STUDIES ON CHEMICAL CONTROL OF WHEAT STEM RUST BY JOHANNES STEPHANUS KOMEN A dissertation submitted in accordance with the requirements for the degree Master of Science in Agriculture In the Faculty of Natural and Agricultural Science Department of Plant Sciences (Division of Plant Pathology) at the University of the Free State Supervisor: Professor Z.A. Pretorius NOVEMBER 2007 BLOEMFONTEIN SOUTH AFRICA I declare that the dissertation hereby submitted by me for the degree of Master of Science in Agriculture at the University of the Free State is my own independent work and has not previously been submitted by me at another University/faculty. I furthermore cede copyright of the dissertation in favour of the University of the Free State. J.S. Komen i ACKNOWLEDGEMENTS I would like to convey my sincere gratitude and appreciation to: • My supervisor, Prof. Z.A. Pretoruis, for his continued support and assistance with the preparation of this dissertation; • Pannar Pty. Ltd., for the opportunity to conduct this research and for financial support and assistance, in particular Dr. F.J. Kloppers and Mr. A. Babooram; • The University of the Free State, for using their facilities to conduct my research; • The ARC-Small Grain Institute for data; • My parents, for all the opportunities granted to me through their hard work and sacrifices; • The love of my life, Natasha, for her love, understanding and assistance, and • Finally, my Heavenly Father for His abundant grace. ii CONTENTS Preface vi 1. AN OVERVIEW OF PUCCINIA GRAMINIS F. SP. TRITICI 1.1. Introduction 1 1.2. Pathogen 2 1.2.1. Host specialization 2 1.2.2. Pathotype differentiation 3 1.2.3. Prevalence of Puccinia graminis f. sp. tritici worldwide 4 1.2.4. History of wheat stem rust in South Africa 4 1.3. Symptoms 5 1.4. Epidemiology 6 1.4.1. Life cycle 6 1.4.2. Inoculum sources 7 1.4.3. Environmental conditions 8 1.5. Economic Importance 9 1.5.1. Global threat of stem rust pathotype Ug99 10 1.6. Disease Control 11 1.6.1. Cultural practices 11 1.6.2. Barberry eradication 12 1.6.3. Fungicides 12 • Foliar application 13 • Seed treatment 13 1.6.4. Breeding for resistance 14 1.6.5. Types of resistance 15 • Seedling resistance 15 • Adult-plant resistance (APR) 15 • Slow rusting 16 1.6.6. Sources of resistance 17 1.6.7. Molecular markers 17 iii 1.6.8. Durability 18 1.6.9. The effect of environment on host-pathogen interactions 19 1.7. Conclusion 20 1.8. Literature Cited 22 2. PATHOTYPES OF PUCCINIA GRAMINIS F. SP. TRITICI DETECTED IN SOUTH AFRICA FROM 1998-2004 2.1. Abstract 40 2.2. Introduction 41 2.3. Materials and Methods 42 2.4. Results 43 • 1998 to 1999 44 • 2000 44 • 2001 44 • 2002 44 • 2003 45 • 2004 45 2.4.1 Comparison of differential sets 45 2.5. Discussion 46 2.6. Acknowledgements 49 2.7. Literature Cited 49 3. IN VITRO TOXICITY AND RESIDUAL EFFECTS OF SELECTED FUNGICIDES TO PUCCINIA GRAMINIS F. SP. TRITICI 3.1. Abstract 70 3.2. Introduction 71 3.3. Materials and Methods 72 3.3.1. Rust pathogen 72 3.3.2. Effects of temperature and incubation period 72 3.3.3. Fungicide experiments 73 • EC50 determination 73 iv • Germination response to fungicide exposure 73 • Residual effects 74 3.3.4. Statistical analysis 75 3.4. Results and Discussion 75 3.4.1. Effects of temperature and incubation period 75 3.4.2. EC50 determination 76 3.4.3. Germination response to fungicide exposure 77 3.4.4. Residual effects 78 3.5. Literature Cited 79 4. THE EFFECT OF STEM AND LEAF RUST ON YIELD OF BREAD WHEAT 4.1. Abstract 92 4.2. Introduction 92 4.3. Materials and Methods 93 4.3.1. 2005 Experiment 94 4.3.2. 2006 Experiment 94 4.3.3. Statistical analysis 95 4.4 Results 95 4.4.1. 2005 Experiment 95 4.4.2. 2006 Experiment 96 4.5 Discussion 97 4.6 Literature Cited 98 5. SUMMARY 109 OPSOMMING 110 v PREFACE The reemergence of wheat stem rust as an important disease of bread wheat not only in South Africa, but also worldwide, served as justification for this study. A significant increase in stem rust occurrence in the Western Cape suggested that there is a major build-up of inoculum and that cultivars lack resistance. Although the emphasis was on chemical control, a chapter dealing with pathogenic variability in Puccinia graminis f. sp. tritici is included. Most of the survey data were obtained while the candidate was employed by the ARC Small Grain Institute at Bethlehem. The dissertation is arranged as independent chapters and a degree of duplication was therefore unavoidable. vi CHAPTER 1 AN OVERVIEW OF PUCCINIA GRAMINIS F. SP. TRITICI 1.1 INTRODUCTION Stem rust caused by Puccinia graminis Pers.:Pers. f. sp. tritici Eriks. & E. Henn., is an economically important disease of bread wheat (Triticum aestivum L.) worldwide (Roelfs, Singh & Saari, 1992; McIntosh, Wellings & Park, 1995). The disease has often reached epidemic proportions in South Africa, the most recent being the stem rust epidemic on Sr24-derived wheat cultivars in 1985 (Le Roux, 1985; Le Roux & Rijkenberg, 1987a,b). According to McIntosh et al. (1995), effective genetic control of rust diseases requires a coordinated effort, including pathotype monitoring, collection and characterization of sources of resistance, and resistance breeding. Data generated by pathotype surveys thus form an essential component of breeding programmes and are conducted by most wheat producing countries, principally to recognize pathogenic changes (McIntosh et al., 1995). The timely detection of stem rust pathotypes with new virulence is considered important to the South African wheat industry. The effect of a shift in virulence on cultivar susceptibility, and thus potential losses, is important to producers aiming for maximum yields. Similarly information on pathogenicity enables geneticists to utilise the most effective resistance genes in their breeding programmes. In South Africa, 22 pathotypes of P. graminis f. sp. tritici that occur on wheat and triticale (X Triticosecale Wittmack) have been characterised from 1981-1997 (Le Roux & Rijkenberg, 1987b; Le Roux, 1989; Smith & Le Roux, 1992; Boshoff, Van Niekerk & Pretorius, 2000). A recent review of P. graminis, including the wheat stem rust pathogen, focused on the importance, taxonomy, host range, symptoms and genetics of this fungal species (Leonard & Szabo, 2005). The objective of the present overview is to provide a summary of the wheat stem rust pathogen, its symptoms, epidemiology and disease control. Literature review 2 1.2 PATHOGEN Stem rust of wheat is caused by Puccinia graminis Pers. f. sp. tritici Eriks. & Henn. The disease is also known as black rust or summer rust (Roelfs, 1985). Stem rust is an obligate parasite that can only grow in living host material (Wiese, 1987). This particular forma speciales can infect numerous wheat cultivars but also a few barleys, rye and some grasses (Knott, 1989). Apart from wheat and its alternate hosts, the stem rust fungus has a relatively narrow host range, but some grasses are important sources of primary inoculum (Wiese, 1987). In addition to wheat, the fungus completes part of its life cycle on alternate hosts, especially barberries (Berberis vulgaris L. and B. canadensis Mill.) and certain species of Mahonia (Roelfs, 1985). In 1916, Stakman found that P. graminis f. sp. tritici comprises different physiological types (races), and that certain wheat varieties while resistant to some races, were fully susceptible to others. Numerous specified pathotypes of P. graminis occur and are probably formed by mutation or somatic hybridization (Luig, 1977). Burdon et al. (1982) used isozyme phenotypes to assess the origin and evolution of stem rust in Australia and found that the major changes in the wheat stem rust flora resulted from exotic introductions. 1.2.1. Host specialisation Within P. graminis, specialization on particular host genera has occurred to produce formae speciales. Five physiological varieties, some of which consist of a number of races are P. graminis f. sp. avenae on oat and related grasses, P. graminis f. sp. secalis on rye and related grasses, P. graminis f. sp. agrostidis on Agrostis spp., P. graminis f. sp. poae on Poae spp. and most importantly, P. graminis f. sp. tritici on wheat, barley and many of the relatives of wheat (Verwoerd, 1935; Knott, 1989). Many species of Berberis, Mahonia, and their hybrid (X Mahoberberis) are susceptible to P. graminis. According to Le Roux & Rijkenberg (1987a), stem rust oversummers on grasses in South Africa namely Hordeum murinum Huds. and Festuca arundinacea. Pretorius et al. (2007) reviewed literature describing Agropyron distichum, Hordeum vulgare, Lolium italicum, Bromus maximus and Dactylis glomerata as accessory hosts in South Africa. Infection with P. graminis f. sp. tritici has been found on 112 species of wild and cultivated grasses from Israel (Gerechter-Amitai & Wahl, 1966). Literature review 3 1.2.2. Pathotype differentiation The use of race–specific resistance to control wheat stem rust requires continued monitoring of the variation in the pathogen population for virulence (Roelfs, 1985). In 1916, Stakman and colleagues at St. Paul, Minnesota, U.S.A., showed that P. graminis f. sp. tritici comprised different biotypes. This discovery revolutionised the science of plant pathology and different biotypes were subsequently discovered in many other pathogens (Luig, 1983). These variants have been termed races, physiologic forms, pathotypes and strains. Several systems are or have been used for the identification of physiological races in P. graminis f. sp. tritici. • International system (Stakman, Stewart & Loegering, 1962). Twelve standard differentials were used worldwide until about 1950. This set is still in use for communication purpose worldwide. Physiologic races are based on combinations of host reaction classes. • Modified potato – Phytophthora infestans system (Watson & Luig, 1962). Nomenclature consists of the international race number, followed by ANZ for New Zealand and then the numbers of the additional differential hosts that are susceptible. According to McIntosh et al. (1995) a selection of six genotypes, an additional 11 wheats and two triticale lines, are used as a supplementary differential set in Australia. • Formula-method (Green, 1965). Used in Canada to identify physiologic races according to the avirulence / virulence formula for each pathotype. • Coded sets (Roelfs & Martens, 1988; Roelfs, Long & Roberts, 1993). Four sets of four host lines each with a single gene for resistance are used to identify physiologic races in the U.S.A and Canada. New subsets can be added without affecting the previously used coding. Differential hosts are placed in sets of four, and a letter is assigned to each of the possible resistant/susceptible combinations. • South African system (Pretorius et al., 2007). New races are identified by an alpha-numeric code. The identity consists of three components. A number denote the rust type viz. 2 denote P. graminis f. Literature review 4 sp. tritici and 3 Puccinia triticina, followed by SA and an accession number for each distinctive pathotype. 1.2.3. Prevalence of virulence in Puccinia graminis f. sp. tritici worldwide Worldwide virulence for Sr2, 13, 22, 24, 25, 26, 27, 29, 31, 32, 33, 34, 37, Gt and Wld-1 is absent or limited. Sr13 is ineffective at lower temperatures, whereas Sr29 and 34 may be ineffective under high inoculum densities (Roelfs et al., 1992). Virulence for Sr24 and Sr27 exists in South Africa (Le Roux & Rijkenberg, 1987a). Sr26 virulence is undetected despite the widespread use of the cultivar Eagle and its derivatives in Australia (Luig, 1983; McIntosh et al., 1995). The extensive use of Sr31 in Kavkaz and similar wheats with the 1B/1R translocation did not result in virulence for this gene worldwide until Sr31 succumbed in Uganda in 1999 (Pretorius et al., 2000). Virulence for Sr6, 11 and 17 is common wherever these genes have been used. Virulence for Sr5, 9e and 21 appears to be common in some areas, but remains low or absent in other areas. Virulence is common for Sr8b (except in Australia, New Zealand and South Africa), Sr9a, Sr9d and Sr14 (except North America), Sr12 (except North America, Australia and New Zealand), Sr15 (except Africa, North America, Australia and New Zealand), and for Sr16, Sr18, Sr19, Sr20 and Sr28 (except in China, India, Nepal, Pakistan and Ethiopia) (McIntosh et al., 1995). Table 1 contains references to published virulence surveys. Luig (1983) summarised the distribution of wheat stem rust races from 1955 through 1966 on a worldwide basis and Green (1975) determined the evolution of virulence combinations in Canada. 1.2.4. History of wheat stem rust in South Africa The history of South Africa is rich in examples of the battle between the wheat industry and stem rust (Pienaar, 1975). Commercial production of wheat in South Africa dates back to 1658 just after the Dutch settlement in 1652 (Lombard, 1986). At the turn of the previous century Dr. E.A. Nobbs, agricultural advisor at the Cape, made the first crosses at the Robertson Experimental Station to improve the milling quality of South African wheat. In 1891, twelve wheat cultivars where imported from England together with the local cultivar ‘Du Toits.’ Ten of these cultivars were ‘rust- proof’ in England, but were severely damaged in trials at Somerset East (Lombard, 1986). The first cultivars bred in South Africa were Union, Darlvan and Nobbs, which Literature review 5 were released in 1910 (Pienaar, 1975). In 1925, the cultivars Bobriet and Gluretty were released by Neethling with stem rust resistance derived from the Italian cultivar Rieti (Pienaar, 1975). In 1948, the cultivar Hoopvol was released because its horizontal resistance to stem rust has withstood the test of time (Pienaar, 1975). In 1930 several physiological forms of P. graminis f. sp. tritici were recorded (Verwoerd, 1935). The first major stem rust epidemic occurred in 1726, when the entire crop was lost (Lombard, 1986). It has been suggested that urediniospores were carried and dispersed with the cereal and grass hay that accompanied livestock on ships en route to South Africa. When urediniospores are well dried and protected from temperature and humidity variations, they can remain viable for a long time (Lombard, 1986). A well-documented epidemic occurred in 1957/1958, when winter wheat production in the Free State was devastated by rust (Lombard, 1986). A potential epidemic threatened in the 1980’s due to the over-utilisation and geographically wide deployment of resistance gene Sr24 in South Africa (Le Roux & Rijkenberg, 1987a). Severe losses eventually occurred on Sr24 cultivars, particularly on wheat production in the Cape. In 1980, the Department of Agriculture decided to make stem rust resistance mandatory for all released cultivars and it became imperative to conduct annual pathotype surveys (Le Roux & Rijkenberg, 1987b). Stem rust pathotypes recorded in South Africa between 1935 and 2000 are documented in Table 2. 1.3 SYMPTOMS Raised orange-red pustules occur on leaves, leaf sheaths, stems and occasionally on glumes, awns and even seed of susceptible wheat cultivars (Scott, 1990). Figure 1 shows the orange-red pustules on the stem. Uredinial pustules are conspicuously erumpent, with shredded epidermal tissues at their margins. They may erupt through both leaf surfaces but tend to be larger on the lower surface (Wiese, 1987). The pustules are oval, elongated or spindle-shaped and up to 3 X 10 mm in size (Roelfs et al., 1992). Urediniospores are 15-24 x 21-40 µm in size, orange-red, dehiscent and oval, oblong or ellipsoid. Four median germ pores indent their thick, spiny walls (Wiese, 1987). Literature review 6 Teliospores are formed later in the season and the pustules become almost black in colour (Scott, 1990). Teliospores are ellipsoid to clavate, 15-20 x 40-60 µm in size and two-celled. They are tapered at their apex and have smooth, thick walls (Wiese, 1987). Germination of the teliospore normally follows after several weeks of cold dormancy and yields a hyaline basidium (promycelium) on which basidiospores develop on sterigmata (Wiese, 1987). The basidiospores are small, 7.6 x 6 µm, hyaline and oval-shaped. They are windborne and are spread to alternate hosts such as Berberis spp. (Roelfs, 1985). Pycnia on barberry are small, flask-shaped, sunken and the supporting leaf tissues are typically discoloured yellow-red. Pycnia produce pycniospores (1.6 x 3.6 µm) in small sticky droplets that attract insects (Roelfs, 1985). Aecia on the underside of barberry leaves are yellow and hornlike, projecting up to 5 mm from the leaf surface. Aeciospores produced by aecia in long dry chains are subglobose, 15-19 x 16-23 µm, smooth and light orange-yellow (Roelfs, 1985). Neither Berberis vulgaris nor the aecial stage of P. graminis f. sp. tritici is known to occur in South Africa. 1.4 EPIDEMIOLOGY 1.4.1. Life cycle In most areas of the world, the life cycle (Fig. 2) of P. graminis f. sp. tritici consists of continual uredinial generations. The fungus spreads by airborne urediniospores from one wheat plant to another and from field to field (Roelfs et al., 1992). Primary inoculum may originate locally (endemic) from volunteer plants or be carried long distances (exodemic) by wind and deposited by rain (Wiese, 1987). In certain areas snow can provide cover that occasionally permits the fungus to survive on winter wheat even at subzero temperatures (Roelfs & Long, 1987). Urediniospores germinate and produce a germ tube and an appressorium. Light stimulates the formation of a penetration peg that enters a closed stoma (Staples & Macko, 1984). The repeating asexual cycle then involves urediniospores producing uredinia in about a 14-day cycle under optimum conditions (Joshi & Palmer, 1973). Literature review 7 The sexual cycle seldom occurs except in certain regions of the United States. Although the sexual stage gives rise to genetic diversity, it also produces large numbers of individuals that are less fit (Roelfs & Groth, 1980). As the host matures, telia are produced directly from urediniospore infections or teliospores can be produced in mature uredinial pustules. Basidiospores germinate and penetrate barberry leaves directly. Infection results in the production of pycnia and formation of pycniospores. Pycniospores must mate to produce aeciospores (Luig & Watson, 1972). Aeciospores are hydroscopically released from the aecium and are airborne to wheat over distances of meters to perhaps a few kilometres. Aeciospores require similar conditions for infection to that of urediniospores. Infection by aeciospores results in the production of uredinia with urediniospores (Roelfs et al., 1992). Wind frequently transports urediniospores up to 100 km and sometimes up to 2000 km (Roelfs, 1985). Stem rust urediniospores are rather resistant to atmospheric conditions if their moisture content is moderate (20-30%). Long distance transport occurs annually 800 km across the North American Great Plains, nearly annually 2000 km from Australia to New Zealand, and at least three times in the past 75 years 8000 km from East Africa to Australia (Roelfs et al., 1992). 