degree of resistance per­ haps due to environmental effects was noted for inoculated seedlings. Since this study was carried out in the green­ house, temperature and light variation may have affected the disease rating. To determine the effect of varying environ­ mental conditions on the disease rating, seedlings grown and inoculated at constant temperature and light were compared with those grown and inoculated in the greenhouse. Two replicates of the hybrids and parents of the diallel cross were grown, following inoculation in 2 growthchambers, at 19°C ± 1 with a 13 hour day. Two other repli­ cates were kept in the greenhouse under natural light with a minimum temperature of 10°C and a maximum temperature reach­ ing 30° on sunny days. Twenty six hybrids and parents were planted, with 7 plants per ehtry per replicate. The seed­ lings were stem-inoculated with isolate H, 32 days after seeding, and scored for their resistance 23, 32, and 40 days later. Progenies resulting from hybridization of resis­ tant plants were resistant at both temperatures. The results in Table 4 are of the susceptible entries; the F^ hybrids of 26 27 Table 4. The cumulative mean score of disease resistance of 7 plants for Fj hybrids and parents grown in the greenhouse and in growth chambers. Growth chamber Greenhouse Hybrid or parent Days after inoculation 23 32 40 mean Days after inoculation 23 32 40 mean MSU 72-279 x A 129 33 33 22 29.3* 28 17 18 21 Earliana x A 129 40 37 38 38.3 28 24 34 28.7 MSU 72-279 x A 134 38 31 20 29.7 20 14 18 17.3 B-12 x Earliana 35 29 30 31.3 25 24 27 25.3 Earliana x B-12 32 30 31 31.0 28 20 26 24.7 8 3 0 3.7 6 2 0 2.7 MSU 72-279 x B-12 26 17 15 19.3 20 14 13 15.7 B-12 x MSU 72-279 14 11 12 12.3 25 15 17 19 A 129 x MSU 72-279 33 20 18 23.6 19 11 10 13.3 A 134 x MSU 72-279 31 25 17 24.3 18 11 13 14 MSU 72-279 x Earliana 13 4 1 6.0 14 2 1 5.7 9 0 0 3.0 18 2 0 6.7 MSU 72-279 x L. hirsutum 15 10 7 10.7 24 10 11 15.0 Earliana x A 134 38 38 41 39.0 34 30 39 34.3 Earliana MSU 72-279 * The higher the scale, the greater the resistance. LSD (for means within progenies) LSD (for each entry) (.05) = 4.9 (.01) = 6 .5 (.05) = 4.6 (.01 )= 6.2 28 resistant x susceptible cultivars and the susceptible parents The disease rating of individual entries was effect ed by the time of observation. The rating of the F^ hybrids that involved Earliana as the susceptible parent generally did not change from one rating to the next, while most hybrids of MSU 72-2 79 were rated less resistant at the 2nd or 3rd reading (Table 4). The variable environment of the greenhouse generally decreased resistance more than the con­ stant temperature in the growth chamber, with the exception of the Fi hybrids of Bulgaria 12 x MSU 72-279 and MSU 72-279 x L » hirsutum where the resistance is higher in the green­ house as compared to the growth chamber. It is concluded that the resistance of seedlings may be affected by experimental factors. However, the over­ all resistance relationship between genotypes within an environment or observation date did not change. Inoculum concentration Thyr (44) reported loss in viability of C . michiganense in aqueous solutions. This may affect the inoculum load between the first and the last seedling in­ oculated in a large experiment. To determine the effect of reduced inoculum load on the disease rating, seedlings of the cultivars Rapids, Bulgaria 12, and A 129 were petiole inoculated with isolates cm 21 and H, cultured on nutrient agar plates. Prior to inoculation the agar was placed in distilled water and mixed with a blender. The concentrations 29 Table 5. Number of days from inoculation until 50% of the plants showed wilting. Concentration of inoculum Cultivar Rapids Bulgaria 12 Isolate 8x10" 8 x 10 8 x 106 Days Days Days cm 21 14.5 13.5 15.5 H 18.0 18.0 18.0 cm 21 17.0 20.5 15.0 14.0 13.0 15.0 H A 129 cm 21 H — = Less than 50% wilting observed 33 days