1.4.2. Inoculum sources In areas where no alternate host is found, e.g. Australia, Argentina and South Africa, stem rust oversummers on grasses or volunteer wheat. The amount of oversummering rust depends on the number of volunteer plants, the alternate hosts and the ability of urediniospores to travel over long distances (Roelfs, 1985). Sporulating uredinia are active in tropical and some subtropical areas throughout the winter. Occasional dormant mycelium may survive beneath the snow pack in northern temperate regions (Knott, 1989). In the northern hemisphere spring-sown wheat is particularly vulnerable in the higher latitudes if sources of inoculum are located downwind. Large areas of autumn-sown wheat occur in the southern American Great Plains, providing inoculum for the northern spring-sown wheat crop (Roelfs et al., 1992). According to Luig & Watson (1977) population shifts in the pathogen are mainly attributed to factors other than survival ability on grasses. In terms of epidemic potential it should be noted that a stem rust pustule can produce 10 000 urediniospores per day (Roelfs et al., 1992). Literature review 8 1.4.3. Environmental conditions The development, extent and severity to which rust develops are, apart from host and pathogenicity factors, strongly influenced by the environment. Suitable temperature and moisture conditions, as well as a susceptible host, are necessary to ensure the development of the different spore types. In addition, the direction and velocity of winds ensure their rapid dissemination and distribution (Staples & Macko, 1984). The development and nature of rust are dependent on: • Abundant viable inoculum. • Rapid distribution of such inoculum. • Favourable climatic conditions, e.g. sufficient moisture (from rains, mist or dews) for germination, infection and rapid development of successive cycles of urediniospores. • Dense and widespread population of susceptible hosts. • The stage of growth of the susceptible host. High temperatures and abundant moisture (especially light drizzles, heavy mists or dews) favour rapid rust development while cool, dry weather tends to retard such development. The minimum, optimum and maximum temperatures for spore germination are 2°C, 15-24°C and 30°C and for sporulation, 5°C, 30°C and 40°C, respectively (Hogg et al., 1969). According to Knights & Lucas (1980), fully dehydrated spores failed to germinate and there is a positive correlation between hydration time and percentage spore germination. Stem rust is more important late in the growing period, on late-sown and maturing wheat cultivars and at lower altitudes. In warm, humid climates, stem rust can be especially severe due to the long period of favourable conditions for disease development when a local inoculum source is available (Wiese, 1987). Stem rust requires a relatively long dew period (6 to 8 h). Maximum infections are obtained with 8-12 h of dew at 18°C followed by 120 µE/m2/s light while the dew slowly dries and the temperature rises to 30°C (Rowell, 1984). Fluctuations in moisture conditions or paucity of viable inoculum may result in a light epidemic or confine the development of rust to small areas or patches in a field. The occurrence of suitable climatic conditions will determine the subsequent development and extension of rust from such centres (Roelfs et al., 1992). Literature review 9 Rust severity may be further increased by excessive nitrogen fertilization resulting in denser stands and delayed maturity (Knott, 1989). The locality and nature of soil conditions may be a factor only in as far as it affects moisture and temperature conditions and the period of growth of the cereal host plant (Staples & Macko, 1984). 1.5 ECONOMIC IMPORTANCE The damage caused by wheat stem rust can be more spectacular than any other cereal disease. Millions of hectares of an apparently healthy crop with a high yield potential can be totally destroyed in less than a month (Roelfs, 1985). Stem rust is a destructive disease in most wheat regions of the world. Its severity globally is well documented in published information from India, China, South Africa, Asia, America Europe and East Africa (Harder, Mathenge & Mwaura, 1972; Bahadur, 1985; Knott, 1989; Roelfs et al., 1992). The fear of stem rust is understandable because a crop could be reduced to broken stems and shrivelled grain by the time of harvest (Roelfs et al., 1992). The four basic components of grain yield in wheat are the number of heads or productive tillers per square meter, the number of kernels per head, the number of kernels per spikelet, and the mass of individual kernels (Teng & Gaunt, 1980). The number of tillers and spikelets per tiller are determined before booting, the kernels per spikelet from stem elongation to the early milk stage, and the kernel mass after anthesis. Therefore, stress factors influencing the plant at different growth stages affect the respective yield components differentially (Teng & Gaunt, 1980). Lesions of rust can occupy a significant portion of the host plant tissue. Most of the sites where infection occur are sources of nutrients that are transported to the developing grain (Wiese, 1987). Stem rust is the most devastating of the rust diseases and can cause losses of 50% in one month when conditions for its development are favourable. Losses of 100% can occur on susceptible cultivars (Roelfs et al., 1992). When stem rust is extremely severe on a portion of the stem, the straw may break or lodge. In the U.S.A., stem rust epidemics have occurred over large areas on the Central plains. It was eventually controlled by the development and production of resistant cultivars. Windborne urediniospores, nevertheless, resulted in devastating epidemics of stem rust on wheat in 1935, 1937, 1953 and 1954 and the last major Literature review 10 epidemic occurred in 1954 (Roelfs & Martens, 1988). In 1954, a combined epidemic of leaf and stem rust probably caused losses of more than $500,000,000 in Canada and the U.S.A. (Knott, 1989). Green and Campbell (1979) estimated the annual savings from growing rust-resistant cultivars on the Canadian prairies to be $217 million. During the years 1930, 1931, 1948 and 1955, epidemics of stem rust of wheat caused significant losses of crops in Greece (Skorda, 1966). Rees & Syme (1981) reported that severe stem rust epidemics occurred on the cultivar Oxley, while grain yields reduced by 50%. Despite rapid changes in virulence in the pathogen, damage has been minimised by the timely release of resistant cultivars. On three occasions there have been sudden changes in rust races in Australia that appear to have originated from Southern Africa (Luig, 1983). In Australia severe stem rust epidemics in the 1880’s resulted in sufficient public concern and political pressure to combat the disease. Crop losses ranged from 30% to 55% in wheats susceptible to both stem and leaf rust. In 1973, national losses due to stem rust ranged from $A100-200 million (McIntosh et al., 1995). In a study done by Pretorius (1983), the effect of stem rust on the reduction of 1000 kernel mass and plot yield was between 7% and 35% depending on the cultivar and environmental conditions. 1.5.1. Global threat of stem rust pathotype Ug99 The deployment of the 1BL.1RS translocation, containing the Sr31 gene, provided the main component for stem rust resistance in many wheats for more than 30 years (Wanyera et al., 2006). Recently, isolates of P. graminis f. sp. tritici with virulence to Sr31 (pathotype Ug99) were detected in Uganda in 1999 (Pretorius et al., 2000). The detection of the new pathotype poses a major threat to most of the wheat production areas of the world. According to surveys, virulence for Sr31 is now widespread in the Eastern African highlands (Wanyera et al., 2006). During 2001, the cultivar Shinna became highly susceptible to stem rust in Ethiopia as well as the two major cultivars grown (56% of small scale production). In 2002 the yield losses in Kenya were estimated as much as 71% under experimental conditions (Global Rust Initiative, 2005). During January 2007, pathologists announced that Ug99 had crossed the Red Sea into the Arabian Peninsula and that it now threatens the major wheat production Literature review 11 areas of Asia (Stokstad, 2007). About 70% of U.S. wheat varieties are thought to be susceptible to Ug99. Between 70 and 75% of wheat grown in India and Pakistan are also susceptible and wheat in Egypt and China is thought to have similar vulnerabilities (Jin & Singh, 2006). 1.6 DISEASE CONTROL Plant disease control strategies should be directed at reducing the probability of epidemics, as well as reducing the magnitude of losses. Integrated cereal rust management, including cultural control practices, genetic resistance and fungicide applications, should contribute to the successful control of stem rust. Due to the airborne nature of stem rust spores, quarantine measures usually delay, but do not prevent entry of pathotypes with specific virulence combinations into new areas (Roelfs, 1985; Knott, 1989 and Roelfs et al., 1992). The profitability of any crop protection strategy depends on the prevalence of the disease, eventual yield loss, cost and efficacy of control measures (Fabre, Plantegenest & Yeun, 2007). 1.6.1. Cultural practices Cultural practices provide an alternative measure for reducing the risk of wheat rust epidemics, but no single practice is effective under all conditions and a series of activities are necessary for effective control (Roelfs et al., 1992). Cultural methods focus on early maturing cultivars and early planting of spring wheat (Roelfs, 1985). Delaying planting and the removal of volunteer grasses a few weeks before planting may prevent early infections and reduce the primary inoculum (Knott, 1989). Whatever the situation, each cultural practice must be tested against the anticipated types of epidemic that occur in the area (Roelfs et al., 1992) Dill-Macky & Roelfs (2000) found that reduced stand densities may promote the development of stem rust in barley. High levels of stem rust occasionally occurred in commercial fields where sparse stands are encountered. Diversity in the cultivars grown on a farm and spacing between fields can provide substantial benefits. On large farms, it may help if fields are arranged so that early maturing cultivars are down-wind from late maturing cultivars (Roelfs et al., 1992). Avoiding excess nitrogen and frequent irrigation are helpful in controlling stem rust (Roelfs, 1985). Literature review 12 Separation of winter and spring wheat grown in the same area by another crop, can delay the spread between fields (Roelfs, 1985). 1.6.2. Barberry eradication Eradication of the alternate host was started in Rouen, France, in 1660. Successful eradication programmes in Northern Europe and the North Central States of the U.S.A. are well documented. Except for Eastern Europe and the North-Western U.S.A., no areas of the world are known where alternate hosts play a role in stem rust epidemiology (Roelfs et al., 1992). After the disastrous epidemic of 1916, laws against the growing of barberry (Berberis vulgaris L.) were passed in the important wheat-producing areas of the U.S.A. A cooperative federal and state programme on barberry eradication was started in 1918, with E.C. Stakman as its head (Roelfs, 1982). Eradication of common barberry from the wheat fields of the central Great Plains of North America was completed, for all practical purposes, by 1928. The effect of this eradication on populations of stem rust on the source of local, often early epidemics of stem rust, was eliminated. Barberry eradication increased the average useful life of resistance genes in wheat (Roelfs & Groth, 1980). Barberry eradication affected the frequency and severity of stem rust epidemics by delaying disease onset by about 10 days, reducing the initial inoculum level, decreasing the number of pathogenic races and stabilising pathogenic races (Roelfs, 1985). 1.6.3. Fungicides Chemical control of cereal diseases is usually not desirable due to the high costs of fungicide application as well as potential environmental hazards. However, fungicides are used world-wide to maintain production levels in wheat cultivars lacking adequate levels of disease resistance (Ireta & Gilchrist, 1994). Chemicals have so far played a minor role in stem rust control (Knott, 1989). The history of fungicides in agricultural crops was described by Kuck, Scheinpflug & Pontzen (1995), Dunne (2002) and Russell (2006). The first foliar applications were done with elemental sulphur in the early 19th century to control powdery mildew of grape vines. The dithiocarbamates were patented in 1934 by Tisdale & Williams working for DuPont and these compounds developed into one of the most important classes of broad-spectrum, protectant fungicides. Between 1945 and 1970 other major classes of chemicals were introduced. In the mid-seventies the Literature review 13 DMI (demethylation inhibitors) fungicide group, which contains the triazole fungicides, was introduced. The STAR or strobilurin type fungicides were introduced in 1997. According to Hewitt (1998), chemical sprays on wheat and barley result in 87% of the total pesticide input into cereals which at the time were equivalent to $3967 million in sales. Since 1986, 16 out of 26 new fungicides have been aimed initially at the cereal sector. The leading cereal fungicides are mobile, specific compounds and are mainly triazoles or morpholines. • Foliar application The grain yield of cereals depends to a large degree on how long the leaves are able to remain green. Some fungicides have the ability to have a stay-green effect (Gordon & De Villiers, 1989). The reason for the minor role of fungicides in stem rust control can be the effectiveness of host resistance, the rate of disease increase for wheat stem rust under ideal conditions and the relatively low economic return per hectare of wheat in comparison to the cost of fungicide applications (Rowell, 1985). According to Nel et al. (1999) only tebuconazole (Folicur®) is registered for control of wheat stem rust in South Africa. However, ACDASA and CropLife South Africa list tebuconazole, triadimefon, propiconazole, and cyproconazole plus propiconazole, as registered wheat stem rust fungicides (ProCrop™ Professional software version 1.30). Gordon & De Villiers (1989) found that stem rust control with tebuconazole was comparatively modest, possibly because the lower parts of the stem may not have been adequately wetted, particularly in the later stages of growth when the flag leaf had appeared. According to research done by Loughman, Jayasena & Majewski (2005), the fungicide Folicur® was more effective at reducing disease and increasing yield or quality than Impact® or Triad®. Fungicides reduced stem rust severity on plant parts that were only slightly infected at that time, but were not effective on heavily affected plants. Fungicides applied at head emergence increased yield by 0.8-1.5 t.ha-1, depending on the control. Literature review 14 • Seed treatment The control of rust diseases by treating seed with a systemic fungicide is an attractive option. Research has shown that compounds with extremely low dosage response against the pathogen can be used effectively in controlling rust diseases (Rowell, 1985). Seed treatments can be used in combinations with foliar sprays when cultivars are very susceptible, alone when yields are too low to justify foliar sprays, and in combination with slow rusting resistance to stem rust (Knott, 1989). Hagborg (in Mills & Kotze, 1978) stated that the sulphone analogue, oxycarboxin or DCMO, reduced the incidence of stem rust significantly under field conditions. He also found that oxycarboxin 75% wettable powder (WP) applied at 4.2 kg/ha, controlled stem rust just as effectively as nickel compounds plus zineb. Formulations of DCMO plus thiram or carboxin/thiram have been developed for use on small grains (Mills & Kotze, 1978). Fungicide resistance often develops in fungal populations. The use of integrated pest management (IPM) practices, disease forecasting and research inputs may reduce the frequency of resistance development (Hewitt, 1998). A committee (FRAC, Fungicide Resistance Action Committee) has been established to review each fungicide type and to determine the compounds at risk. They have classified fungicides into groups according to their mode of action. Fungicides belonging to the same category can result in resistance if that specific group is used too frequently. Due to high costs the research into and development of new compounds, have been reduced to a few multi-national pesticide companies. The effective use of modern fungicides and thus the desired economic benefits, require a high level of education for all role players. The use of modern fungicides requires a more integrated approach and complete control of most diseases of wheat can be done by one to three applications of fungicides (Lyr, 1995). 1.6.4. Breeding for resistance Genetic resistance is the most effective, environment-friendly and economical way to control stem rust of wheat. The use of resistant cultivars adds no extra cost to farmers because there are no chemicals to buy or additional cultural operations to be carried out (Knott, 1989). Genetic resistance occurs when a resistance allele is present in the host along with a corresponding avirulence gene in the pathogen (Johnson & Knott, Literature review 15 1992). Several types of resistance to stem rust in wheat, including seedling resistance, adult-plant resistance, and slow rusting, have been described. The primary objective of a rust genetics programme is to understand the expression and inheritance of resistance and to know the range of available genetic diversity (McIntosh, 1988a). However resistance breeding is regularly confronted by genetic adaptation in the pathogen. Virulent pathotypes often increase in frequency and either render the resistant cultivars vulnerable to disease or actually cause crop losses (“boom and bust” cycles) (McIntosh, 1988a). 1.6.5. Types of resistance • Seedling resistance Seedling (qualitative) resistance, also known as complete resistance, usually protects the plant against avirulent pathogen isolates during their entire growing period. Seedling resistance is often of the pathotype-specific, major gene type (Knott, 1989). When used extensively over time and space, new pathotypes usually circumvent seedling resistance in a relatively short period after the release of such cultivars (Rajaram, Singh & Torres, 1988). The longevity of resistance based on major genes may be limited and stepwise mutation can eventually lead to susceptibility, but this strategy has been successfully employed in Australia (Rajaram et al., 1988). A range of designated and temporary designated Sr genes, controlling stem rust resistance, has been described (Table 3). Most of the seedling genes have become ineffective after their incorporation in cultivars in different countries, but by managing seedling resistance, the lifespan of these genes can be lengthened. The stacking of genes into a specific cultivar can provide protection for several years (Roelfs et al., 1992). In addition, post-seedling resistance genes used in combination with seedling resistance should reduce the rate of build-up of a new race with virulence on seedling resistance genes (De Pauw & Buchannon, 1975). • Adult-plant resistance (APR) Plants with APR are susceptible at the seedling stage and develop resistance in post-seedling phases (Roelfs, 1985). General resistance is effective against all races of the pathogen, while specific resistance (seedling) is effective against Literature review 16 some races and ineffective against others (Knott, 1982). Nazareno & Roelfs (1981) described the value of APR in protecting wheat cultivars against stem rust. Efforts should be made to combine those genes in a single cultivar that confer some degree of general resistance against the pathogen. Preliminary genetic studies of a large number of selected lines that have APR, indicated that resistance was generally recessive and controlled by several genes with small effects (Knott & Padidam, 1988). The APR gene Sr2, although less effective individually, proved to be a durable source of resistance in many parts of the world. There is no doubt that Sr2 provides a desirable genetic background into which more effective, but less durable genes can be placed (McIntosh, 1988b). Stem rust resistance in wheat conferred by Sr2 has remained effective against stem rust worldwide for more than 50 years (Spielmeyer & Lagudah, 2003). The presence of Sr2 in South African wheat cultivars such as Palmiet was considered valuable in terms of anticipated durability of resistance to stem rust of wheat (Pretorius & Brown, 1997). According to Sunderwirth & Roelfs (1980) Sr2 should be useful in combinations with other genes as a back-up resistance in high-risk stem rust regions. • Slow rusting Slow rusting is incomplete or quantitative resistance that is associated with a reduced rate of epidemic development (Roelfs & McVey, 1979). Slow rusting may be the result of fewer and smaller uredinia, longer latent periods and slower growing lesions resulting in less stem area infected (Martin et al., 1979). The ability to retard development of wheat stem rust is apparently effective in reducing yield losses (Wilcoxson, Skovmand & Atif, 1975). Some “defeated” resistance genes expressed significant residual effects in the form of reduced infections and in sporulation capacity. By combining a number of “defeated” race-specific resistance genes non-specific or rate-reducing resistance could be obtained (Brodny, Nelson & Gregory, 1986). Slow rusting results from limited growth in the host after penetration has occurred (Martin, Littlefield & Miller., 1977). According to Palmer & Wilcoxson (1982), infection frequency and latent period were not affected by the plant’s Literature review 17 anatomical and morphological characteristics, but that enlargement of uredinia and sporulation by the pathogen could be affected by plant structure. The term ‘slow rusting’ originates from the disease phenotype and has no genetic meaning. 1.6.6. Sources of resistance Due to the continual evolution of rust pathotypes, there must be a constant search for germplasm possessing resistance to the cereal rusts. Several natural sources of the specific type of resistance are available, namely cultivars, land races or primitive cultivars, wild or cultivated relatives and mutations (Dyck & Kerber, 1985). Numerous sources of resistance exist to rust diseases, although not all are of equal value (McIntosh et al., 1995). Genes conferring resistance to stem rust were identified in T. aestivum, T. boeoticum, T. durum, T. dicoccum, T. monococcum, T. timopheevi, T. turgidum, A. elongatum, Ae. squarrosa and S. cereale (Dyck & Kerber, 1985). When the various related species have a genome(s) that is homologous with at least one of the genomes of cultivated wheat, transfer of resistance is relatively simple (Dyck & Kerber, 1985). Bridge crosses may be used where the transfer of genetic material, usually between different levels of ploidy, is difficult or impossible by direct hybridisation (Dyck & Kerber, 1977). The use of cytogenetic procedures is available to genetically exchange chromosomes from distantly related species (Dyck & Kerber, 1985). 1.6.7. Molecular markers Markers used in plant breeding programmes fall into three broad categories, namely morphological linked disease resistance genes, biochemical and DNA based markers. Markers should be closely linked to the gene controlling an economically important trait (Eagles et al., 2001). The large scale application of markers that are available in wheat breeding are PCR based markers such as microsatellites (SSR) and STS markers developed from sequencing such RFLP, AFLP or RAPD markers (William, Trethowan & Crosby-Galvan., 2007). Some STS markers for resistance genes such as Sr24, 25, 26 and 38 are available and are implemented in the marker-assisted selection (MAS) wheat breeding programme at CIMMYT (William et al., 2007). Literature review 18 The use of pseudo-black chaff, or high-temperature-induced seedling chlorosis as morphological markers to detect the stem rust resistant gene Sr2, has been of economic importance in breeding programmes in Australia (Eagles et al., 2001). Other morphological markers used in their breeding programme are red-glumes associated with Yr10 and leaf tip necrosis (LTN) linked to Lr34 (Eagles et al., 2001). In South Africa, marker development has focused primarily on the leaf rust resistance gene Lr19 and stripe rust resistance in cv. Kariega. The use of markers in South Africa to identify useful rust resistance genes has been successful to some extent, but their application in breeding programmes could be improved (Pretorius et al., 2007). Useful markers are likely to become available for traits of importance in the near future. However, the cost associated with MAS assays is the limiting factor for its adoption in public wheat breeding programmes. The use of markers also depends on the breeding objectives and target traits (William et al., 2007). The complexity of the genome of hexaploid wheat has made it a relatively difficult species for marker application (Eagles et al., 2001). 1.6.8. Durability In breeding for resistance, the main objective is to produce cultivars with durable resistance. Plant breeders generally agree that breeding for resistance should not depend solely on race-specific genes. In a wheat breeding programme it would be desirable and in some cases essential, to incorporate a number of genes into one rust- resistant variety. Thus, resistance would be provided to more pathotypes and resistance to individual pathotypes would probably be increased (Knott & Anderson, 1956). Long-term resistance to stem rust is dependent on a continuing availability of resistance sources. Various procedures assist in ensuring that potential resistance sources carry new or different genes for resistance (McIntosh, 1988a). By interpreting pathogenic surveys and a reasonable knowledge of the genetics of resistance, a durable source of resistance could be found (Knott, 1989). Resistance that is race non-specific and controlled by a number of genes may be long-lasting because directional selection pressure on the pathogen will be minimal (Knott & Padidam, 1988). Although theoretically non-durable, losses due to stem rust have been successfully reduced through the use of wheat cultivars possessing vertical resistance (Eaton, McVey & Busch, 1984). Literature review 19 Knowledge of genetic variation for virulence and of host resistance effective to these variants, is the basis for utilisation of diverse resistance in the development of wheat cultivars resistant to the pathogen. It is well known that diversity can be introduced into a crop community at a number of levels and in a number of ways. One of the methods for extending the field life of varieties, the development of “multigene and multiline” cultivars, also requires information on the effectiveness of known genes for resistance against local virulences (Sawhney & Goel, 1981). Kolmer, Dyck & Roelfs (1991) researched stem rust resistance of wheat grown in Western Canada and found that cultivars with a combination of effective genes have been more resistant over a long period of time than closely related cultivars that have fewer of the same resistance genes. Breeding for durable resistance should be based on tests with pathotypes that enables the desired resistance genes to be selected (Lombard, 1986). This necessitates the regular surveying of wheat regions for prevailing pathotypes. 1.6.9. The effect of environment on host-pathogen interactions Watson (1970) proposed that virulence is controlled by specific genes whose products interact with those of corresponding resistance genes in the host, under a relatively simple genetic system. Growth rate, lesion size, spore production, aggressiveness and characters related to pathogenicity are controlled by a polygenic system. Watson (1970) further suggested that survival ability was related to characters controlled by a genetically similar, but different system. In reviewing the literature Browder (1985) stated that the mechanisms controlling host specificity are influenced by the environment. Loegering (1963) referred to the effects of environment upon both growth and expression of the host-pathogen interaction. Hart (1955) reported that both ‘Kenya 58’ and ‘Kenya 117A’ became moderately susceptible to race 15B at high temperatures. Loegering (1963) studied near-isogenic lines, with and without Sr6, at different temperatures and found that the lines possessing Sr6 were resistant to an avirulent pathogen culture only at low temperatures. Luig & Rajaram (1972) noted the interaction of higher temperatures with susceptible genetic backgrounds in decreasing the degree of resistance conferred by all the genes with which they worked. In a study on resistance controlled by Sr6, Knott (1981) concluded that a complex interaction involving genotype, temperature Literature review 20 and light exists. Knott & Anderson (1956) found that increased post-inoculation temperatures applied to seedlings with Sr6 resulted in reduced resistance following subsequent inoculation with avirulent cultures. Silverman (1959) concluded that the development of necrosis associated with rust infection was particularly sensitive to temperature. According to McIntosh et al. (1995) genes Sr6, 8b, 10, 12, 13, 14, 15, 17, 22, 23, 34, 36, and 38 are sensitive to environmental variation. In addition to qualitative differences in Sr gene phenotype, Browder (1985) concluded that quantitative variation in host response is likely to be the result of parasite-host-environment specificity. 1.7 CONCLUSIONS Wheat is the world’s most important crop and the rusts are present wherever wheat is grown. Stem rust is the most researched host-pathogen system in agriculture. However, much remains to be discovered in the areas of both applied and basic research. The occurrence of stem rust epidemics in South Africa depends largely on the ability of the fungus to over-summer on volunteer plants, environmental conditions suitable for pathogen survival and growth, and the cultivation of susceptible cultivars. Since there are no known alternate hosts for the fungus in South Africa, the pathogen changes through mutation and introduction of exotic pathotypes. Breeding resistant cultivars is the most effective method of stem rust control. Breeding efforts in South Africa should concentrate on combining quantitative resistance and effective seedling genes (gene-pyramiding). To increase the possibility of achieving durable resistance, breeders should select parents with satisfactory agronomical traits and proven long-lasting resistance. The deployment of cultivars with APR genes and a combination of seedling genes are recommended in areas where stem rust is a major problem. Effective genetic control of rust diseases requires a coordinated effort, including pathotype monitoring, collection and characterisation of sources of resistance and resistance breeding. More efficient traditional and molecular-based selection techniques should be maintained or developed. The continued development Literature review 21 of cultivars with a combination of different types of resistance can reduce inoculum and prevent the pathogen population from increasing to epidemic levels in future. The primary objective of this study was to identify fungicides toxic to P. graminis f. sp. tritici. A chapter on pathogenic variability of wheat stem rust is also included. Literature review 22 1.8 LITERATURE CITED Bahadur, P. 1985. Pattern of virulence in Puccinia graminis f. sp. tritici in India. Cereal Rusts Bulletin 13, 16-17. Boshoff, W.H.P., Van Niekerk, B.D. & Pretorius, Z.A. 2000. Pathotypes of Puccinia graminis f. sp. tritici detected in South Africa during 1991-1997. S.A. Journal of Plant and Soil 17, 60-62. Boshoff, W.H.P., Pretorius, Z.A., Van Niekerk, B.D. & Komen, J.S. 2002. First report of virulence in Puccinia graminis f. sp. tritici to wheat stem rust resistance genes Sr8b and Sr38 in South Africa. Plant Disease 86, 922. Brodny, U., Nelson, R.R. & Gregory, L.V. 1986. The residual and interactive expression of “defeated” wheat stem rust resistance genes. Phytopathology 76, 546-549. Browder, L.E. 1985. Parasite, host, environment specificity in the cereal rusts. Annual Review of Phytopathology 23, 201-222. Burdon, J.J., Marshall, D.R., Luig, N.H. & Gow, D.J.S. 1982. Isozyme studies on the origin and evolution of Puccinia graminis f. sp. tritici in Australia. Australian Journal of Biological Science 35, 231-238. De Pauw, R.M. & Buchannon, K.W. 1975. Postseedling response of wheat to stem rust. Canadian Journal of Plant Science 55, 385-390. Dill-Macky, R. & Roelfs, A.P. 2000. The effect of stand density on the development of Puccinia graminis f. sp. tritici in barley. Plant Disease 84, 29-34. Dunne, B. 2002. New fungicides and their role in disease control programmes. http://www.teagasc.ie/publications/2002/nattillageconf/paper06.htm. (accessed 2005/05/06) Dyck, P.L. 1992. Transfer of a gene for stem rust resistance from Triticum araraticum to hexaploid wheat. Genome 35, 788-792 Dyck, P.L. & Kerber, E.R. 1977. Chromosome location of gene Sr29 for reaction to stem rust. Canadian Journal of Genetics and Cytology 19, 371-373. Dyck, P.L. & Kerber, E.R. 1985. Resistance of the race-specific type. Pp. 469-500 in A.P. Roelfs & W.R. Busnell, eds. Cereal Rusts Diseases Vol. II. Distribution, Epidemiology and Control, Academic Press, Orlando. Eagles, H.A., Bariana, H.S., Ogbonnaya, F.C., Rebetzke, G.J., Hollamby, G.J., Henry, R.J., Henschke, P.H. & Carter, M. 2001. Implementation of markers in Literature review 23 Australian wheat breeding. Australian Journal of Agricultural Research 52, 1349-1356. Eaton, D.L., McVey, D.V. & Busch, R.H. 1984. Quantification of infection levels in wheat genotypes varying in stem rust resistance. Crop Science 24, 122-126. Fabre, F., Plantegenest, M, & Yeun, J. 2007. Financial benefit of using crop protection decision rules over systematic spraying strategies. Phytopathology 97, 1484-1490. Gerechter-Amitai, Z.K. & Wahl, I 1966. Wheat stem rust on wild grasses in Israel: Role of wild grasses in the development of the parasite and in breeding for resistance. National and University Institute of Agriculture, Rehovot, Israel. Global Rust Initiative. 2005. Sounding the alarm on global stem rust – An assessment of race Ug99 in Kenia and Ethiopia and the potential for impact in neighboring regions and beyond. www.globalrust.org (accessed 29/05/2005). Gordon, M.N.B. & De Villiers, V. 1989. Field trails with ®Folicur (tebuconazole) for the control of foliar, ear and stem diseases of wheat (Triticum aestivum) in the Republic of South Africa. Pflanzenschutz-Nachrichten Bayer 42, 91-120. Green, G.J. 1965. Stem rust of wheat, rye and barley in Canada in 1964. Canadian Plant Disease Survey 45, 23-29. Green, G.J. 1975. Some observations on the evolution of virulence in Puccinia graminis tritici in Western Canada. Rept. Tottori Mycol. Inst. (Japan) 12, 93- 97. Green, G.J. & Campbell, A.B. 1979. Wheat cultivars resistant to Puccinia graminis tritici in western Canada: Their development, performance and economic value. Canadian Journal of Plant Pathology 1, 3-11. Harder, D.E., Mathenge, G.R. & Mwaura, L.K. 1972. Physiologic specialization and epidemiology of wheat stem rust in East Africa. Phytopathology 62, 166-171. Harder, D.E., Dunsmore, K.M. & Anema, P.K. 1994. Stem rust on wheat, barley and oat in Canada in 1992. Canadian Journal of Plant Pathology 16, 56-60. Hart, H. 1955. Complexities of the wheat stem rust situation. Transactions of the American Association of Cereal Chemists XIII, 1-14. Hewitt, H.G. 1998. Fungicides in crop protection, Cab International, New York, 221pp. Literature review 24 Hogg, W.H., Hounam, C.E., Mallik, A.K. & Zadoks, J.C. 1969. Meteorological factors affecting the epidemiology of wheat rusts. WMO Tech. Note 99, 143pp. Ireta, M.J. & Gilchrist, S.L. 1994. Fusarium head scab of wheat (Fusarium graminearum Schwabe). Wheat Special Report No. 21b. Mexico, D.F.: CIMMIT. Jin, Y. & Singh, R.P. 2006. Resistance in U.S. wheat to recent Eastern African isolates of Puccinia graminis f. sp. tritici with virulence to resistance gene Sr31. Plant Disease 90, 476-480. Johnson, R. & Knott, D.R. 1992. Specificity in gene-for-gene interactions between plants and pathogens. Plant Pathology 41, 1-4. Joshi, L.M. & Palmer, L.T. 1973. Epidemiology of stem, leaf and stripe rust of wheat in Northern India. Plant Disease Replication 57, 8-12. Knights, I.K. & Lucas, J.A. 1980. Photosensitivity of Puccinia graminis f. sp. tritici urediniospores in vitro and on the leaf surface. Transactions of the British Mycology Society 74, 543-549. Knott, D.R. 1981. The effects of genotype and temperature on the resistance to Puccinia graminis tritici controlled by the gene Sr6 in Triticum aestivum. Canadian Journal of Genetics and Cytology 23, 183-190. Knott, D.R. 1982. Multigenic inheritance of stem rust resistance in wheat. Crop Science 22, 393-399. Knott, D.R. 1989. The Wheat Rusts – Breeding for Resistance, Springer Verslag, Berlin Heidelberg, 201pp. Knott, D.R. & Anderson, R.G. 1956. The inheritance of rust resistance. I. The inheritance of stem rust resistance in ten varieties of common wheat. Canadian Journal of Agricultural Science 36, 174-195. Knott, D.R. & Padidam, M. 1988. Inheritance of resistance to stem rust in six wheat lines having adult plant resistance. Genome 30, 283-288. Kolmer, J.A., Dyck, P.L. & Roelfs, A.P. 1991. An appraisal of stem and leaf rust resistance in North American hard red spring wheats and the probability of multiple mutations to virulence in populations of cereal rust fungi. Phytopathology 81, 237-242. Literature review 25 Kuck, K.H., Scheinpflug, H. & Pontzen, R. 1995. DMI fungicides. Pp. 205-258 in H. Lyr, ed. Modern Selective Fungicides, Properties, Applications and Mechanism of Action, Gustav Fischer Verlag, New York, 595pp. Leonard, K.J. & Szabo, L.J. 2005. Stem rust of small grains and grasses caused by Puccinia graminis. Molecular Plant Pathology 6, 99-111. Le Roux, J. 1985. First report of a Puccinia graminis f. sp. tritici race with virulence for Sr24 in South Africa. Plant Disease 69, 1007. Le Roux, J. 1989. Physiologic specialisation of Puccinia graminis f. sp. tritici in southern Africa during 1986-1987. Phytophylactica 21, 255-258. Le Roux, J. & Rijkenberg, F.H.J. 1987a. Pathotypes of Puccinia graminis f. sp. tritici with increased virulence for Sr24. Plant Disease 71, 1115-1119. Le Roux, J. & Rijkenberg, F.H.J. 1987b. Occurrence and pathogenicity of Puccinia graminis f. sp. tritici in South Africa during the period 1981-1985. Phytophylactica 19, 456-472. Loegering, W.Q. 1963. The relationship between host and pathogen in stem rust of wheat. Proceedings of the Second International Wheat Genetics Symposium. (Ed. J. MacKey.) Lund, Sweden 1963, 167-177. Lombard, B. 1986. Host-pathogen interactions involving wheat and Puccinia graminis tritici in South Africa. Phd Thesis, University of Stellenbosch, Stellenbosch, South Africa. Loughman, R. Jayasena, K. & Majewski, J., 2005. Yield loss and fungicide control of stem rust of wheat. Australian Journal of Agricultural Research 56, 91-96. Luig, N.H. 1977. The establishment and success of exotic strains of Puccinia graminis tritici in Australia. Proceedings of the Ecological Society of Australia 10, 89-96. Luig, N.H. 1983. A survey of virulence genes in wheat stem rust, Puccinia graminis f. sp. tritici. Berlin, Hamburg : Parey, 1983 (Advances in Plant Breeding, 11). Luig, N.H. & Rajaram, S. 1972. The effects of temperature and genetic background on host gene expression and interaction to Puccinia graminis tritici. Phytopathology 62, 1171-1174. Luig, N.H. & Watson, I.A. 1972. The role of wild and cultivated grasses in the hybridization of formae speciales of Puccinia graminis. Australian Journal of Biological Science 25, 335-342. Literature review 26 Luig, N.H. & Watson, I.A. 1977. The role of barley, rye and grasses in the 1973-74 wheat stem rust epiphytotic in Southern and Eastern Australia. Proceedings of Linn. Society. N.S.W. 101, 65-76. Lyr, H. 1995. Outlooks. Pp. 579-584 in H. Lyr, ed. Modern Selective Fungicides, Properties, Applications and Mechanism of Action, Gustav Fischer Verlag, New York, 595pp. Martin, C.D., Littlefield, L.J. & Miller, J.D. 1977. Development of Puccinia graminis f. sp. tritici in seedling plants of slow-rusting wheats. Transactions of the British Mycology Society 68, 161-166. Martin, C.D., Miller, J.D., Busch, R.H. & Littlefield, L.J. 1979. Quantitation of slow rusting in seedling and adult spring wheat. Canadian Journal of Botany 57, 1550-1556. McIntosh, R.A. 1988a. The role of specific genes in breeding for durable stem rust resistance in wheat and triticale. Pages 1-9 in: Breeding Strategies for Resistance to the Rusts of Wheat: N.W. Simmonds and S. Rajaram, eds. Cimmyt, Mexico, D.F., Mexico. McIntosh, R.A. 1988b. Catalogue of gene symbols for wheat. Proceedings of the Seventh International Wheat Genetics Symposium Vol 2. (Eds TE Miller & RMD Koebner.) pp 1225-1323. McIntosh, R.A., Wellings, C.R. & Park, R.F. 1995. Wheat rusts: An atlas of resistance genes. Kluwer, Dordrecht. 200pp. McVey, D.V., Long, D.L. & Roberts, J.J. 1996. Races of Puccinia graminis in the United States during 1994. Plant Disease 80, 85-89. Mills, L.J. & Kotze, J.M. 1978. Control of stem rust of wheat with systemic fungicides. Phytophylactica 10, 17-20. Nazareno, N.R.X. & Roelfs, A.P. 1981. Adult plant resistance of Thatcher wheat to stem rust. Phytopathology 71, 181-185. Nel, A., Krause, M., Ramautor, N. & Van Zyl, K. 1999. A guide for the control of plant diseases. National Department of Agriculture, Pretoria. 1ste edition. 