after inoculation 30 were adjusted to 8 x 10®, 8 x 10^, and 8 x 10® bacteria, / cm^. A randomized block design with 2 replicates and 14 plants per treatment was used. was recorded on alternate days. The number of wilted plants Wilting, the first symptom to be expressed by the diseased seedling, was considered to be most indicative of the concentration effect. The results are expressed as number of days following inoculation, when 50% of the plants showed signs of wilting (Table 5). Rapids seedlings inoculated with 3 concentrations of isolate H were wilted 18 days after inoculation, while the seedlings of Bulgaria 12 and A 129 recovered from the initial wilting following inoculation. All seedlings wilted,when inoculated with cm 21. Concentration difference did not affect the time between inoculation and wilting of the seedlings, suggesting that at the level used, 100 x changes in the concentration of the inoculum did not change the reaction of the seedling to the pathogen. CRITERIA FOR RESISTANCE In previous experiments 2 symptoms were used to indicate susceptibility wilting and vascular discoloration. These symptoms were inadequate, since some of the susceptible plants failed to wilt and the degree of the vascular dis­ coloration was variable. Other symptoms commonly associated with the diseased seedlings are stunted growth and cracks in the stem or petiole, commonly called cankers. In the follow­ ing experiment the manifestation of bacterial canker of toma­ to was studied with the purpose of determining the best pos­ sible symptom for differentiating between resistant and sus­ ceptible reactions. Twenty-five day old seedlings of the cultivars A 134, Bulgaria 12 and Spartan Red 8 were petiole inocula­ ted with isolates cm 21 and F. The bacterial concentration was adiusted to 15 x 10® cells / cm® and 3 x 10® cells / cm**; a 500 x dilution. The experiment was replicated 3 times with 7 plants per treatment. The wilting and occurrence of cankers were recorded d a i l y . Canker size, plant height, and the percentage of vascular discoloration were recorded at the termination of the study. The concentration of the inoculum did not have an effect on the reaction of the cultivars, therefore, the data 31 32 for concentration were pooled. Wilting Seedlings of susceptible and resistant cultivars inoculated with isolate cm 21 wilted and died 25 days after inoculation (Table 6). Wilting did not occur on all plants inoculated with isolate F, but seedlings of the susceptible cultivar Spartan Red 8 differed significantly in wilting from those of the resistant cultivars A 134 and Bulgaria 12. It appears that the resistance of a cultivar may be deter­ mined by the percent of wilted plants, but not the resis­ tance of individual plants (Table 6). 33 Table 6. The reaction, 25 days after petiole-inoculation, of 3 cultivars of tomato to 2 isolates of C. michiganense. Cultivar Percent plants with cankers (isolate F) Percent wilted plants (isolate F) Percent wilted plants (isolate cm 21) % % % Spartan Red 8 74 43 100 Bulgaria 12 17 7 97 4 3 93 20% 10% A 134 LSD (.05) : 10% 34 Cankers Seedlings inoculated with isolate cm 21 succumbed prior to canker development. Inoculation with isolate F caused canker to develop on 74% of Spartan Red 8 seedlings, while only 43% of the seedlings showed wilting (Table 6). Cultivars can be classified as resistant or susceptible, based on the mean number of plants showing canker on the stem (Table 6), however, the determination of resistance of indi­ vidual plants was not always possible since many plants did not develop canker. Vascular discoloration The percent of vascular discoloration in Spartan Red 8 ranged from 0 to 100%, while for Bulgaria 12 the range was from 0 to 16%. Inoculation at the 3 leaf stage of Bul­ garia 12 showed a range of discoloration from 0 to 42%, and for the Spartan Red 8 from 6 to 100%. The coefficient of variation for Spartan Red 8 is 52%, which makes discoloration unsuitable for the determination of the resistance of indi­ vidual plants, but the mean score may be used for the deter­ mination of the resistance of a cultivar (Table 5). Thyr (47) , using mean ratings between cultivars. has shown that the size of the canker was positively corre­ lated with the degree of vascular discoloration. A similar observation was noted in this study, however, within a cul­ tivar no relationship was found. 