122pp. Palmer, M.L.A. & Wilcoxson, R.D. 1982. Wheat peduncle structure in relation to slow rusting by Puccinia graminis f.sp. tritici. Phytopathology 72, 505-506. Literature review 27 Park, R.F. 1991. The University of Sydney Plant Breeding Institute, Wheat Rust Survey – 1990-1991. Plant Breeding Institute, Cobbitty Road, Cobbitty, N.S.W., 2570. Pienaar, R. 1975. Wheat breeding in South Africa: The history of wheat breeding in the Cape Province up to 1975. University of Stellenbosch, Department of Genetics, Stellenbosch, 7600. Pretorius, Z.A. 1983. Disease progress and yield response in spring wheat cultivars and lines infected with Puccinia graminis f. sp. tritici. Phytophylactica 15, 35-45. Pretorius, Z.A. & Brown, G.N. 1997. Detecting the Sr2-linked gene for seedling chlorosis in South African wheat cultivars. Department of Plant Pathology, University of the Orange Free State, South Africa. Pretorius, Z.A., Pakendorf, K.W., Marais, G.F., Prins, R. & Komen, J.S., 2007. Challenges for sustainable cereal rust control in South Africa. Australian Journal of Agricultural Research 58, 593-601. Pretorius, Z.A., Singh, R.P., Wagoire, W.W. & Payne, T.S. 2000. Detection of virulence to wheat stem rust resistance gene Sr31 in Puccinia graminis f. sp. tritici in Uganda. Plant Disease 84, 203. Rajaram, S., Singh, R.P. & Torres, E. 1988. Current Cimmyt approaches in breeding wheat for rust resistance. Pages 101-118 in: Breeding Strategies for Resistance to the Rusts of Wheat: N.W. Simmonds and S. Rajaram, eds. Cimmyt, Mexico, D.F., Mexico. Rees, R.G. & Syme, J.R. 1981. Epidemics of Stem rusts and their effects on grain yield in the wheat WW15 and some of its derivatives. Australian Journal of Agricultural Research 32, 725-730. Roelfs, A.P. 1982. Effects of barberry eradication on stem rust in the United States. Plant Disease 66, 177-181. Roelfs A.P. 1985. Wheat and rye stem rust Pp. 3-37 in A.P. Roelfs & W.R. Busnell, eds. Cereal Rusts Diseases Vol. II. Disease, Distribution, Epidemiology and Control, Academic Press, Orlando, 606pp. Roelfs, A.P. & Martens, J.W., 1988. An International system of nomenclature for Puccinia graminis f. sp. tritici. Phytopathology 78, 526-533. Literature review 28 Roelfs, A.P. & McVey, D.V. 1979. Low infection types produced by Puccinia graminis f.sp. tritici and wheat lines with designated genes for resistance. Phytopathology 69, 722-730. Roelfs, A.P. & Groth, J.V. 1980. A comparison of virulence phenotypes in wheat stem rust populations reproducing sexually and asexually. Phytopathology 70, 855-862. Roelfs, A.P. & Long, D.L. 1987. Puccinia graminis development in North America during 1986. Plant Disease 71, 1089-1093. Roelfs, A.P., Long, D.L. and Roberts, J.J. 1993. Races of Puccinia graminis in the United States during 1990. Plant Disease 77, 125-128. Roelfs, A.P., Singh, R.P. & Saari, E.E. 1992. Rust Diseases of Wheat: Concepts and methods of disease management. Mexico, D.F.: Cimmyt. 81pp. Rowell, J.B. 1984. Controlled infection by Puccinia graminis f. sp. tritici under artificial conditions. Pp. 292-332 in W.R. Busnell & A.P. Roelfs, eds. Cereal Rusts Diseases Vol. I. Origins, Specificity, Structure and Physiology, Academic Press, Orlando, 546pp. Rowell, J.B. 1985. Evaluation of chemicals for rust control. Pp. 561-589 in A.P. Roelfs & W.R. Busnell, eds. Cereal Rusts Diseases Vol. II. Disease, Distribution, Epidemiology and Control, Academic Press, Orlando, 606pp. Russell, P.E. 2006. The development of commercial disease control. Plant Pathology 55, 585-594. Sawhney, R.N. & Goel, L.B. 1981. Race-specific interactions between wheat genotypes and Indian cultures of stem rust. Theoretical Applied Genetics 60, 161-166. Scott, D.B. 1990. Wheat Diseases in South Africa. Department of Agricultural Development, South Africa, Technical Communication No. 220, 62pp. Silverman, W. 1959. The effect of variation in temperature on the necrosis associated with infection type 2 uredia of the wheat stem rust fungus. Phytopathology 49, 827-830. Singh, R.P. 1991. Pathogenic variations of Puccinia recondita f. sp. tritici and P. graminis f. sp. tritici in wheat-growing areas of Mexico during 1988 and 1989. Plant Disease 75, 790-794. Literature review 29 Skorda, E.A. 1966. Studies on the physiologic races of wheat stem rust (P. graminis tritici) in Greece during the eight years 1955-1962. Ann. Inst. Phytopath. Benaki, N.S. 7, 157-176. Smith, J. & Le Roux, J. 1992. First report of wheat stem rust virulence for Sr27 in South Africa. Vorträge für Pflanzenzüchtung 24, 109-110. Spielmeyer, W. & Lagudah, E.S. 2003. Rice genome sequence expedites fine mapping of durable, broad-spectrum stem rust resistance gene Sr2 in wheat (Triticum aestivum). In ‘Proceedings of the Tenth International Wheat Genetics Symposium’, Paestum, Italy. pp.414-416. Stakman, E.C., Stewart, D.M. & Loegering, W.Q. 1962. Identification of physiological races of Puccinia graminis var. tritici. U.S. Agricultural Research Service, A.R.S. E617, 1-53. Staples, R.C. & Macko, V. 1984. Germination of urediospores and differentiation of infection structures. Pp. 255-289 in W.R. Busnell & A.P. Roelfs, eds. Cereal Rusts Diseases Vol. I. Origins, Specificity, Structure and Physiology, Academic Press, Orlando, 546pp. Stokstad, E. 2007. Deadly wheat fungus threatens world’s breadbaskets. Sience 315, 1786-1787. Sunderwirth, S.D. & Roelfs, A.P. 1980. Greenhouse evaluation of the adult plant resistance of Sr2 to wheat stem rust. Phytopathology 70, 634-637. Teng, P.S. & Guant, R.E. 1980. Modelling systems of disease and yield loss in cereals. Agricultural Systems 6, 131-154. Wanyera, R., Kinyua, M.G., Jin, Y. & Singh, R.P. 2006. The spread of stem rust caused by Puccinia graminis f.sp. tritici, with virulence on Sr31 in wheat in Eastern Africa. Plant Disease 90, 113. Watson, I.A. 1970. Changes in virulence and population shifts in plant pathogens. Annual Review of Phytopathology 8, 209-230. Watson, I.A. & Luig, N.H. 1962. Selecting for virulence on wheat while inbreeding Puccinia graminis var. secalis. Proceedings of Linn. Society. N.S.W. 87, 39- 44. Wiese, M.V. 1987. Compendium of Wheat Diseases, Second Edition. St Paul, Minnesota: APS Press. Literature review 30 Wilcoxson, R.D., Skovmand, B. & Atif, A.H. 1975. Evaluation of wheat cultivars for ability to retard development of stem rust. Annals of Applied Biology 80, 275-281. Williams, H.M., Trethowan, R. & Crosby-Galvan, E.M. 2007. Wheat breeding assisted by markers: CIMMYT’s experience. Euphytica 157, 307-319. Literature review 31 Table 1. Virulence surveys of Puccinia graminis f. sp. tritici that is generally available in international literature. Country Year Reference* Australia 1990-91 Park (1991) Brazil 1982-85 Roelfs et al. (1992) Bulgaria 1974-78 Roelfs et al. (1992) Canada 1992 Harder et al. (1994) Czechoslovakia 1981-83 Roelfs et al. (1992) Egypt 1974-76 Roelfs et al. (1992) Ethiopia 1982-83 Roelfs et al. (1992) France 1977 Roelfs et al. (1992) Germany 1965-66 Roelfs et al. (1992) Greece 1955-62 Skorda (1966) Hungary 1969-72 Roelfs et al. (1992) India 1985 Bahadur (1985) Iraq 1967-69 Roelfs et al. (1992) Italy 1984 Roelfs et al. (1992) Kenya 1969-70 Roelfs et al. (1992) Korea 1971-72 Roelfs et al. (1992) Mexico 1982 Singh (1991) Pakistan 1976 Roelfs et al. (1992) Portugal 1980 Roelfs et al. (1992) Romania 1968-70 Roelfs et al. (1992) South Africa 1991-97 Boshoff et al., 2000 Spain 1968-71 Roelfs et al. (1992) USA 1994 McVey et al. (1996) USSR 1971 Roelfs et al. (1992) Uruguay 1968 Roelfs et al. (1992) Yugoslavia 1976-83 Roelfs et al. (1992) * Updated references Literature review 32 Table 2. Avirulence/virulence formulae for stem rust pathotypes in South Africa up to the end of 2000. Pathotype Avirulence/virulence formula (Sr-genes) 2SA2 Sr5,6,8b,9b,9e,17,24,27,30,36,38/Sr7b,9g* 2SA4 Sr8b,9g,24,27,36,38/Sr5,6,7b,9b,9e,11,17,30* 2SA6 Sr8b,9e,24,27,36,38/Sr5,6,7b,9b,9g,17,30* 2SA10 Sr6,7b,8b,9b,9e,17,24,27,30,36,38/Sr5,9g* 2SA18 Sr7b,8b,9b,9e,24,27,30,36,38/Sr5,6,9g,17* 2SA20 Sr8b,9e,24,27,30,38/Sr5,6,7b,9b,9g,17,36* 2SA32 Sr5,6,7b,8b,9b,9e,9g,17,24,30,36,38/Sr11* 2SA33 Sr7b,8b,9e,9g,17,24,27,30,38/Sr5,6,9b,36* 2SA36 Sr7b,8b,9e,9g,24,27,30,38/Sr5,6,9b,11,17,36* 2SA39 Sr5,8b,9b,9e,17,24,27,30,36,38/Sr6,7b,9g* 2SA43 Sr8b,24,27,36,38/Sr5,6,7b,9b,9e,9g,17,30* 2SA45 Sr8b,9e,9g,24,27,36,38/Sr5,6,7b,9b,17,30* 2SA48 Sr8b,9e,9g,24,27,30,36,38/Sr5,6,7b,9b,17* 2SA49 Sr8b,9e,9g,24,27,38/Sr5,6,7b,9b,17,30,36* 2SA51 Sr8b,9b,9e,9g,17,24,27,30,36,38/Sr5,6,7b * 2SA52 Sr8b,9e,24,27,30,36,38/Sr5,6,7b,9b,9g,17* 2SA53 Sr8b,24,27,38/Sr5,6,7b,9b,9e,9g,17,30,36* 2SA54 Sr7b,8b,9e,9g,24,27,36,38/Sr5,6,9b,17,30* 2SA55 Sr11/Sr5,6,7b,8b,9b,9e,9g,17,24,27,30,36,38** 2SA88 Sr24,27,36/Sr5,6,7b,8b,9b,9e,9g,17,30,38** 2SA100 Sr7b,8b,9e,9g,27,30,36,38/Sr5,6,9b,17,24* 2SA101 Sr8b,9e,9g,30,36,38/Sr5,6,7b,9b,9g,17,24* 2SA102 Sr5,6,7b,8b,9b,9e,11,17,24,36,38/Sr9g,27* 2SA102(K) Sr5,6,7b,8b,9b,9e,11,17,24,36,38/Sr9g,27,30,(Kiewiet)*** 2SA103 Sr5,6,7b,8b,9b,9e,9g,11,17,24,36,38/Sr27,30* * Le Roux & Rijkenberg, 1987b ** Boshoff et al., 2002 *** Unpublished data Literature review 33 Table 3. Designated and temporarily designated resistance genes for Puccinia graminis f. sp. tritici (http://www.ars.usda.gov) Sr Genome gene location Linkage Original source Tester line Remarks Designated 1 see Sr9d 2 3BS Triticum turgidum CnS(Hope3B) Few uredinia (Yaroslav emmer) APR 5 6DS Reliance ISr5-Ra 6 2DS Lr2, L 15 Red Egyptian ISr6-Ra Test at 18°C 7a 4BL Kenya 117 A Line G Sel 7b 4BL Marquis ISr7b-Ra 8a 6AS Red Egyptian ISr8-Ra 8b 6AS Barletta Benvenuto Barletta Benvenuto 9a 2BL Red Egyptian ISr9a-Ra 9b 2BL Kenya 117 A W2691 Sr9b 9d 2BL Hope ISr9d-Ra 9e 2BL Triticum turgidum Vernstein (Vernal emmer) 9f 2BL Chinese Spring Chinese Spring 9g 2BL Yr7 Lee CnSSr9g Literature review 34 Table 3 (cont.). Designated and temporarily designated resistance genes for Puccinia graminis f. sp. tritici (http://www.ars.usda.gov ) Sr Genome gene location Linkage Original source Tester line Remarks Designated 10 2B Egypt NA95 W2691 Sr10 11 6BL Lee ISr11-Ra 12 3BS Thatcher BtSr12Tc Test at 18°C 13 6AL Triticum turgidum W2691 Sr13 Test at 25°C (Kaphli emmer) 14 1BL Triticum turgidum Line A Sel (Kaphli emmer) 15 7AL Norka W2691 Sr15 Test at 18°C 16 2BL Thatcher ISr16-Ra 17 7BL Triticum turgidum CS (Hope7B) Test at 18°C (Yaroslav emmer) 18 1D Marquis LcSr18Mq 19 2BS Marquis LcSr19Mq 20 2BL Marquis LcSr20Mq 21 2AL T. monococcum Einkorn 22 7AL T. monococcum SwSr22 T.B. Literature review 35 Table 3 (cont.). Designated and temporarily designated resistance genes for Puccinia graminis f. sp. tritici (http://www.ars.usda.gov ) Sr Genome gene location Linkage Original source Tester line Remarks Designated 23 2BS Lr16 Exchange Exchange 24 3DL Lr24 A. elongatum BtSr24Agt 25 7DL Lr19 A. elongatum LcSr25Ars 26 6AL A. elongatum Eagle 27 3A Secalis cereale W2691 Sr27 (Imperial) 28 2BL Kota W2691 Sr28Kt 29 6DL Etiole de Choisy Pusa 30 5DL Pm2 Webster BtSr30Wst 31 1BL Lr26, Yr9 S. cereale (Imperial) Kavkaz 32 2A, 2B T. speltoides ER 5155 33 1DL Lr21 T. tauschii RL 5288 34 2A, 2B Yr8 T. comosa Compare 35 3AL T. monococcum Mq(2)5xG2919 36 2BS T. timopheevi W2691 SrTt-1 37 4AL T. timopheevi W2691 SrTt-2 Off type plants Literature review 36 Table 3 (cont.). Designated and temporarily designated resistance genes for Puccinia graminis f. sp. tritici (http://www.ars.usda.gov ) Sr Genome gene location Linkage Original source Tester line Remarks Designated 38 2AS Lr37, Yr17 T. ventricosa VPM1 Test at 18°C) 39 2B Lr35 T. speltoides RL 5711 40 2BS Lr13,23,16 T. araraticum RL 6087 41 4D Waldron Waldron 42 6D Norin 10 Norin 10 43 7D A. elongatum KS10-2 44 7DS A. intermedium Taf-2 45 1DS T. tauschii RL 5289 Temporary designated Bj Bejon Bejon Charter Charter Charter Dp-2 6AS Golden Ball Media Em Entrelargo de Montijo Gt Gamut Gamut H H-44 H-44 Literature review 37 Table 3 (cont.). Designated and temporarily designated resistance genes for Puccinia graminis f. sp. tritici (http://www.ars.usda.gov ) Sr Genome gene location Linkage Original source Tester line Remarks Temporary designated Kt2 2BL Kota Line AE Sel LC Little Club Little Club McN McNair 701 McNair 701 Satu Satu triticale Satu Tmp 4B Truimph 64 Truimph Tt-3 T. timopheevi Fed*2/SrTt-3 U 2D Red Egyptian CnsSrUre Wld1 Waldron BtSrWld Wld2 Webster Webster Zdar 1B Zdar A 2D Coteau Coteau B 2BL Coteau Coteau C 2B Len Len Literature review 38 Figure 1. Orange-red pustules of Puccinia graminis f. sp. tritici on the stem of a susceptible plant (image courtesy of Z.A. Pretorius). Literature review 39 Figure 2. The life cycle of Puccinia graminis f. sp. tritici as noted in the northern hemisphere (Roelfs et al., 1992). CHAPTER 2 PATHOTYPES OF PUCCINIA GRAMINIS F. SP. TRITICI DETECTED IN SOUTH AFRICA FROM 1998-2004 2.1 ABSTRACT A stem rust survey is conducted annually to monitor pathogenic variability in Puccinia graminis f. sp. tritici in the major bread wheat and triticale (X Triticosecale Wittmack) producing areas of South Africa. Data generated from pathotype surveys form an essential component of breeding programmes. Collections from trap nurseries and commercial wheat fields were multiplied on McNair 701 seedlings in a glasshouse. Single uredinium-derived isolates were inoculated onto a set of differential wheat lines and pathotypes determined according to the avirulence and virulence profile of each isolate. During 1998 to 1999 the incidence of stem rust on susceptible wheat and triticale cultivars and lines planted in trap nurseries was low. Four pathotypes were detected from 1998 to 1999, with 2SA4 (43%) and 2SA102 (45%) being dominant. A new pathotype, 2SA99, was found in 1999 and, in 2003, a new pathotype similar to 2SA102, but virulent to Kiewiet triticale (2SA102K), was detected. During the 2000 season, 40 single pustule isolates were established. Two new pathotypes were identified, namely 2SA88 and 2SA55. Pathotype 2SA55 was most frequently isolated, constituting 65% of all collections. In 2001 16 isolates were evaluated of which 2SA88 constituted 68%. Sixty two, 114 and 122 stem rust collections were analyzed in 2002, 2003 and in 2004, respectively. More than 70% of the isolates obtained from 2002 to 2004 were characterized as pathotype 2SA88, which unusually was virulent for Sr8b and Sr38. During the six year survey period, stem rust pathotypes 2SA4, 2SA55, 2SA88, 2SA99 and 2SA100 were detected on bread wheat and 2SA102, 2SA102K and 2SA103 on triticale. The steady increase in stem rust occurrence appears to be directly related to the market share of susceptible cultivars and continued breeding for durable resistance is required. Finally some suggestions are made to improve the stem rust differentiating set in South Africa. Pathogenic variability 41 2.2 INTRODUCTION Stem rust caused by Puccinia graminis Pers.:Pers. f. sp. tritici Eriks. & E. Henn. (Pgt) is an economically important disease of bread wheat (Triticum aestivum L.) worldwide (Roelfs, Singh & Saari, 1992; McIntosh, Wellings & Park, 1995). Variation in severity of annual rust epidemics is influenced by the rust species, environmental factors, cultural practices and cultivar resistance (Eversmeyer & Kramer, 2000). In South Africa wheat and barley (Hordeum vulgare L.) were first produced by Dutch settlers at the Cape of Good Hope during 1652 (Lombard, 1986). Currently bread wheat is produced in three production systems in South Africa, namely rain-fed spring wheat (Western and Southern Cape), irrigated wheat (Northern Cape, Northwest and KwaZulu-Natal) and rain-fed winter wheat (Free State). Stem rust infections have historically occurred in most production regions of South Africa but have more recently been restricted to spring wheat in the Western Cape. Stem rust has often reached epidemic proportions in South Africa (Lombard, 1986), with the most recent being severe and widespread infection of Sr24-derived wheat cultivars in 1985 (Le Roux, 1985; Le Roux & Rijkenberg, 1987a,b). However, due to efficient resistance breeding levels of stem rust in wheat crops, and the diversity of Pgt pathotypes recovered in surveys, have been lower during the past two decades (Boshoff, Van Niekerk & Pretorius, 2000). The timely detection of stem rust pathotypes with new virulence is considered important to the South African wheat industry and a shift in virulence that can contribute to cultivar susceptibility, and thus potential losses, is important to producers aiming for maximum yields. The first pathotyping of stem rust in South Africa was initiated by Verwoerd in the 1920s (Verwoerd, 1931; 1935). Although annual surveys were not conducted for most of the 20th century, available information on pathogenic variability was summarized by Lombard (1986). In South Africa, 22 pathotypes of Pgt that occur on wheat and triticale have been characterised between 1981 and 1997 (Le Roux & Rijkenberg, 1987a; Le Roux, 1989; Smith & Le Roux, 1992; Boshoff et al., 2000). The objectives of the surveys were to (a) determine the occurrence and distribution of stem rust in South Africa; (b) timeously detect new pathotypes originating either by means of mutation or exotic introduction; (c) determine the Pathogenic variability 42 avirulence/virulence reactions for each pathotype; (d) maintain an appropriate pathotype collection to test cultivars and breeders’ lines, and (e) enhance the knowledge of the pathogen and its distribution from year to year. In this chapter the results of pathogenicity surveys are presented for wheat, barley and triticale stem rust collections made during 1998-2004. 2.3 MATERIALS AND METHODS Wheat fields and disease nurseries were surveyed periodically from 1998 - 2004 by researchers of the ARC-Small Grain Institute (SGI), Bethlehem, South Africa, to monitor the occurrence, development and distribution of stem rust in South Africa. Disease nurseries, planted annually in the most important wheat production areas of South Africa, were visited at least once each wheat season. Disease nurseries included two universally susceptible cultivars McNair (stem and stripe rust susceptible) and Morocco (leaf, stem, and stripe rust susceptible), a stripe rust differential set, several tester lines carrying different stem and leaf rust resistance genes, a number of commercial wheat cultivars and advanced breeding lines. During surveys, sections of bread wheat, barley and triticale stems infected by stem rust, and exhibiting sporulating uredinia, were placed in glassine bags and transported to the SGI. In addition, rust samples were sent to the SGI by co-workers responsible for the planting and maintaining of disease nurseries. Details were recorded of the date of collection, location, cultivar if known and rust severity. On receipt of samples, urediniospores were harvested using a cyclone collector. The collected spores, suspended in a light-weight mineral oil (Soltrol® 170), were applied to primary leaves of wheat cv. McNair 701 seedlings which had previously been treated with maleic hydrazide (50 ml/pot of 0.3 gl-1) to enhance urediniospore production and to retard plant growth. Inoculated seedlings were placed in an incubation chamber, held at >96% relative humidity for a 14 h dark period, followed by a 3 h gradual drying period at a light intensity of 120 µE/m2/s. Temperature inside the chamber was maintained at 20 ±2°C. Plants were then moved to a greenhouse where natural daylight were supplemented with illumination from cool-white fluorescent tubes for 16 h each day and where temperature was controlled Pathogenic variability 43 at 22±2°C. After 10 days, urediniospores were collected from a single uredinium and increased on 7-d-old McNair 701 seedlings to produce sufficient spores to inoculate a differential set. The differential set consisted of Reliance (Sr5), McMurachy (Sr6), Marquis (Sr7b), ISr8A-Ra (Sr8a), Barleta Benvenuto (Sr8b), W2402 (Sr9b), Vernal (Sr9e), Acme (Sr9g), Yalta (Sr11), Renown (Sr17), Einkorn (Sr21), Gamka (Sr24), Coorong (Sr27), Festiguay (Sr30), Gamtoos (Sr31), W2691 (Sr36), VPM1 (Sr38), Taf-2 (Sr44) and McNair 701 (susceptible control). Inoculated sets were exposed to the same incubation procedures as described above. Stem rust pathotypes were identified according to their avirulence/virulence profiles. To confirm host responses to different Pgt pathotypes and compare differential sets, isolates representative of 2SA4, 2SA36, 2SA88, 2SA55, 2SA100, 2SA102, 2SA102K and 2SA103 were inoculated onto the South African and North American stem rust differential sets. Trident (Sr38), Kiewiet and Tobie (SrSatu), both triticales, were added to the South African set. Seed of the North American differentials were obtained from the ARS-USDA Cereal Disease Laboratory (CDL), St Paul, Minnesota. Original CDL stocks, without a cycle of seed multiplication, were used. These tests, which were conducted at the University of the Free State under conditions similar to those described above, allowed allocation of North American Pgt race codes. 