35 Stunting Stem growth of the Spartan Red 8 seedlings, beyond the point of inoculation, ranged from 2 to 43 mm, with a mean of 15 mm and a coefficient of variation of 74%. For Bul­ garia 12 the growth ranged from 10 to 72 m m with a mean of 27 mm and a coefficient of variation of 46%. A significant correlation (r = . 80) between stunting and vascular discoloration was noted for Spartan Red 8 . There was no relationship between the size of the canker and plant size (r = .20). Apparently, each of the described symptoms alone does not appear to be a reliable indicator of the reaction of individual plants to the pathogen. The criteria for re­ sistance used in the genetic analysis included wilting, stun­ ting and the presence of canker (Figure 1). 0= Death of the growing point. 1= Extensive wilting, stunted growth, or large cankers (3 mm or larger). 2= Plant approaching normal size with some wilting of leaves or leaflets. 3= Healthy seedlings with no apparent symptoms of the disease. Wounds of 2 mm or smaller at the point of inoculation may occur. Disease ratings based on symptoms of plants inoculated with C. michiganense: A. B. C. D. E. F. Plant showing one wilted leaf; Rating #2 Plant showing progressive wilting of leaves; Rating #1 Plant with wilted leaves and dead growing point; Rating #10 Plant showing several wilted leaves, growing point alive; Rating #1 Plant showing unilateral wilting of a leaf; Rating #2 Plant with stem canker; Rating #1 37 Figure 1 RESISTANCE OF MATURE PLANTS In the summer of 1973 . 5 cultivars of tomato with variable resistance to C . michiganense were planted in the field at intervals of 91 cm in rows, 152 cm apart. Bulgaria 12 is characterized by small fruits and resistance to C. michiganense which was derived from L. pimpinellifolium (8) . G-14565 is a resistant cherry tomato (46) . Earliana is a susceptible cultivar whose hybrids with L. pimpinellifolium remained healthy when inoculated with a combination of isolates of the pathogen C . michiganense. (8). MH-1 is susceptible to Saturn is resistant to bacterial wilt caused by Pseudomonas solanacearum E . F. Sm. Thyr (person­ al communication) has reported that resistance to P. sola­ nacearum provides some protection to C . michiganense. The plants of each genotype were petiole inocula­ ted at anthesis with isolate F or cm 21. The mean fruit weight and the number of fruits were recorded weekly during the harvest season, except for G-14565, where due to small fruits, the total fruit weight was determined at the end of the season. Measurement of the vascular discoloration in plants sampled from the guard rows at various times during the growing season showed that the plants were infected with C. michiganense. 38 39 Fruit number was not affected by infection with either isolate, however, fruit size was reduced. shows the reduction in the mean fruit weight. Table 7 With the ex­ ception of Bulgaria 12 the mean fruit weight was' reduced by isolate cm 21. second harvest. This reduction became significant after the Only MH-1 showed a significant decrease in the mean fruit weight when inoculated with the less virulent isolate F. Isolate F did not affect the total fruit yield of any of the cultivars, while cultivars inoculated with isolate cm 21 showed an average reduction in yield of 17% (Table 8). The absence of a isolate x cultivar interaction suggests there was no cultivar difference in their