2.4 RESULTS During 1998-2004, 395 single stem rust pustules were obtained from stem rust samples collected from wheat and triticale cultivars and lines throughout the small grain production areas of South Africa. Of the 395 isolates, 60% originated from the Southern Cape, 28% from the Western Cape, 6.3% from Mpumalanga, 3.1% from KwaZulu-Natal, 1.5% from Free State and 1.1% from Lesotho (Table 1). Eight stem rust pathotypes were detected, namely 2SA4, 2SA55, 2SA88, 2SA99 and 2SA100 from wheat, and 2SA102, 2SA102K and 2SA103 from triticale. The avirulence/virulence combination of each pathotype is presented in Table 3. Pathogenic variability 44 • 1998 to 1999 During 1998 to 1999 the incidence of stem rust was low. Fungicides were regularly used on the commercial fields throughout the season to control P striiformis f. sp. tritici. Two pathotypes were detected during the 1998 season, namely 2SA4 (66%) and 2SA103 (33%) out of six samples. A new pathotype, 2SA99, was identified during the 1999 season. A total of 36 samples were collected and pathotypes 2SA4 (39%), 2SA99(5%), 2SA102 (53%) and 2SA103 (3%) were recorded. • 2000 No incidences of stem rust in commercial fields were found, most probably due to below normal rainfall during the production season. Samples were collected only from trap nurseries in the Southern and Western Cape and 40 single pustule isolates were established. Two new pathotypes were identified, namely 2SA88, unusually virulent to Sr8b and 38, and 2SA55, virulent to Sr11 and 44. Pathotype 2SA55 was most frequently isolated, constituting 65% of all collections and 2SA88 (35%). • 2001 The Swartland area of the Western Cape experienced a wetter period from June to October and the climatic conditions were favourable for stem rust throughout the season. Fungicides were frequently applied to control Septoria diseases and in some instances eyespot. The Rûens area had little rainfall and crop losses were experienced due to drought stress. Only 16 samples were collected from these regions, yielding pathotypes 2SA88 (68%), 2SA99 (18%) and 2SA102 (12%). • 2002 Stem rust was first found in the Sandveld area of the Western Cape on the cultivar SST88 in September 2002. It was the first sample collected in commercial fields since the late eighties and only isolated pustules were found. Environmental conditions were favourable for stem rust in the Western Cape but in the southern Cape it was again a dry year. During the season, 62 samples were obtained from which pathotypes 2SA88, 2SA99, 2SA102 and 2SA103 were described. Pathotype 2SA88 consisted of 81% of the samples collected during this season. Pathogenic variability 45 • 2003 Weather records for the Swartland indicated the lowest rainfall in decades, while the rainfall in the Rûens was on average 140 mm more than the Swartland. In total 114 isolates of Pgt were processed from 57 samples collected during September to December. Most samples originated from the Western Cape. The most common pathotype was 2SA88, representing 85% of the pathotypes found. A new pathotype on the triticale cultivar Kiewiet was detected in the southern Cape near Caledon. Based on the differentials used its avirulence/virulence combination was the same as for pathotype 2SA102, therefore the new pathotype was described as 2SA102K. • 2004 One hundred and twenty two isolations were made from 64 field collections. Eight pathotypes were characterized. The Western Cape yielded pathotypes 2SA88 and 2SA102, the southern Cape 2SA36, 2SA55, 2SA88, 2SA99, 2SA100 and 2SA102 and in Mpumalanga pathotypes 2SA4, 2SA88, 2SA102 and 2SA103 were detected. During the summer, isolates were collected in the highlands of Lesotho and pathotypes 2SA88 and 2SA102 were identified. The predominant pathotype was 2SA88, consisting of 70% of isolates tested. 2.4.1. Comparison of differential sets The experiment in which the South African and North American differentiating sets were compared produced high quality infection types (Table 2, Figure 1). Except for subtle differences in stem rust phenotype, entries containing the same resistance gene mostly displayed similar differentiating abilities. Different responses were, however, obtained for the Sr9b and Sr21 tester lines. W2402-Sr9b, the South African differential, showed low infection types to pathotypes 2SA55 and 2SA102K. Line W2691-Sr9b, the North American entry, was susceptible to these two pathotypes (Figure 2). Based on the reaction of Einkorn, the original T. monococcum accession containing Sr21, no South African Pgt isolate had virulence for this gene. In contrast, pathotypes 2SA102 and 2SA103 produced high infection types on the North American tester Cns T. mono deriv. (Sr21) (Figure 3). The North American race Pathogenic variability 46 codes (Roelfs, Long & Roberts, 1993) for the pathotypes retested at UFS are given in Table 2. In addition to the above comparison, published data (Le Roux & Rijkenberg, 1987a,b; Le Roux, 1989; Boshoff et al., 2000) and unpublished ARC data were tabulated (Table 3). Infection types 0 to 2 were considered to indicate resistance (R) and 3 to 4 susceptibility (S). According to this comparison discrepancies occurred for classifying avirulence or virulence to Sr7b, 8a, 9a, 9b, 9d, 10, 11, 17, 21, 30 and Wld- 1. The variation was observed mostly for 2SA36, 2SA100 and the triticale pathotypes 2SA102, 2SA102K and 2SA103 (Table 2). 2.5 DISCUSSION In South Africa ideal conditions for stem rust incidence normally prevail in the Western Cape, especially the southern Cape coastal regions. In exceptional seasons stem rust occurs in other wheat production areas, but severe epidemics have been absent in the summer rainfall regions for many years. Similar to other countries, pathogenicity of the South African P. graminis f. sp. tritici populations indicate a considerable degree of variation. Such variability in rust pathogens is influenced by migration, mutation, recombination, selection and chance (McIntosh, 1992). During 1991-1996 no new pathotypes of stem rust were found. The low incidence of the disease can be attributed to resistance of the cultivar Palmiet (Sr2+Sr24) which comprised approximately 60% of the wheat cultivated in the Western Cape (Boshoff et al., 2000). The absence of rust in the southern parts of South Africa subsequently reduced inoculum movement to the interior of the country. During September and October, when the southern spring wheat crop is maturing, weather systems moving in a north-easterly direction commonly occur and it is assumed that spores are transported by these frontal systems to the summer rainfall regions. The years 1996 – 1999 produced only one pathotype namely 2SA99 (Boshoff et al., 2000). In 1996 the outbreak of P. striiformis f. sp. tritici resulted in a high incidence of stripe rust on the cultivar Palmiet. Stem rust prevalence was low partly because of the new cultivar SST57 (resistant to stem and stripe rust) and the widespread use of fungicides to control stripe rust. At that time the most dominant Pathogenic variability 47 pathotypes found in the area were 2SA4 and 2SA102. Boshoff et al. (2000) mentioned that the planting of susceptible triticale in the summer and winter rainfall areas resulted in an increase of pathotype 2SA102. The stripe rust epidemics seriously challenged wheat breeding and considerable time and effort were put into controlling this disease. Consequently stem rust resistance was overlooked in most breeding programmes. Based on survey data no or little stem rust was found in commercial fields in the 1990s. Since 2002 almost every commercial field in the Western and southern Cape had a substantial amount of stem rust infections. The increase in stem rust inoculum, and subsequent Pgt collections made by the ARC (Figure 4), can by attributed to the release of susceptible cultivars from 2000 onwards. Pathotypes 2SA88 and 2SA55 found in 2000 differed from those previously identified in South Africa. According to Boshoff et al. (2002), 2SA88 was the first local isolate to have virulence towards Sr8b and the T. ventricosum-derived gene Sr38. Since these virulences were uncommon in South Africa, it seems unlikely that 2SA88 was a mutant of existing pathotypes. Pretorius et al. (2007) mentioned that 2SA88 was similar to race Ug99 except for avirulence to Sr31. It is thus possible that 2SA88 originated from Eastern Africa and entered South Africa as an exotic introduction. Seedling and field reactions recorded for the barley cultivars Sterling, SSG532, Puma and Jaguar, showed an increase in susceptibility to 2SA55. The previously resistant cultivars SST57 (heterogeneous), Tugela, Tugela DN and PAN 3377 were susceptible in the seedling stage to pathotype 2SA88 (Boshoff et al., 2002). According to Boshoff et al. (2000) 2SA102 was expected to decline because of the release of the triticale cultivars Falcon, Korhaan, Arend and Kiewiet that were resistant to this pathotype. In 2003 an apparent single-step mutation in 2SA102 occurred and Kiewiet became susceptible. In seedling tests done at the Small Grain Institute, none of the other triticale cultivars or Sr genes was affected by this mutation. Based on recent infection studies, the original 2SA102 pathotype appeared avirulent to W2691-Sr9b. However, 2SA102K was virulent on W2691-Sr9b, suggesting that this pathotype mutated for virulence to both Sr9b and the unidentified gene in Kiewiet (Table 2). The authenticity of W2691-Sr9b, obtained from the Cereal Disease Laboratory in St Paul, must be considered correct (Y. Jin, ARS-USDA, personal communication). It is proposed that this line replaces W2402-Sr9b, which also contains Sr7b (Park, 2007). Pathogenic variability 48 The comparison of data sources suggested that South African rust pathologists should redefine its differential set and decide on entries that clearly and consistently distinguish between stem rust pathotypes. The phenotype of some genes giving variable results, e.g. Sr7b, Sr17 and Sr30, falls in the intermediate category and misclassification is easy if appropriate control reactions are absent. Furthermore, pathotypes such as 2SA36 and 2SA100 are inherently not as aggressive as others and seldomly produce typical ‘4’ infection types, even on susceptible hosts. What appears to be a 2 to 2++ infection type may in fact represent virulence. When compiling a new differential set, the seed should come from accredited sources. Although the North American set will be an obvious choice, present data showed that some of their entries can be replaced with other genotypes. For example, it is suggested that Combination VII be replaced with Renown as a tester for Sr17. The Bt/Wld (SrWld- 1) and CnSr9g lines were heterogeneous and should be reselected for purity or replaced. It may also be meaningful to retain the North American Cns T. mono deriv. as a source of Sr21 rather than Einkorn. Since Einkorn is not a high-yielding bread wheat, it is difficult to maintain sufficient seed sources and this accession appears to contain a gene in addition to Sr21. Pathotypes 2SA2, 2SA6, 2SA32, 2SA43 2SA45, 2SA48, 2SA101 were last found in 1987 (Le Roux & Rijkenberg, 1987a,b; Le Roux, 1989; Boshoff et al., 2000). Data presented in surveys done from 1991-1997 showed that pathotypes 2SA4, 2SA36, 2SA100 and 2SA102 prevailed during this time (Boshoff et al., 2000). During the current survey period it became evident that pathotype 2SA88 dominated the stem rust populations and was found on several occasions on the cultivar SST88 in commercial fields. With the exception of the new pathotypes 2SA55, 2SA88, 2SA99 and 2SA102K, it is not clear if the others survived independently or whether they originated from inoculated breeders’ plots. The fact that few stem rust samples were initially collected from commercial fields, seems to indicate that pathotypes such as 2SA4, 2SA36, 2SA100 and 2SA102 may have spread from experimental plots. According to research done at the Small Grain Institute, the high susceptibility on cultivars tested under field conditions during the 2003 survey showed that stem rust resistance is lacking. Forty two percent of the dryland spring cultivars and 86% of irrigation cultivars tested were susceptible to stem rust in the field. High levels of susceptibility in the seedling and adult plant stages were expressed by the dryland spring cultivars SST88, SST015 and SST027 as well as by the irrigation cultivars Pathogenic variability 49 Kariega, Olifants, Krokodil, CRN826, SST 806, SST 822 and SST876 (SGI, unpublished data). Winter wheats were only tested as seedlings and, as a group, appeared more resistance to stem rust than the spring types. The use of race–specific resistance to control wheat stem rust requires continued monitoring of the variation in the pathogen population for virulence (Dyck & Kerber, 1985). Long-term resistance to stem rust is dependent on a continuing availability of resistance sources. It is also unlikely that durable resistance will be achieved using single Sr genes. Breeders should access new germplasm on a continual basis and establish if potential resistance sources carry new or different genes for resistance (McIntosh, 1988). Worldwide virulence for Sr2, 13, 22, 24, 26, 27, 32, 33, 34, 35, 36, 37, 39, 40, 44, Tmp and Tt-3 is limited (Jin et al., 2007). In South Africa even fewer genes are effective against the local pathotypes. In 1980 the Department of Agriculture decided to make stem rust resistance mandatory for wheat. Unfortunately this decision has not been upheld in recent releases of commercial cultivars. The higher incidence of stem rust in commercial fields and experimental plots can be ascribed to the growing of susceptible cultivars in the Western Cape production areas. The detection of new stem rust pathotypes is of concern to the local small grain industry and requires continued research of and breeding for durable rust resistance. 2.6 ACKNOWLEDGEMENTS The ARC-Small Grain Institute, Bethlehem, South Africa, is thanked for the use of unpublished information. 2.7 LITERATURE CITED Boshoff, W.H.P., Van Niekerk, B.D. & Pretorius, Z.A. 2000. Pathotypes of Puccinia graminis f. sp. tritici detected in South Africa during 1991-1997. S.A. Journal of Plant and Soil 17, 60-62. Pathogenic variability 50 Boshoff, W.H.P., Pretorius, Z.A., Van Niekerk, B.D. & Komen, J.S. 2002. First report of virulence in Puccinia graminis f. sp. tritici to wheat stem rust resistance genes Sr8b and Sr38 in South Africa. Plant Disease 86, 922. Dyck, P.L. & Kerber, E.R. 1985. Resistance of the race-specific type. Pp. 469-500 in A.P. Roelfs & W.R. Bushnell, eds. Cereal Rusts Diseases Vol. II. Distribution, Epidemiology and Control, Academic Press, Orlando. Eversmeyer, M.G. & Kramer, C.L. 2000. Epidemiology of wheat leaf and stem rust in the Central Great Plains of the USA. Annu. Rev. Phytopathology 38, 491- 513. Jin, Y., Singh, R.P., Ward, R.W., Wanyera, R., Kinyua, M., Njau, P., Fetch, T., Pretorius, Z.A., & Yahyaoui, A. 2007. Characterization of seedling infection types and adult plant infection response of monogenic Sr gene lines to race TTKS of Puccinia graminis f.sp. tritici. Plant Disease 91, 1096-1099. Le Roux, J. 1985. First report of a Puccinia graminis f. sp. tritici race with virulence for Sr24 in South Africa. Plant Disease 69, 1007. Le Roux, J. 1989. Physiologic specialization of Puccinia graminis f. sp. tritici in southern Africa during 1986-1987. Phytophylactica 21, 255-258. Le Roux, J. & Rijkenberg, F.H.J. 1987a. Pathotypes of Puccinia graminis f. sp. tritici with increased virulence for Sr24. Plant Disease 71, 1115-1119. Le Roux, J. & Rijkenberg, F.H.J. 1987b. Occurrence and pathogenicity of Puccinia graminis f. sp. tritici in South Africa during the period 1981-1985. Phytophylactica 19, 456-472. Lombard, B. 1986. Host-pathogen interactions involving wheat and Puccinia graminis tritici in South Africa. Phd Thesis, University of Stellenbosch, Stellenbosch, South Africa. McIntosh, R.A. 1988. The role of specific genes in breeding for durable stem rust resistance in wheat and triticale. Pages 1-9 in: Breeding Strategies for Resistance to the Rusts of Wheat: N.W. Simmonds and S. Rajaram, eds. Cimmyt, Mexico, D.F., Mexico. McIntosh, R.A. 1992. Pre-emptive breeding to control wheat rusts. Euphytica 16, 103-113. McIntosh, R.A., Wellings, C.R. & Park, R.F. 1995. Wheat rusts: An atlas of resistance genes. Kluwer, Dordrecht. 200pp. Pathogenic variability 51 Park, R.F. 2007. Stem rust of wheat in Australia. Australian Journal of Agricultural Research 58, 558-566. Pretorius, Z.A., Pakendorf, K.W., Marais, G.F., Prins, R. & Komen, J.S. 2007. Challenges for sustainable cereal rust control in South Africa. Australian Journal of Agricultural Research 58, 593-601. Roelfs, A.P., Long, D.L. and Roberts, J.J. 1993. Races of Puccinia graminis in the United States during 1990. Plant Disease 77, 125-128. Roelfs, A.P., Singh, R.P. & Saari, E.E. 1992. Rust Diseases of Wheat: Concepts and methods of disease management. Mexico, D.F.: Cimmyt. 81pp Smith, J. & Le Roux, J. 1992. First report of wheat stem rust virulence for Sr27 in South Africa. Vortrage fur Pflanzenzuchtung 24, 109-110. Verwoerd, L. 1931. Die fisiologiese vorms van Puccinia graminis Pers. wat in Suid Afrika voorkom. South African Journal of Science 28, 274-279. Verwoerd, L. 1935. A review of the black stem rust (Puccinia graminis Pers.) situation. University of Stellenbosch, Sience Bullettin 138. Pathogenic variability 52 Table 1. Frequency of pathotypes of Puccinia graminis f. sp. tritici detected in South Africa during 1998-2004. Pathotypes Year and region 2SA4 2SA55 2SA88 2SA99 2SA100 2SA102 2SA103 1998 Southern Cape 4 0 0 0 0 0 2 % 66.6 0 0 0 0 0 33.3 1999 Western Cape 0 0 0 2 0 4 0 Southern Cape 10 0 0 0 0 15 1 Mpumalanga 4 0 0 0 0 0 0 % 39 0 0 5 0 53 3 2000 Western Cape 0 4 2 0 0 0 0 Southern Cape 0 22 12 0 0 0 0 % 0 65 35 0 0 0 0 2001 Western Cape 0 0 8 0 0 2 0 Southern Cape 0 0 3 3 0 0 0 % 0 0 68.75 18.75 0 12.5 0 2002 Western Cape 0 0 30 2 0 3 1 Southern Cape 0 0 2 0 0 3 3 Free State 0 0 6 0 0 0 0 Mpumalanga 0 0 12 0 0 0 0 % 0 0 81 3 0 10 6 2003* Western Cape 0 0 30 0 0 7 0 Southern Cape 1 0 56 1 1 2 0 Mpumalanga 0 0 2 0 0 0 0 KwaZulu-Natal 0 0 8 0 0 4 0 % 1 0 85 1 1 12 0 2004 Western Cape 0 0 12 0 0 4 0 Southern Cape 0 4 71 3 1 17 0 Mpumalanga 1 0 1 0 0 5 1 Lesotho 0 0 1 0 0 3 0 % 0.8 3.3 69.7 0.8 0.8 23.7 0.8 *During the 2003 survey 2 samples collected from the Southern Cape contributed to a new pathotype nl 2SA102K. Pathogenic variability 53 Table 2. Differentiation of Puccinia graminis f. sp. tritici pathotypes. Tester line Sr gene Source Pathotypea 2SA4 2SA36 2SA55 2SA88 2SA100 2SA102 2SA102K 2SA103 PSKS MJQS BNGQ PTKS PJBS GFBS BFGS GDBS MJRS TTKS PJCS Reliance 5 BV 2005 3+ 3 0; 4 3 0; 0; 0 I Sr5 Ra 5 CDL 2006 3++ 3 0; 3++ 3 0 0; 0 I Sr6 Ra 6 CDL 2006 3+ 3 ;1 4 3 ;1 ;1 ;1 Marquis 7b BV 2005 4 3+ 2 3++ 2++ 2+ 2+ 2,3 I Sr7b Ra 7b CDL 2006 3++ 3 2 3+ 2+ 2++ 2 3 I Sr8a Ra 8a BV 2005 3+ 2+ 4 3++ 3 4 4 4 I Sr8a Ra 8a CDL 2006 4 2+ 3++ 3++ 3 3++ 3+ 3+ Barletta Benvenuto 8b BV 1999 ;1 ;1 ;1 4 ; ; ;1 ;1 I Sr9a Ra 9a CDL 2006 4 2+ 4 4 3 4 4 4 W2042 9b BV 2005 3+ 2++ 2- 3+ 2+ 2 2 2 W2691 Sr9b 9b CDL 2006 3+ 2++ 3 3+ 2+ 2 3 2++ I Sr9d Ra 9d CDL 2006 4 3 4 3+ 3 4 4 3++ Vernal 9e BV 2005 3++ 0; ;1 4 ; ;1 1+ ;1 Vernstein 9e CDL 2006 4 ;1 1 4 ;1 1 1 1 Acme 9g BV 2005 1 ;1 2+ 4 1 4 4 2 Pathogenic variability 54 Tester line Sr gene Source Pathotypea 2SA4 2SA36 2SA55 2SA88 2SA100 2SA102 2SA102K 2SA103 PSKS MJQS BNGQ PTKS PJBS GFBS BFGS GDBS MJRS TTKS PJCS CnSr9g 9g CDL 2006 1,3 1 2+ 3++ 1+ 4 4 2 W2691 Sr10 10 CDL 2006 3++ 2++ 1 3++ 3 3+ 4 4 Yalta 11 BV 2005 4 ;1 3++ 4 3 2 2 1+ I Sr11 Ra 11 CDL 2006 3++ ;1 3++ 4 2- 1+ 1+ 1 Renown 17 GH 1999 3+ 2 X 4 2++ ;1 ;1 0; Combination VII 13,17 CDL 2006 2+ 1 ;1 2+ 1- ; ;1= ;1 Einkorn 21 GH 1999 1 ;1= 1 1 ;1 1 1 1 Cns T. mono deriv 21 CDL 2006 2+ 2 2 2+ 1- 3 2+ 2+3 Gamka 24 GH 2005 1 ;1 ;1 1 3 ; ; ; LcSr24Ag 24 CDL 2006 1 1 1 2 3 1 1 1 Coorong 27 GH 2004 ;1 ; 1- ; ;1 4 4 4 Festiquay 30 GH 1999 3+ 1 2 4 2= 2 2- 2+ BtSr30Wst 30 CDL 2006 4 1 2 3++ 1 2 2- 2+ Sr31/6*LMPG 31 CDL 2006 1 ; 2 1 ;1 1 1 1 W2691 36 BV 1999 0 3 ; 0 0 0 0 0 W2691 SrTt-1 36 CDL 2006 0 3 0 0 0 0 0 0 Pathogenic variability 55 Tester line Sr gene Source Pathotypea 2SA4 2SA36 2SA55 2SA88 2SA100 2SA102 2SA102K 2SA103 PSKS MJQS BNGQ PTKS PJBS GFBS BFGS GDBS MJRS TTKS PJCS Trident 38 BV 2002 X- ; ; 4 ; 0 ; ; McNair 701 McN BV 1998 3++ 3+ 4 4 3 4 4 4 Kiewiet GH 2004 ;1 ;1 1 ; ;1 1 3++ 1 Tobie Satu Triticale ; 0; ; 0 ; 0; 0; 0; Cns SrTmp Tmp CDL 2006 2 1 1 2+ 1 1 1 2 Bt/Wld Wld-1 CDL 2006 1 ; 3+ 1 3++ 3+ 3++ a Provisional North American race codes were allocated. Acme (Sr9g) and Renown (Sr17) replaced CnSr9g and Combination VII in the USA set. Variation between current and previous data resulted in uncertainty regarding the expression of Sr17 and Sr21. In those cases the alternative codes are also given. Pathogenic variability 56 Table 3. Comparison of Puccinia graminis f. sp. tritici pathotype classification in South Africa. Sr UFSa CDLb ARCc Boshoffd Le Rouxe UFSa CDLb ARCc Boshoffd Le Rouxe Gene 2SA4 2SA36 5 S S S S S S S S S S 6 S S S S S S S 7b S S S S S S S S S S 8a S S S S S R R R S S 8b R R R R R R R R 9a S S S S R S S 9b S S S S S S S S S S 9d S S S S S R S 9e S S S S S R R R R R 9g R R R R R R R R R R Pathogenic variability 57 Sr UFSa CDLb ARCc Boshoffd Le Rouxe UFSa CDLb ARCc Boshoffd Le Rouxe Gene 10 S S S S R S S S 11 S S S S S R R S S S 17 S S S S S R R S S S 21 R R R R R R R R R R 24 R R R R R R R R R R 27 R R R R R R R R 30 S S S S S R R R R R 31 R R R R R R 36 R R R R R S S S S S 38 R R R R R Tmp R R R R Wld-1 R S R S Pathogenic variability 58 Sr UFSa CDLb ARCc Boshoffd Le Rouxe UFSa CDLb ARCc Boshoffd Le Rouxe Gene McN S S S S Kiewiet R R R R Satu R R R R 2SA55 2SA88 5 R R R R S S S S 6 R R R S S S 7b R R R R S S S S 8a S S S S S S S 8b R R R S S S 9a S S S S S 9b R S R R S S S S 9d S S S S S Pathogenic variability 59 Sr UFSa CDLb ARCc Boshoffd Le Rouxe UFSa CDLb ARCc Boshoffd Le Rouxe Gene 9e R R R R S S S S 9g R R R R S S S S 10 S S S S S 11 S S S S S S S S 17 R R R R S S S 21 R R R R R R R R 24 R R R R R R R R 27 R R R R R R 30 R R R R S S S S 31 R R R R R R 36 R R R R R R R 38 R R R S S S Pathogenic variability 60 Sr UFSa CDLb ARCc Boshoffd Le Rouxe UFSa CDLb ARCc Boshoffd Le Rouxe Gene Tmp R R R Wld-1 R S S McN S S S Kiewiet R R Satu R R R 2SA100 2SA103 5 S S S S S R R R R 6 S S S R R R 7b R S S S S S S S 8a S S S S S S S S S 8b R R R R R R R 9a S S S S S S S Pathogenic variability 61 Sr UFSa CDLb ARCc Boshoffd Le Rouxe UFSa CDLb ARCc Boshoffd Le Rouxe Gene 9b R S S S R R R 9d S S S S S R S 9e R R R R R R R R R 9g R R R R R R R R R 10 S S S S S S S 11 S R S S S R R R R 17 R R S S S R R R R 21 R R R R R R R R R 24 S S S S S R R R R 27 R R R R S S S 30 R R R R S R S S S 31 R R R R R R R Pathogenic variability 62 Sr UFSa CDLb ARCc Boshoffd Le Rouxe UFSa CDLb ARCc Boshoffd Le Rouxe Gene 36 R R R R R R R R R 38 R R R R R Tmp R R R R Wld-1 R S S S McN S S S S Kiewiet R R Satu R R R 2SA102 2SA102K 5 R R R R R R R 6 R R R R R 7b R R S S R R S 8a S S S S S S S Pathogenic variability 63 Sr UFSa CDLb ARCc Boshoffd Le Rouxe UFSa CDLb ARCc Boshoffd Le Rouxe Gene 8b R R R R R 9a S S S S S 9b R R R R R S R 9d S R S S R 9e R R R R R R R 9g S S S S S S S 10 S S S S S 11 R R R R R R R 17 R R R R R R R 21 R S R R R R 24 R R R R R R R 27 S S S S S S Pathogenic variability 64 Sr UFSa CDLb ARCc Boshoffd Le Rouxe UFSa CDLb ARCc Boshoffd Le Rouxe Gene 30 R R S S R R R 31 R R R R R 36 R R R R R R R 38 R R R R Tmp R R R R Wld-1 S S S S McN S S S S Kiewiet R S S Satu R R R a University of the Free State tester lines used. b Cereal Disease Laboratory tester lines used. c Data collected from the ARC Small Grain Institute, Bethlehem. d Le Roux & Rijkenberg (1987a,b) and Le Roux (1989). e Boshoff, van Niekerk & Pretorius (2000) and Boshoff et al. (2002). Pathogenic variability 65 Figure 1. Primary leaf infection types displayed by selected stem rust differential lines to pathotypes from left to right, 2SA88, 2SA100, 2SA36, 2SA102, 2SA102K, 2SA55 and 2SA4 of Puccinia graminis f. sp. tritici. The following differential lines were photographed: Reliance (Sr5), Marquis (Sr7b), Vernal (Sr9e), Acme (Sr9g), Renown (Sr17), Coorong (Sr27), Festiguay (Sr30), Barletta Benvenuto (Sr8b), Trident (Sr30) and Yalta (Sr11). Pathogenic variability 66 Sr5 Sr7b Sr9e Sr9g Sr17 Sr27 Sr30 Sr8b Sr38 Sr11 Pathogenic variability 67 Figure 2. Variation in infection types described for Sr9b on stem rust differentials obtained from the University of the Free State and Cereal Disease Laboratory. From left to right: W2402 (Sr9b) and W2691 (Sr9b) infected with pathotypes 2SA102K, 2SA55 and 2SA88, respectively. Pathogenic variability 68 Figure 3. Variation in infection types described for Sr21 on stem rust differentials obtained from the University of the Free State and Cereal Disease Laboratory. From left to right: Einkorn (Sr21) and Cns T. mono deriv. (Sr21) infected with pathotypes 2SA88, 2SA102K and 2SA102, respectively. Pathogenic variability 69 140 120 100 80 60 40 20 0 1998 1999 2000 2001 2002 2003 2004 Years Figure 4. The gradual increase in number of stem rust isolates obtained and tested during the survey period 1998-2004. No. of isolates CHAPTER 3 IN VITRO TOXICITY AND RESIDUAL EFFECTS OF SELECTED FUNGICIDES TO PUCCINIA GRAMINIS F. SP. TRITICI 3.1 ABSTRACT The recent increase of stem rust in the Western Cape production area of South Africa has led to a renewed interest in the use of fungicides. At present only tebuconazole is registered to control stem rust in South Africa. To optimize in vitro testing conditions the effect of temperature on germination and germ tube growth of Puccinia graminis f. sp. tritici as well as incubation periods were evaluated. Seventy eight percent of urediniospores had germinated on water agar in petri dishes after 6 h of incubation at 20 and 25°C. Low germination rates were observed at 30°C. Twenty-nine fungicides in nine chemical classes were evaluated in vitro for toxicity to stem rust urediniospores. Water agar petri dish plates were amended with fungicides at concentrations ranging from 10 to 0.0001 μl active ingredient / ml medium. After incubation the percentage germination and average germ tube length were determined. Low EC50 (effective concentration that results in 50% inhibition) values were obtained with azoxystrobin, trifloxystrobin, kresoxim-methyl, mancozeb, azoxystrobin/difenoconazole, iprodione, chlorothalonil and hexoconazole. Fungicides such as quinoxyfen did not exhibit any toxic effect to urediniospores. When testing the effect of exposure time of stem rust urediniospores to selected fungicides, germination decreased when spores were exposed to trifloxystrobin and azoxystrobin/difenoconazole for 1 min and was completely inhibited after 30 min. Tebuconazole and flusilazole/carbendazim inhibited germination after 90 min. A shorter residual effect of trifloxystrobin occurred in wheat infected with stem rust in a glasshouse assay whereas longer efficacy periods were observed for azoxystrobin/difenoconazole and tebuconazole. In vitro fungicide toxicity 71 3.2 INTRODUCTION The occurrence of stem rust, caused by Puccinia graminis Pers.:Pers. f. sp. tritici Eriks. & E. Henn., has increased on commercial wheat in the Western Cape over the past seasons. This increased incidence can be attributed to the widespread cultivation of susceptible cultivars, and is of concern to the South African small grain industry. Fungicide sprays have become part of the production practice of wheat producers in the Western Cape. Fungicides used for the control of wheat leaf diseases can account for more than 20% of the variable cost of wheat production in the Western Cape and are an integral part of cereal crop management in that region. According to Nel et al. (1999) only tebuconazole is registered for control of wheat stem rust in South Africa. However, ACDASA and CropLife South Africa list tebuconazole, triadimefon, propiconazole, and cyproconazole plus propiconazole, as registered wheat stem rust fungicides (ProCrop™ Professional software version 1.30). Limited data are available on other fungicides which, in some instances, are used extensively by producers. Many fungicides are applied at specific growth stages and until recently, surveys have indicated that up to one-third of applications may be too late for optimal effectiveness. The severity of any disease is related to inoculum pressure, meteorological conditions and cultivar susceptibility (Cook, Hims & Vaughan, 1999). According to Milne et al. (2007) product choice and the timing of applications are often poor in commercial crops. Crops receiving up to three badly-timed sprays often suffer as much disease as untreated crops, suggesting that certain fungicides are applied too late or do not control the disease effectively. Research has often shown that fungicides applied during the period from flag leaf emergence to ear emergence (GS 37-59) offer the best prospects for cost-effective disease control on wheat (Bradley, 2004). The management of stem rust through resistant cultivars is undoubtedly the most desirable means of control. However, fungicides have been used against various diseases of cereals for over 100 years. With the introduction of systemic fungicides in the 1970’s, fungicides became widely used on a routine basis in cereal production. They have become established as an essential input in the growing of cereals and without them yields would be severely reduced (Dunne, 2002). Despite a progressive tightening of restrictions on chemical development and usage, such limitations can In vitro fungicide toxicity 72 also serve as a driving force in the discovery of new compounds (Sbragia, 1975). This is supported by the development of new fungicide classes with novel modes of action in the 1990’s. These include the strobilurins, phenylpyrroles, anilinopyrimides, phenoxyquinolines and compounds that trigger defence mechanisms in the plant (Knight et al., 1997). The East African epidemics of wheat stem rust due to race Ug99 of P. graminis f. sp. tritici (Pretorius et al., 2000) have stimulated a renewed interest in the search for effective and economically feasible chemicals to control this disease. Research on chemical control of stem rust has been documented, amongst others, by Mayfield (1985), Rowell (1985) and Loughman, Jayasena & Majewski (2005). Little research has recently been done to determine the direct toxic effect of fungicides to stem rust urediniospores in active lesions. The application of fungicides eliminating spores on plant surfaces, as well as through systemic activity, is beneficial from a chemical control perspective. The objective of this study was to optimize procedures for in vitro testing, to determine the toxic effect of fungicides labelled for use on rust diseases and to determine the efficacy of four foliar fungicides over time on wheat stem rust under controlled conditions. 3.3 MATERIALS AND METHODS 3.3.1. Rust pathogen. Inoculum of the stem rust pathotype UVPgt55 (virulent to Sr5, 6, 7b, 8b, 9b, 9e, 11, 17, 30, 38) was increased in a glasshouse on the susceptible cultivar Trident. Freshly collected urediniospores were used in all experiments. 3.3.2. Effects of temperature and incubation period on germination and germ tube growth of urediniospores. The effect of four temperatures, viz. 15, 20, 25 and 30°C on germination and germ tube growth of urediniospores of pathotype PgtUV55 incubated in the dark for 3, 6, 9, 12 and 15 h was examined. Water agar plates (1.5% and 9-cm diameter) and four replicates of each treatment were used. The plates were uniformly inoculated with 8 mg dry, freshly harvested urediniospores by placing them in a previously calibrated spore-settling tower. The spores were evenly dispersed by an air gun and allowed to settle for 3 min. In vitro fungicide toxicity 73 The inoculated plates were placed in four different growth chambers, each set at the desired temperature. The first group of 20 plates (four plates per temperature) was removed from the growth chambers after 3 h and continued until the final group was removed after 15 h. Upon removal the urediniospores were killed with lactophenol (Ellison et al., 1990). The percentage of germinated urediniospores was determined by microscopically counting 100 randomly selected spores on each plate. The spores were considered germinated if the tube was at least twice as long as the diameter of the spore. The length of germ tubes was measured at 20X magnification using a calibrated micrometer in the microscope eyepiece. The length of 10 germ tubes selected on each plate was measured. 3.3.3. Fungicide experiments Twenty nine fungicides, either registered, in experimental phase, or mentioned as having activity to rust diseases, irrespective of the host plant, were evaluated. • EC50 determination Water agar was autoclaved and allowed to cool to 50°C before a fungicide was incorporated into the medium. For each fungicide concentrations of 0, 0.01, 0.1, 1 and 10 μg active ingredient per ml of water agar were tested in 9-cm diameter petri dishes. All treatments were replicated four times. The pH of each fungicide-amended medium was recorded. Using a settling tower, the fungicide-amended petri dishes were inoculated with spores of pathotype UVPgt55 as described earlier. The inoculated petri dishes were incubated in the dark for 6 h at 20°C in a growth chamber. After incubation the urediniospores were killed and the percentage germination and germ tube length were recorded for each fungicide treatment. The EC50 values for each fungicide were determined according to Bliss (1935). For some compounds, e.g. the strobilurin fungicides, a lower concentration of 0.0001 μl a.i / ml agar was included. • Germination response to fungicide exposure Five fungicides were used to determine the effect of urediniospore survival after exposure for certain time intervals. The fungicides, namely Amistar Top® (azoxystrobin/difenoconazole), Punch Xtra® (flusilazole/carbendazim), Folicur® In vitro fungicide toxicity 74 (tebuconazole), Twist® (trifloxystrobin) and Legend® (quinoxyfen), and a water control, were included. Treatments were replicated four times. The fungicides were sprayed at recommended rates [azoxystrobin/difenoconazole (100/62.5 g a.i. ha-1), flusilazole/carbendazim (50/100 g a.i. ha-1), tebuconazole (187.5 g a.i. ha-1), trifloxystrobin (50 g a.i. ha-1) and quinoxyfen (125 g a.i. ha-1)] onto sporulating wheat seedlings. After 1, 5, 10, 30, 60, 90 and 120 min intervals the spores were collected with a cyclone collector. The collected spores were placed in 2 ml eppendorf tubes containing distilled, sterilized water. The tubes were centrifuged for 10 sec before collection of the supernatant. The process was repeated twice before the spores were pipetted onto water agar petri dishes. The inoculated petri dishes were placed in the dark in a growth chamber for 6 h at 20°C. The spores were killed by lactophenol after incubation and percentage germination and germ tube length were determined. The experiment was conducted twice. • Residual effects The wheat cultivar Trident, which is susceptible to pathotype UVPgt55, was grown in a sterilized soil-peat moss mixture in 9-cm diameter pots in a disease-free glasshouse cubicle. After emergence seedlings were thinned to ten plants per pot. Fourteen days after planting seedlings were sprayed with tebuconazole, flusilazole/carbendazim, trifloxystrobin and azoxystrobin/difenoconazole and a water control. Trade names, formulations and recommended rates are listed in Table 1. Applications were done with a pressurized knapsack sprayer at 200 kPa using hollow cone nozzles, recommended rates and water volumes simulating 300 ℓ ha-1. The fungicide-treated plants were kept in a glasshouse cubicle at 20± 4°C and fertilized weekly by a water- soluble solution of 2 g/ℓ Multifeed®. The first sets of plants were inoculated 3 h after fungicide applications. This was followed by consecutive inoculations 7, 14, 21 and 28 days after fungicide treatment. The plants (four pots per treatment) were inoculated with pathotype UVPgt55 using 8 mg of freshly collected urediniospores, suspended in light mineral oil. Inoculations were done in an enclosed inoculation booth. After drying for 2 h in an air-conditioned room, inoculated plants were placed in the dark in a dew chamber at 20±1°C and >96% relative humidity for 16 h. Upon removal from the chamber plants were dried down in a growth chamber at 20°C for 3 h before placement in a glasshouse cubicle at 20±4°C. Daylight was supplemented with cool-white In vitro fungicide toxicity 75 fluorescent tubes emitting photosynthetic active radiation of 120 μE/m2/s for 14 h each day. Fourteen days after inoculation plants were evaluated for stem rust incidence (percentage of plants showing symptoms), total number of pustules observed on the second leaf of each plant, infection type (0 to 4 scale, Stakman, Steward & Loegering, 1962) and pustule size. The length (mm) of 10 randomly selected pustules per leaf was measured using a digital caliper. The experiment was arranged as a randomized complete block with four replications. The trial was conducted twice. 3.3.4. Statistical analysis Data were analyzed for variance using the statistical software NCSS 2000 for Windows. Means were compared using Tukey’s least significant difference test (Steel & Torrie, 1980). 3.4 RESULTS AND DISCUSSION 3.4.1. Effects of temperature and incubation period on germination and germ tube growth of urediniospores The main objective for the first experiment was to determine the optimum temperature and dew period for in vitro germination of stem rust urediniospores. Significant (P ≤ 0.05) effects due to temperature, duration of incubation period and temperature x duration for germination were detected. Germination varied from 78.7% (20°C, 6 h incubation period) to 9.7% (30°C, 12 h incubation period). As temperature increased from 15 to 30°C, spore germination increased to a maximum of 78.3% at 20°C and then declined (Figure 1). Spore germination was not significantly affected by the duration of incubation. Germination was initiated at 3 h and reached a maximum after 6 h. Results demonstrated that urediniospores of P. graminis f. sp. tritici germinated over all temperatures tested but that the process was sub-optimal at 15 and 30°C. According to McIntosh, Wellings & Park, (1995) stem rust develops successfully within the range 18-30°C. Wiese (1987) stated that stem rust develops optimally near 20°C and is seriously hampered below 15°C and above 40°C. Spore In vitro fungicide toxicity 76 germination in vitro tests done by Rowell (1985) showed that the optimum temperature for stem rust incubation was at 18°C for 24 h. In general similar results were obtained for germ tube growth in response to temperature and incubation period (data not shown). According to the present data, the optimum incubation period for stem rust urediniospores to germinate was 6 h and this time was employed in further experimentation. 