reaction to isolate F and cm 21. However, F tests performed on the yield of individual cultivars showed that M H - 1 , inoculated with cm 21, yielded significantly less than the control. It is apparent that MH-1 was the most susceptible cultivar in this study, with Earliana and Saturn showing a low level of tolerance to the pathogen. The resistance of Bulgaria 12 was superior since the mean fruit weight of this variety was not reduced when infected with the virulent isolate cm 21. It would appear that the seedling resistance as shown by Bulgaria 12 in greenhouse tests could be used as a criterium for selecting resistant plants in a segregating population. Table 7. The mean fruit weight of 4 cultivars inoculated with isolate F or cm 21. Mean fruit weight in grams Cultivar Control Bulgaria 12 F cm! 2.1 42 41 38 MH 1 179 150 97 Earliana 119 130 87 Saturn 122 113 81 Means joined by a common line are not different from each other (.05) Table 8. The yield of 5 cultivars of tomato inoculated at anthesis with isolate cm 21 and control. Total yield per plant in kg Cultivar Control F cm 21 Bulgaria 12 8.2 8.6 7.6 MH 1 8.1 7.8 4.7 Earliana 6.9 7.6 6.1 G 14565 3.6 3.9 3.6 Saturn 6.9 7.7 5.6 6J)____________ 7;2 5.7 mean Figures joind by a common line are not different from each other (.05) 41 Mature plants vs seedling resistance. Four week old seedlings of the cultivars Earliana, Bulgaria 12, Saturn, MH-1 and G-14565 were grown in the greenhouse following stem-inoculation with isolate H . The disease ratings, on the scale 0 to 3 are presented in Table 9. Earliana, which showed a low degree of tolerance as a mature plant was as susceptible as MH-1 in the seed­ ling stage, 23 days after inoculation. Saturn showed a higher tolerance than either MH-1 or Earliana. was resistant in the seedling stage. Bulgaria 12 Thirty two days after inoculation the resistance of Saturn diminished while Bulgaria 12 and G-l4564 remained resistant. It is apparent that the seedling screening technique used may be reliable, however small differences in resistance may go undetected especially when scored 30 days, or later after inoculation. Table 9. Disease rating of 5 cultivars of tomato stemC>c is proposed to confer resistance to plants of the genotype AaBBC^-D-. 2 2 The genotype AAbbC C dd is proposed for Earliana. The F 2 (Table 16) fits the ratio 54 resistant: 202 susceptible (P= .80) with resistance controlled by the genotypes 2 2 aaB-C - D -, aaB-CCD-, and AaBBC - D - . The observed and ex­ pected frequencies of the P^, P 2 , F l/ F 2 ' BC to P^, and BC to P 2 are presented in Table 16. Chi-square test for 4-gene model for resistance to isolate H in the F2 , BC to P !, and BC to P2 of the cross Bulgaria 12 (Px) x Earliana (P2). Generation Expected segregation Unweighted R S R Number R S Weighted S Ratio R S 72 10 82 0 1 : 0 0 75 0 75 0 : 1 Fi 24 58 0 82 0 : 1 f2 78 253 71 260 70 261 BC to Pt 82 50 79 53 83 BC to P2 10 110 6 114 77 Bulgaria 12 (Pi) Earliana (P2) 32 X2 P 49 10 : 6 .40 0 120 0 : 1 5 82 0 1 : 0 0 75 0 75 0 : 1 F, 22 60 0 82 0 : 1 f2 83 248 72 259 70 261 54 : 202 .16 00 BC to P! 86 46 87 45 83 49 10 : 6 .30 .50 - .70 BC to P2 7 113 3 117 0 120 0 : 1 Bulgaria 12 (Pi) Earliana (P2) 40 o o 54 : 202 CM O• • (0 Days after inoculation Observed segregation CO Table 16. .50-.70 o 05 1 o Table 16 (continued) Days after inoculation Generation Bulgaria 12 (Pi) Earliana (P2) Observed segregation Expected segregation Unweighted R S Number R S Weighted R S Ratio R S X? P 54 : 202 4.07 .02 - .05 1.82 .1 0 -.2 0 54 : 202 8.0 .001 - .01 1.36 .20 -.3 0 77 5 82 0 1 : 0 0 75 0 75 0 : 1 Fi 13 69 0 82 0 : 1 F* 60 271 55 276 70 261 BC to Pj 91 41 90 42 83 49 10 : 6 BC to P2 2 118 0 120 0 120 0 : 1 78 4 82 0 1 : 0 0 75 0 75 0 : 1 Fi 12 70 0 82 0 : 1 f2 55 276 49 282 70 261 BC to Pj 91 41 89 43 83 49 10 : 6 BC to P2 1 119 0 120 0 120 0 : 1 Bulgaria 12 (P J Earliana (P2) 47 93 60 The weighting