3.4.2. EC50 determination Germination of stem rust urediniospores was significantly affected by the fungicides tested and by the concentration of active ingredient used (Table 2). The EC50 values for germination and germ tube length differed significantly between fungicides (P ≤ 0.05). There was also a significant difference between concentrations for each fungicide tested. Of the 29 fungicides tested, azoxystrobin, trifloxystrobin, kresoxim-methyl, mancozeb, azoxystrobin/difenoconazole, iprodione, chlorothalonil and hexoconazole inhibited or almost restricted germination. EC50 values ranged from 0.001 to 0.004 µg a.i./ mℓ for the strobilurins, 0.118 to 20.944 for the triazoles, 0.002 to 199.1 for seed treatment fungicides, and 0.053 to 1060.679 µg a.i./ mℓ for other compounds tested. Some fungicides, e.g. cyproconazole, triticonazole, quinoxyfen and the experimental seed treatment Jockey Flex, gave high EC50 values for spore germination, while only quinoxyfen and Jockey Flex responded similarly with high EC50 values for germ tube length. The coefficient of variability for the latter two fungicides was exceptionally high, indicating a lack of accuracy in this method when using ineffective fungicides. The five strobilurin fungicides and the mancozeb seed treatments were fungicidal and killed the urediniospores at low concentrations. Eight fungicides from the triazole group, four fungicides from other groups and two experimental fungicides displayed low EC50 values. With the exception of triticonazole and Jockey Flex, the seed treatment chemicals were effective. Despite their efficacy, it is unlikely that seed treatments will be effective in controlling stem rust infections late in the season. Due to their toxicity the strobilurin fungicides had to be tested at a concentration lower than the originally planned 0.01 μg active ingredient per ml agar to determine their EC50 values. According to Dunne (2002) the strobilurin group is excellent inhibitors of spore germination and the present investigation supported this statement. In vitro fungicide toxicity 77 3.4.3. Germination response to fungicide exposure The experiments did not differ significantly and data were pooled. Germination was affected differentially when urediniospores were exposed to five fungicides for different lengths of time. The interaction between fungicide and time was significant (P ≤ 0.05) and fungicides influenced germination significantly. Over the duration of this experiment, the effects of trifloxystrobin and azoxystrobin / difenoconazole were similar, and the effects of tebuconazole and flusilazole / carbendazim were similar. Quinoxyfen did not have a toxic effect on stem rust urediniospores and gave results similar to the control (Figure 2). Urediniospores were killed by trifloxystrobin and azoxystrobin / difenoconazole when exposed for 30 min (Figure 2). In the tebuconazole and flusilazole / carbendazim treatments spore germination decreased gradually, completely inhibiting germination after 120 min exposure. In vitro fungicide efficacy has been evaluated by testing the ability of a compound to inhibit spore germination. The information is valuable in identifying fungicides that kill non-germinating urediniospores on contact. Therefore, four fungicides were selected according to their EC50 values determined in the first toxicity experiment. These fungicides were fungicidal to urediniospores and were presumably able to penetrate the spore wall to accumulate to a lethal dose. Trifloxystrobin and azoxystrobin/difenoconazole appeared to quickly enter the spore. Significant reductions in germination were observed after exposure for only 1 min reaching zero germination after 30 min of fungicide contact. In contrast, uptake of tebuconazole and flusilazole/carbendazim into the urediniospores took longer (Figure 2). The fungicide quinoxyfen, which was included as an ineffective control, did not have a toxic effect on the urediniospores and gave results similar to the water control. The identification of fungicides that kill quiescent urediniospores on contact can serve several purposes. These compounds will be useful in reducing spore viability in sporulating uredinia, thereby slowing disease progress within the crop as well as the potential development of fungicide resistant isolates (Mueller, Jeffers & Buck, 2005). The efficacy of fungicides in the triazole group against stem rust was expected. In vitro fungicide toxicity 78 Previously Rowell (1981) had indicated that triadimefon was effective for control of wheat stem rust epidemics on spring wheat in the North-central United States. Likewise, due to their fungitoxic effect on urediniospores, the strobilurins appear to have potential for stem rust control. At present no strobilurin fungicide is registered on wheat in South Africa and further field experimentation may prove useful. Questions often arise if in vitro exposure of rust mycelia to fungicides is similar to what can be expected in the host. The external concentration required for a lethal dose is not always an accurate measure of the actual efficacy of a chemical. According to Rowell (1985) fungicidal action of a compound is a function of both its ability to be taken up in the host plant and its inherent toxicity. 3.4.4. Residual effects In an analysis to test similarity of data in the two independent experiments, the effect of fungicides on pustule length (P=0.0005) and pustule size (P=0.0003) was significant. The data are therefore presented separately (Figures 3-6). The incidence of stem rust ranged from no visual infection (tebuconazole and azoxystrobin/difenoconazole) to an average of 90% of plants infected (trifloxystrobin and water control). Infection type differed between treatments over time. High infection types (3 and 4) were found for the control and trifloxystrobin 7 days after treatment while the treatment with flusilazole/carbendazim gave infection types of 1C and 2+3 on individual plants, 21 days after treatment. No significant difference was found for pustule size and number of pustules between the control and trifloxystrobin after seven days. Therefore, it can be assumed that trifloxystrobin has a low residual effect on plants infected with stem rust. After seven days, a slow increase in pustule number and size was found on plants treated with the fungicide flusilazole/carbendazim. Compared to the control, significantly lower values were found for pustule size and pustule number even after 28 days. In both experiments no infection was found on plants treated with tebuconazole. Low infection levels were found with azoxystrobin/difenoconazole after 28 days in the first experiment. Tebuconazole showed a high residual effect and provided protection throughout the duration of the experiment. From this study, it can be assumed that fungicides belonging to the triazole group have a longer residual effect in the plant than the fungicides from the strobilurin group. When combining the data from the in vitro and glasshouse experiments, the In vitro fungicide toxicity 79 strobilurin fungicides have an excellent ability to kill stem rust urediniospores on contact but showed a shorter residual effect in plants. The triazole fungicides took longer to kill stem rust urediniospores but exhibited a longer residual effect. Triazole derivatives have exhibited both fungicidal and plant growth regulating properties and are inhibitors of gibberellin biosynthesis (Goa, Hofstra & Fletcher, 1988). A promising formulation was observed for the triazole and strobilurin combination. The azoxystrobin/difenoconazole has also shown promising results on rust diseases on maize (F.J. Kloppers, Pannar Research, Greytown, personal communication). According to Mueller et al. (2005), azoxystrobin is known to be actively taken up by the plant cells and is upwardly systemic where it can provide protection from certain pathogens for up to 8 weeks. Continual pressure for improved safety with agricultural chemicals will provide a strong impetus for discovering alternative types of activity. The demand for new innovative chemicals is very strong, but chemicals alone are not the answer to reduce crop losses. An integrated programme includes sanitation, good cultural practices and resistant varieties, all linked to accurate disease forecasting and sensible applications of chemicals (Sbragia, 1975). 3.5 LITERATURE CITED Bliss, C.I. 1935. The calculation of the dosage-mortality curve. Annuals of Applied Biology 22, 134-167. Bradley, R.S. 2004. UK rules for growing wheat: Might Australian crops comply. http://www.grdc.com.au/growers/res_upd/south/s04/sylvester.htm (2007/02/22). Cook, R.J., Hims M.J. & Vaughan, 1999. Effects of fungicide spray timing on winter wheat disease control. Plant Pathology 48, 33-50. Dunne, B. 2002. New fungicides and their role in disease control programmes. http://www.teagasc.ie/publications/2002/nattillageconf/paper06.htm. (2005/05/06) In vitro fungicide toxicity 80 Ellison, P.J., Cullis, B.R., Bambach, R.W. & Kable, P.F. 1990. The effect of temperature on in vitro germination and germ tube growth of urediniospores of Tranzshelia discolor. Australian Journal of Agricultural Research 42, 479- 488. Gao, J., Hofstra, G. & Fletcher, R.A. 1988. Anatomic changes induced by triazoles in wheat seedlings. Canadian Journal of Botany 66, 1178-1185. Knight, S.C., Anthony, V.M., Brady, A.M., Greenland, A.J., Heaney, S.P., Murray, D.C., Powell, K.A., Schulz, M.A., Spinks, C.A., Worthington, P.A. & Youle,D. 1997. Rational and perspectives on the development of fungicides. Annual Review of Phytopathology 35, 349-372. Loughman, R. Jayasena, K. & Majewski, J., 2005. Yield loss and fungicide control of stem rust of wheat. Australian Journal of Agricultural Research 56, 91-96. Mayfield, A.H., 1985. Efficacies of fungicides for the control of stem rust of wheat. Australian Journal of Experimental Agriculture 25, 440-443. McIntosh, R.A., Wellings, C.R. & Park, R.F. 1995. Wheat rusts: An atlas of resistance genes. Kluwer, Dordrecht. 200pp. Meuller, D.S., Jeffers, S.N. & Buck, J.W. 2005. Toxicity of fungicides to urediniospores of six rust fungi that occur on ornamental crops. Plant Disease 89, 255-261. Milne, A., Paveley, N., Audsley, E. & Parsons, D. 2007. A model of the effect of fungicides on disease-induced yield loss, for use in wheat disease management dicision support systems. Annals of Applied Biology 151, 113-125. Nel, A., Krause, M., Ramautor, N. & Van Zyl, K. 1999. A guide for the control of plant diseases. National Department of Agriculture, Pretoria. 1ste edition. 122pp. Pretorius, Z.A., Singh, R.P., Wagoire, W.W. & Payne, T.S. 2000. Detection of virulence to wheat stem rust resistance gene Sr31 in Puccinia graminis f. sp. tritici in Uganda. Plant Disease 84, 203. Rowell, J.B. 1981. Control of stem rust on spring wheat by triadimefon and fenapanil. Plant Disease 65, 235-263. In vitro fungicide toxicity 81 Rowell, J.B. 1985. Evaluation of chemicals for rust control. Pages 561-589 in: The Cereal Rusts Vol. II. Distribution, Epidemiology and Control. A.P. Roelfs and W.R. Bushnell, eds. Academic Press, Orlando, 606pp. Sbragia, R.J. 1975. Chemical control of plant disease: an exciting future. Annual Review of Phytopathology 13, 257-269. Stakman, E.C., Steward, D.M. & Loegering, W.Q. 1962. Identification of physiologic races of Puccinia graminis f. sp. tritici. Agricultural Research Services E617, United States Department of Agriculture, Washington DC. Steel, R.G.D. & Torrie, J.H. 1980. Principles and procedures of statistics, a biometrical approach. McGraw-Hill Book Company, New York, 2de edition, 633pp. Wiese, M.V. 1987. Compendium of Wheat Diseases, Second Edition. St Paul, Minnesota: APS Press. Table 1. Fungicides registered or in experimental phase used that were evaluated for toxicity to stem rust urediniospores Fungicide category Active ingredient Trade name Formulation Rec. rate Labeled for rust Strobilurin Azoxystrobin Amistar® 250 SC 300ml/ha Yes Kresoxim-methyl Stroby® 50 WG 15g/ha No Trifloxystrobin Flint® 50 WG 15g/ha No Trifloxystrobin Twist® 500 SC 100ml/ha Yes Strobilurin / DMI triazole Azoxystrobin/Difenoconazole Amistar Top® 200/125 500ml/ha Yes DMI triazole Bitertanol Baycor® 300 EC 650ml/ha Yes Bromuconazole Granit® 100 EC 700ml/ha Yes Cyproconazole Alto® 100 SL 400ml/ha Yes Difenoconazole Score® 250 EC 350ml/ha Yes Epoxiconazole Opus® 125 SC 800ml/ha Yes Flusilazole Capitan® 250 EW 400ml/ha Yes Hexaconazole Anvil® 50 SC 480ml/ha Yes Propiconazole Tilt® 250 EC 500ml/ha Yes Tebuconazole Folicur® 250 EW 750ml/ha Yes Triadimefon Bayleton® 250 EC 500ml/ha Yes Triadimenol Baytan® 150 FS 150g/100kg No Triticonazole Flite® 200 FS 120g/100kg Yes DMI piperazine Triforine Denarin® 190 EC 150ml/ha Yes In vitro fungicide toxicity 83 Table 1 (cont) Fungicides registered or in experimental phase used that were evaluated for toxicity to stem rust urediniospores Fungicide category Active ingredient Trade name Formulationa Rec. rateb Labeled for rust DMI triazole/ Flusilazole / carbendazim Punch Xtra® 125 / 250 400ml/ha Yes benzimidazole Dihiocarbamate Mancozeb Dithane® 800 WP 150g/100kg Yes Mancozeb Ifax® 800 DS 160g/100kg Yes Phthalimide Chlorothalonil Bravo® 720 SC 190ml/ha Yes Dinitrophenyl Dinocap Karathane® 350 EC 50ml/ha No Multi site Quinoxyfen Legend® 250 SC 500ml/ha No Anilide Oxycarboxin Plantvax® 200 EC 1500ml/ha Yes Dicarboximide Iprodione Rovral Flo® 255 SC 200ml/ha No Experimentalc - Jockey Flex® 167 FS 450ml/100kg Yes experimental - JAU 6478 100 FS 100ml/100kg Yes experimental - JAU 6478 / HWG 250 EC 400ml/ha Yes experimental 1608 a Percentages of active ingredients in commercial products formulated as powder of dry seed treatment (DS), emulsifiable concentrate (EC), emulsion in oil or water (EW), flowable concentrate for seed treatment, suspension concentrate (SC), soluble concentrate (SL), water dispersible granule (WG) and wettable powder (WP) (Nel, Krause, Ramautor and van Zyl, 1999). b Recommended use (as per chemical company) of fungicides per hectare or as seed treatment per 100kg of seed. c Fungicides still in experimental phase. In vitro fungicide toxicity 84 Table 2. The EC50 values for germination and germ tube length, with the average CV for each fungicide and rankings according to their performance. Fungicide category Active ingredient Germination Germ tube length (µm) EC50 CV Rank EC50 CV Rank Strobilurin Azoxystrobin 0.002 0.0 6 0.002 0.0 5 Kresoxim-methyl 0.001 0.0 3 0.001 0.0 3 Trifloxystrobin 0.001 0.0 4 0.001 0.0 4 Trifloxystrobin 0.001 0.0 2 0.001 0.0 1 Strobilurin / DMI triazole Azoxystrobin/Difenoconazole 0.003 24.7 7 0.003 24.7 7 DMI triazole Bitertanol 22.972 9.5 24 2.210 4.1 23 Bromuconazole 6.835 11.9 22 1.198 7.9 22 Cyproconazole 1733.460 30.8 27 11.329 15.8 25 Difenoconazole 0.497 3.3 18 0.154 11.5 16 Epoxiconazole 0.569 7.1 20 0.225 8.9 21 Flusilazole 0.520 7.8 19 0.177 11.6 18 Hexaconazole 0.104 11.2 10 0.034 5.1 10 Propiconazole 18.973 7.3 23 23.924 19.5 26 Tebuconazole 0.312 11.6 15 0.131 5.7 14 Triadimefon 0.164 5.8 12 0.043 10.1 11 Triadimenol 28.619 23.4 25 25.569 36.1 27 In vitro fungicide toxicity 85 Fungicide category Active ingredient Germination Germ tube length (µm) EC50 CV Rank EC50 CV Rank Triticonazole 666.100 69.9 26 2.343 2.7 24 DMI piperazine Triforine 0.415 6.0 17 0.142 3.6 15 DMI triazole/ Flusilazole / carbendazim 0.310 1.4 14 0.193 2.1 19 benzimidazole Dihiocarbamate Mancozeb 0.002 0.0 5 0.002 0.0 6 Mancozeb 0.001 0.0 1 0.001 0.0 2 Phthalimide Chlorothalonil 0.063 11.4 9 0.032 6.4 9 Dinitrophenyl Dinocap 0.171 11.6 13 0.061 4.9 12 Multi site Quinoxyfen 1437491.9 172.0 28 320772.95 92.7 28 Anilide Oxycarboxin 0.325 3.5 16 0.163 2.2 17 Dicarboximide Iprodione 0.052 4.0 8 0.026 4.4 8 Experimental*** Jockey Flex 233784655 102.3 29 30371053 129.5 29 JAU 6478 0.116 6.0 11 0.096 6.0 13 JAU 6478 / HWG 1608 0.629 40.7 21 0.216 9.7 20 In vitro fungicide toxicity 86 100 80 60 40 3 hours 6 hours 20 9 hours 12 hours 15 hours 0 15°C 20°C 25°C 30°C Incubation temperature Figure 1. In vitro germination of urediniospores of Puccinia graminis f. sp. tritici as affected by incubation temperature and incubation time. The Tukey (P=0.05) value for comparison of means is 6.43. % Germination In vitro fungicide toxicity 87 100 80 60 40 20 0 1 10 20 30 40 50 60 70 80 90 100 110 120 Exposure time in minutes Tebuconazole Trifloxystrobin Flusilazole/carbendazim Azoxystrobin/difenoconazole Quinoxyfen Control Figure 2. The percentage germination of freshly collected urediniospores of Puccinia graminis f. sp. tritici following exposure to fungicides for different time periods. The Tukey (P=0.05) value for comparison of means is 6.14. % Germination In vitro fungicide toxicity 88 3 2 1 0 0 7 14 21 28 Days after fungicide treatment Tebuconazole Trifloxystrobin Flusilazole/carbendazim Azoxystrobin/difenoconazole Control Figure 3. The mean pustule size of stem rust on wheat plants sprayed with fungicides and inoculated at different time periods after fungicide application in experiment 1. The Tukey (P=0.05) value for comparison of means is 0.42. Pustule size (mm) In vitro fungicide toxicity 89 50 40 30 20 10 0 0 7 14 21 28 Days after fungicide treatment Tebuconazole Trifloxystrobin Flusilazole/carbendazim Azoxystrobin/difenoconazole Control Figure 4. The mean number of stem rust pustules on the 2nd leaf of wheat plants sprayed with fungicides and inoculated at different time periods after fungicide application in experiment 1. The Tukey (P=0.05) value for comparison of means is 7.86. Number of pustules In vitro fungicide toxicity 90 50 40 30 20 10 0 0 7 14 21 28 Days after fungicide treatment Tebuconazole Trifloxystrobin Flusilazole/carbendazim Azoxystrobin/difenoconazole Control Figure 5. The mean number of stem rust pustules on the 2nd leaf of wheat plants sprayed with fungicides and inoculated at different time periods after fungicide application in experiment 2. The Tukey (P=0.05) value for comparison of means is 5.93. Number of pustules In vitro fungicide toxicity 91 3 2 1 0 0 7 14 21 28 Days after fungicide treatment Tebuconazole Trifloxystrobin Flusilazole/carbendazim Azoxystrobin/difenoconazole Control Figure 6. The mean pustule size of stem rust on wheat plants sprayed with fungicides and inoculated at different time periods after fungicide application in experiment 2. The Tukey (P=0.05) value for comparison of means is 0.33. Pustule size (mm) CHAPTER 4 THE EFFECT OF STEM AND LEAF RUST ON YIELD OF BREAD WHEAT 4.1 ABSTRACT Rusts remain the most important diseases of wheat worldwide and localized epidemics have become more frequent in many regions. The release of rust- susceptible cultivars in South Africa is of great concern and therefore research was conducted to determine if chemical control is an economically viable method to control rust diseases. Fungicide field trials were conducted over two years at Greytown, South Africa to determine the effect of rust diseases on yield and test weight. Information on the efficacy of different foliar treatments in controlling these diseases was obtained. The wheat cultivar SST 88 was used in both experiments and four fungicides and three application methods were tested. Although the emphasis was on stem rust, it was not possible to eliminate leaf rust from the field trials. In 2005, 92% of the yield reduction experienced was contributed by leaf rust infection whereas in 2006, 32% of losses could be attributed to this disease. Stem rust on the other hand contributed 8% to losses in 2005 and 68% in 2006. Mean yield increase resulting from fungicide applications ranged from 36% to 45% over the two years. With a flag leaf treatment the mean yield increase per plant was 21% compared to 43% with a multiple-spray treatment in 2005. The highest yield increase was obtained with tebuconazole resulting in a 58% gain. Fungicide application increased test weight by 29% over the control treatment. No significant differences were found between application techniques. 4.2 INTRODUCTION Rusts remain important diseases of wheat (Triticum aestivum L.) worldwide because of their wide distribution, capacity to form new pathotypes that can attack previously resistant cultivars, ability to move long distances, potential to develop rapidly under optimal environmental conditions, and economic impact. Stem rust (Puccinia Effects of rust on yield 93 graminis Pers.:Pers. f. sp. tritici Eriks. & E. Henn.) can cause total losses in grain yield of wheat (Mayfield, 1985). South African regions historically prone to wheat stem rust, and where localised epidemics have been more frequent, are the Western Cape production areas. Stem rust epidemics in South Africa were last documented in 1985 when Sr24 virulence was found (Le Roux, 1985; Le Roux & Rijkenberg, 1987a,b). The increased frequency of the disease since 2000 has been associated with the cultivation of stem rust susceptible cultivars and the occurrence of pathotype 2SA88 (Boshoff, Pretorius, Van Niekerk & Komen, 2002). Due to the comparative rarity of stem rust, effects of fungicide control on yield reduction are poorly understood. Treatment decisions should ideally be supported by disease thresholds, either formalised or developed by crop managers through practical experience. In farm practice, the proportion of yield that is due to fungicide treatment is not usually known, and the success of spray decisions is often judged by the level of disease later in the season (Paveley et al., 1997). Paveley et al. (2001) quantified the reduction in foliar disease severity in wheat by fungicides. They showed that fungicides delayed and reduced epidemic onset and rate, and that the magnitude of the effect was dependent on both spray timing and dosage. Reliable methodology to estimate crop losses could enable the producer to weigh the cost of a control programme against the benefits of yield protection (Calpouzos et al., 1976). The potential threat of pathotype Ug99 (Pretorius et al., 2000) has renewed interest in stem rust impact and control. The objective of this chapter was to assess the damage potential of stem rust on a current spring wheat cultivar. 