factor as based on the reaction of the parental populations and the those described earlier. The hybrid is similar to phenotype is represented by the genotypes AaBbC^ - (or CC) D- which occur at a frequency of 48/256 in the F 2 . Data from the 47 and 93 day observation do not fit the proposed model as resistance in the F 2 continued to break down, possibly due to a delayed reaction of certain genotypes to the pathogen. MSU 72-279 x A 134 F 1 hybrids of the cross MSU 72-279 are less susceptible 12 x MSU 72-279 (P^) x A 134 (P2) (P= .05) than the F^ hybrids of Bulgaria (Table 10), suggesting that A 134 transmits a higher level of resistance to its progeny than Bulgaria 12. The higher resistance of A 134 is explained by the presence of an allel of gene D in A 134 with an order of dominance of D^> D > d. Similar to increases the re­ sistance of plants heterozygous for A, so that the genotype of AaBBC-D^ - are assumed to be resistant. The proposed genotype for MSU 72-279 is AAbbccdd and aaBBCCD2D2 for A, 134. The weighted F 2 ratio (Table 17) fits to the expected ratio of 45 resistant: 211 susceptible (P= .30) with aaB-C-D^- and AaBBC-D^. controlling resistance. The BC to P^ and P2 weighted ratios give an acceptable fit to the expected 0 resistant: 1 susceptible and 3 resistant: 1 susceptible ratios respectively. ^=*T35SC33S5B0^5^S!6SRS$3KE[O Table 17. Chi-square test for 4-gene model for resistance to isolate H, 51 days after inoculation in the F2 , BC to Px, and BC to P2 of the cross MSU 72-279 (Px) x A 134 (P2). Generation Observed segregation Expected segregation Unweighted S R Number R S Weighted R S Ratio R S X2 P .49 .30 - .50 .60 .3 0 -.5 0 MSU 72-279 (Pi) 0 56 0 56 0 : 1 A 134 (P2) 56 0 56 0 1 : 0 Fi 11 45 0 56 0 : 1 f2 49 175 43 181 39 185 45 : 211 BC to Pt 3 85 2 86 0 88 0 : 1 BC to P2 67 13 63 17 60 20 3 : 1 62 Earliana x A 134 hybrids of the cross Earliana (P^) x A 134 (P2 ) have shown a high level of resistance to isolate H (Table 10). A combination of dominant genes in the F1 provides a level of resistance, however misclassification may put a proportion of the F^ hybrids in the susceptible class. A minimum weighting factor is obtained by assuming that under the conditions of this test the F^ is resistant. The proposed genotypes for Earliana and A 134 are AAbbC2C 2dd and aaBBCCD2D 2 respectively. If the F^ AaBbC^CD^d is classified as resistant, then the susceptible genotypes are A A — — of 144/256. — , — bb — — and — — — dd with a frequency The ratio of resistant to susceptible in the F 2 is therefore expected to be 112 : 144. The weighted segregation ratio for 44 days after inoculation shows a good fit to a 4 gene model (Table 18). The genotypes of the BC to P 2 are all expected to be resistant, whereas, in the BC to P^ the genotypes AaBbC2C2D2d and AaBbC2CD2d , with a combined frequency of 2/16, are resistant. With a resistant F^, weighting will increase the number of resistant plants in the segregating populations. For the F 2 (Table 18, day 44) the weighted number of resis­ tant plants is: = 80 + (12/55 x 72/256 x 223) = 94 with 12/55 = number of susceptible F^ hybrid plants of total 72/256 = frequency of the F x phenotype (AaB-C2CD2 AaB-CCD2-) 223 =* total number of plants in the F 2 . Table 18. Chi-square test for 4-gene model for resistance to isolate H in the F2, BC to Earliana (PJ x A 134 (P2). Days after inoculation Generation P t, and BC to P2 of the cross Observed segregation Expected segregation Unweighted R S R Number R S Weighted S Ratio R S 56 0 1 56 0 56 0 1 0 Fi 42 13 55 0 1 0 F2 95 128 110 1.13 98 125 112 BC to Pi 9 72 11 70 10 71 1 7 BC to P2 74 12 84 2 86 0 1 0 0 56 0 56 0 1 A 134 56 0 56 0 1 0 F* 43 12 55 0 1 0 f2 80 143 94 129 98 125 112 BC to Pi 5 76 7 74 10 71 1 7 BC to P2 68 18 77 9 86 0 1 0 A 134 (P2) Earliana (Pi) 44 144 144 2.61 o 0 33 CN • • 56 (Pi) P o 0 Earliana X2 .11 .7 0 -.8 0 .29 .30 - .50 1.12 .20 - .30