4.3 MATERIALS AND METHODS In 2005 and 2006 field trials were established at Redgates, Pannar Experimental Farm, Greytown, South Africa. The test cultivar was SST 88, a popular spring wheat cultivar grown in the Western Cape. Stem rust epidemics were created by injecting fresh urediniospores of pathotype UVPgt55 into the whorls of susceptible wheat plants before the booting stage and by applying spores in a water suspension onto rust spreader rows. These spreader rows were grown between experimental plots and in Effects of rust on yield 94 pathways perpendicular to plots. Supplementary overhead sprinkler irrigation was supplied as needed. 4.3.1. 2005 Experiment The cultivar SST 88 was planted in 1-m rows during June. The spacing between rows was 90 cm. The experiment was designed to determine yield on a single plant basis as described by Campbell & Madden (1990). Treatments included a water control, one fungicide spray with tebuconazole at the flag leaf stage and multiple sprays with tebuconazole to serve as an uninfected control. The fungicide was applied with a knapsack sprayer fitted with a cone nozzle and water volumes of approximately 300 ℓ ha-1. Disease assessments for leaf and stem rust were carried out on tagged plants on three occasions, namely at decimal growth stages 45, 53 and 71. The percentage disease severity on flag leaves and stems of 140 plants per treatment, using a modified Cobb scale (Peterson, Campbell & Hannah, 1948), was determined. Area under the disease progress curve (AUDPC) was calculated. At maturity, each tagged plant was harvested individually and yield and thousand kernel mass were determined. Thousand kernel mass was extrapolated from the mass of 100 kernels per plant. 4.3.2. 2006 Experiment The cultivar SST 88 was space-planted in a randomized, split-plot design with five treatments (four fungicides and one water control), three application techniques and four replications. Each plot consisted of two 2-m rows. Fungicides served as main plots and application technique as sub plots. The fungicides tebuconazole (187.5 g a.i. ha-1), trifloxystrobin (50 g a.i. ha-1), flusilazole/carbendazim (50/100 g a.i. ha-1) and azoxystrobin/difenoconazole (100/62.5 g a.i. ha-1) were evaluated. For unregistered fungicides, the rates applied were recommended by the respective chemical company. The fungicides were applied in single applications at growth stage 61, using a knapsack sprayer and water volumes of approximately 300 ℓ ha-1. Fungicides were applied either on the base of the stems, the flag leaf canopy only, and on both the stems and flag leaves. Disease assessments were carried out on whole plots at the time of application followed by two assessments at 14-day intervals. The data were used to calculate Effects of rust on yield 95 AUDPC for each treatment. At maturity the number of plants per row was determined and two rows per treatment were harvested. Seed were air-dried, cleaned and weighed to determine grain yield. Yield obtained from each plot was used to determine the thousand kernel mass. 4.3.3. Statistical analysis AUDPC values calculated for rust development over time, grain yield and thousand kernel mass data obtained in the different experiments in different years, were analyzed for variance using the statistical software NCSS 2000 for Windows. Means were compared using Tukey’s test (P=0.05) (Steel & Torrie, 1980). 4.4 RESULTS The occurrence of stem and leaf rust infections and the onset of epidemics differed from year to year. 4.4.1. 2005 Experiment Natural leaf rust infection reached epidemic proportions on SST 88 early in the season. The first leaf rust symptoms were observed in mid-August at the time of the first application of tebuconazole. No stem rust symptoms were observed at that stage. When the flag leaf treatments were applied, two weeks after the first spray, no leaf rust was observed on the multiple spray treatment plots whereas severity varied between 40 to 80% in the other treatments. Only traces of stem rust were observed at that time. After six weeks premature leaf necrosis was noted on the control treatment and 50% of the stems were rusted. The mean stem rust infection on the single flag leaf treatment was 20%, whereas no rust symptoms were detected in the multiple spray treatment. Leaf rust infection resulted in high AUDPC values as opposed to stem rust for which a lower AUDPC was calculated (Figure 1.) The single flag leaf application reduced the AUDPC for leaf and stem rust by 32% and 39%, respectively. In a stepwise regression analysis it was shown that leaf rust contributed 92% of the reduction in yield. Effects of rust on yield 96 Yield per plant was significantly higher following fungicide treatments which also differed statistically (P ≤ 0.05). Compared with the control, the mean yield increase per plant was 21% following the flag leaf treatment, and 43% with the multiple spray treatment (Figure 2). Test weight differed statistically (P ≤ 0.05) between treatments. The test weight obtained with the multiple spray treatment was significantly higher (P ≤ 0.05) than those obtained for the single flag leaf treatment and the control. Compared to the control, test weight for the flag leaf and multiple spray treatments increased by 18% and 33%, respectively (Figure 3). 4.4.2. 2006 Experiment Only trace stem rust symptoms were observed during the application of the fungicides at growth stage 59. Leaf rust symptoms appeared about two weeks later on the control plots. Four weeks after fungicide application, stem rust severity varied between 5 and 20% for sprayed plots, and between 40 and 60% for the unsprayed plots. The mean leaf rust severity was 10% for the treated plots and 70% for the untreated plots. The mean stem rust AUDPC for the fungicide treatments was 53% lower than the control whereas the mean leaf rust AUDPC was reduced by 68% (Figure 4). Tebuconazole reduced the AUDPC for stem rust and leaf rust by 83% and 89% respectively, whereas flusilazole/carbendazim reduced the incidence of the rusts by 37% and 61%. Differences occurred between the application treatments. Compared to the untreated control, fungicides applied to the stems reduced the total AUDPC by 42%. The flag leaf, and combined flag leaf and stem application, reduced AUDPC by 41% and 48% respectively (Figure 5). In contrast to 2005, the R2 values showed that stem rust caused 68% yield reduction. Over application positions, yield increased by 58% for tebuconazole followed by trifloxystrobin (40%), flusilazole/carbendazim (27%), and azoxystrobin/ difenoconazole (20%) (Figure 6). The yield did not differ statistically (P ≤ 0.05) between application positions. The test weight was affected by stem and leaf rust infection and increased by 29% in fungicide applications. Over application techniques, the test weight increased by 42% for tebuconazole followed by trifloxystrobin (25%), azoxystrobin/ Effects of rust on yield 97 difenoconazole (25%) and flusilazole/carbendazim (24%) (Figure 7). The arrangement of plots and spreader rows is shown in Figure 8. 4.5 DISCUSSION This study confirmed the potential damage of rusts on wheat in South Africa. Under ideal conditions and the presence of a virulent pathotype and susceptible cultivar, these diseases can cause significant economic losses. Depending on the time, number of fungicide applications and the susceptibility of the cultivar, a yield increase of 21 to 43% can be achieved. In a recent study, yield losses caused by stem rust were reduced by 50% with the application of fungicides at the correct time (Loughman, Jayasena & Majewski, 2005). Previously, Pretorius (1983) showed that losses due to stem rust ranged from 7% to 35%, depending on the cultivar. With regard to wheat leaf rust, Boshoff, Pretorius & Van Niekerk (2002) showed that a combined seven leaf and flag leaf spray increased mean yield by 56%. Averaged over fungicides, one spray at the flag leaf stage increased yield by 50%. A similar yield gain of 56% was reported for wheat leaf rust control by Eversmeyer, Browder & Young (1975). In the present study, the application of fungicides during the flag leaf growth stage resulted in a higher test weight in both trials. The effect of fungicide applications to quantify stem rust damage was not well illustrated in both experiments because of the occurrence of leaf rust. More than 92% of the losses in 2005 were due to leaf rust. Significant differences were found among fungicides in their ability to control stem and leaf rust. McRae & Platt (1987) showed that AUDPC was more reliable than other measures to determine effective rates of fungicides. Low AUDPC values calculated for leaf and stem rust infections were observed with the application of tebuconazole which transpired into higher yield and test weights. In field studies Loughman et al. (2005) found that tebuconazole was more effective than flutriafol. In another study it was shown that chlorothalonil (a non-systemic fungicide) has a reduced efficacy and low eradicant activity as opposed to propiconazole (a systemic fungicide) (Mayfield, 1985). Mayfield (1985) also found a clear relationship between grain yield and disease severity by showing that prevention of a 1% increase in rust severity avoided a 2% loss in grain yield. Effects of rust on yield 98 Different application methods did not differ statistically and more refined experiments are needed to study the effect of a canopy spray on rust infections at the stem base. Loughman et al. (2005) found that when a fungicide is sprayed on slightly infected plants there is a subsequent reduction in stem rust development, while on heavily infected plants poor control was achieved. Application of fungicides at late booting was profitable when stem rust was present at that stage. The results of this study showed that stem and leaf rust could severely reduce the yield of susceptible cultivars in South Africa. Experimenting with fungicide rates and timing on different cultivars over several years and locations are necessary for developing accurate recommendations (Christ & Frank, 1989). More detailed research is needed to provide this information for stem rust control in South Africa. 4.6 LITERATURE CITED Boshoff, W.H.P., Pretorius, Z.A. & Van Niekerk, B.D. 2002. The impact of leaf rust on spring wheat in the winter rainfall region of South Africa. South African Journal for Plant & Soil 19, 84-88. Boshoff, W.H.P., Pretorius, Z.A., Van Niekerk, B.D. & Komen, J.S. 2002. First report of virulence in Puccinia graminis f. sp. tritici to wheat stem rust resistance genes Sr8b and Sr38 in South Africa. Plant Disease 86, 922. Calpouzos, L., Roelfs, A.P., Madson, M.E., Martin, F.B., Welsh, J.R. & Wilcoxson, R.D., 1976. A new model to measure yield losses caused by stem rust in spring wheat. Technical Bulletin 307, Agricultural Experimental Station University of Minnesota. Campbell, C.L. & Madden, L.V. 1990. Introduction to plant disease epidemiology. John Wiley & Sons, New York, 532pp. Christ, B.J. & Frank, J.A., 1989. Influence of foliar fungicides and seed treatments on powdery mildew, Septoria and leaf rust epidemics on winter wheat. Plant Disease 73, 148-150. Eversmeyer, M.G., Browder, L.E. & Young Jr, C., 1975. Effect of leaf and stem rust on 1974 Kansas wheat yields. Plant Disease Reporter 59, 604-607. Effects of rust on yield 99 Le Roux, J. 1985. First report of a Puccinia graminis f. sp. tritici race with virulence for Sr24 in South Africa. Plant Disease 69, 1007. Le Roux, J. & Rijkenberg, F.H.J. 1987a. Pathotypes of Puccinia graminis f. sp. tritici with increased virulence for Sr24. Plant Disease 71, 1115-1119. Le Roux, J. & Rijkenberg, F.H.J. 1987b. Occurrence and pathogenicity of Puccinia graminis f. sp. tritici in South Africa during the period 1981-1985. Phytophylactica 19, 456-472. Loughman, R. Jayasena, K. & Majewski, J., 2005. Yield loss and fungicide control of stem rust of wheat. Australian Journal of Agricultural Research 56, 91-96. Mayfield, A.H., 1985. Efficacies of fungicides for the control of stem rust of wheat. Australian Journal of Experimental Agriculture 25, 440-443. McRae, K.B. & Platt, H.W., 1987. An index for cultivar resistance based on disease progress curves. Phytopathology 77, 1181-1186. Paveley, N.D., Bradley, R.S., Scott, R.K., Craigon, J. & Day, W., 2001. Steps in predicting the relationship of yield on fungicide dose. Phytopathology 91, 708-716. Paveley, N.D., Lockley, K.D., Bradley, R.S. & Thomas, J., 1997. Determinants of fungicide spray decisions for wheat. Pesticide Science 49, 379-388. Peterson, R.F., Campbell, A.B. & Hannah, A.E. 1948. Diagrammatic scale for estimating rust intensity on leaves and stems of cereals. Canadian Journal of Research 26, 496-500. Pretorius, Z.A. 1983. Disease progress and yield response in spring wheat cultivars and lines infected with Puccinia graminis f. sp. tritici. Phytophylactica 15, 35-45. Pretorius, Z.A., Singh, R.P., Wagoire, W.W. & Payne, T.S. 2000. Detection of virulence to wheat stem rust resistance gene Sr31 in Puccinia graminis f. sp. tritici in Uganda. Plant Disease 84, 203. Steel, R.G.D. & Torrie, J.H. 1980. Principles and procedures of statistics, a biometrical approach. McGraw-Hill Book Company, New York, 2de edition, 633pp. Effects of rust on yield 100 3000 AUDPC SR 2500 AUDPC LR 2000 1500 1000 500 0 Control 1 x Spray Multiple Sprays Treatments Figure 1. The mean area under the disease progress curve (AUDPC) calculated for each treatment in the 2005 experiment. The Tukey (P=0.05) values for comparing stem rust and leaf rust means are 36.08 and 40.03, respectively. AUDPC Effects of rust on yield 101 16 14 12 10 8 6 4 2 0 Control 1 X Spray Multiple Sprays Treatments Figure 2. Mean yield for each treatment obtained over fungicides applied in 2005. Error bars indicate the positive standard deviation. The Tukey (P=0.05) value for comparing spray application means is 0.64. Mean yield per plant (g) Effects of rust on yield 102 60 50 40 30 20 10 0 Control 1 x Spray Multiple Sprays Treatments Figure 3. Mean test weight obtained over fungicides for each treatment in 2005. Error bars indicate standard deviation. The Tukey (P=0.05) value for comparing spray application means is 1.13. Test weight (g) Effects of rust on yield 103 1400 AUDPC LR 1200 AUDPC SR 1000 800 600 400 200 0 tro l ole le n on az z o zim i n na da tro b C co co en ys fen o bu arb x i Te e/c ri flo in/ d ol T rob l azi yst lu s zo x F A Fungicides Figure 4. Mean area under disease progress curve (AUDPC) calculated for each fungicide applied over treatments in 2006. Error bars indicate positive standard deviations. The Tukey (P=0.05) values for comparing stem rust and leaf rust means are 178.27 and 99.71, respectively. AUDPC Effects of rust on yield 104 600 500 400 300 200 100 0 Flag leaf & Stem Flag leaf Stem Application position AUDPC LR AUDPC SR Figure 5. Mean area under disease progress curve (AUDPC) calculated for each application position over fungicides applied in 2006. AUDPC values for untreated stem and leaf rust plots were 868 and 661. Error bars indicate positive standard deviations. The Tukey (P=0.05) values for comparing stem rust and leaf rust means are 138.95 and 174.6, respectively. AUDPC Effects of rust on yield 105 1000 800 600 400 200 0 l e e tro ol ol im bin on az z zC on on a da tro c c n s no bu arb e y e e c iflo x if T e/ r bin /d l T o ila zo stry Flu s x Az o Fungicides Figure 6. Mean yield obtained for the control and each fungicide treatment, averaged over application methods, in 2006. Error bars indicate positive standard deviations. The Tukey (P=0.05) value for comparing treatment means is 103.28. Yield per plot (g) Effects of rust on yield 106 40 35 30 25 20 15 10 rolt l e le n on az o azo az im i n n d tro b C co co en s no bu arb lox y fe e /c if /di T le Tr bin laz o o i yst r lus zo x F A Fungicides Figure 7. Mean test weight obtained for the control and each fungicide treatment, averaged over application technique, in 2006. Error bars indicate positive standard deviations. The Tukey (P=0.05) value for comparing treatment means is 5.41. Test weight (g) Effects of rust on yield 107 Figure 8. Wheat stem rust field trial at the Redgates Experimental farm, Greytown, KwaZulu-Natal in 2006. Rust spreader rows are indicated by solid arrows and fungicide-treated plots by broken arrows. SUMMARY Stem rust caused by Puccinia graminis Pers.:Pers. f. sp. tritici Eriks. & E. Henn., is an economically important disease of bread wheat (Triticum aestivum L.) worldwide. The disease has often reached epidemic proportions in South Africa. The release of rust-susceptible cultivars in South Africa is of great concern and therefore research is needed to determine if chemical control is an economically viable method to control rust diseases. The recent increase of stem rust in the Western Cape production area of South Africa has led to a renewed interest in the use of fungicides. In addition to chemical control, effective genetic control of rust diseases requires a coordinated effort, including pathotype monitoring, collection and characterization of sources of resistance, and resistance breeding. Data generated by pathotype surveys thus form an essential component of breeding programmes and are conducted by most wheat producing countries, principally to recognize pathogenic changes. The timely detection of stem rust pathotypes with new virulence is considered important to the South African wheat industry. A stem rust survey is conducted annually to monitor pathogenic variability in Puccinia graminis f. sp. tritici in the major bread wheat and triticale producing areas of South Africa. More than 70% of the isolates obtained from 2002 to 2004 were characterized as pathotype 2SA88, which unusual for local pathotypes, was virulent for Sr8b and Sr38. During the six year survey period, stem rust pathotypes 2SA4, 2SA55, 2SA88, 2SA99 and 2SA100 were detected on bread wheat and 2SA102, 2SA102K and 2SA103 on triticale. To optimize in vitro testing conditions the effect of temperature on germination and germ tube growth of P. graminis f. sp. tritici as well as incubation periods were evaluated. Seventy eight percent of urediniospores had germinated on water agar in petri dishes after 6 h of incubation at 20 and 25°C. Low germination rates were observed at 30°C. Twenty-nine fungicides in nine chemical classes were evaluated in vitro for toxicity to stem rust urediniospores. Water agar petri dish plates were amended with fungicides at concentrations ranging from 10 to 0.0001 μl active ingredient / ml medium. Low EC50 (effective concentration that results in 50% inhibition) values were obtained with azoxystrobin, trifloxystrobin, kresoxim-methyl, mancozeb, azoxystrobin/ difenoconazole, iprodione, chlorothalonil and hexoconazole. Summary 109 A shorter residual effect of trifloxystrobin occurred in wheat infected with stem rust in a glasshouse assay, whereas longer efficacy periods were determined for azoxystrobin/difenoconazole and tebuconazole. Fungicide field trials were conducted at Greytown, South Africa to determine the effect of rust diseases on yield and test weight. Information on the efficacy of different foliar treatments in controlling these diseases was obtained. Although the emphasis was on stem rust, it was not possible to eliminate leaf rust from the field trials. In 2005, 92% of the yield reduction experienced was contributed by leaf rust infection whereas in 2006 32% of losses could be attributed to this disease. Stem rust on the other hand contributed 8% to losses in 2005 and 68% in 2006. Yield increase resulting from fungicide applications ranged from 36% to 45% over the two years. With a flag leaf treatment the mean yield increase per plant was 21% compared to 43% with a multiple-spray treatment in 2005. The highest yield increase was obtained with tebuconazole resulting in a 58% gain. Test weight increased by 29% with fungicide applications over the control plots. Further research is needed to optimize control by means of fungicides. OPSOMMING Stamroes, veroorsaak deur Puccinia graminis Pers.:Pers. f. sp. tritici Eriks. & E. Henn., is ‘n ekonomies belangrike siekte op broodkoring (Triticum aestivum L.) wêreldwyd. Die siekte het gereeld epidemiese afmetings in Suid-Afrika aangeneem. Die vrystelling van vatbare kultivars is kommerwekkend en navorsing is nodig om te bepaal of chemiese beheer ‘n ekonomies-regverdigbare uitweg is om roessiektes te beheer. Die toename in stamroes in koringproduksiegebiede van die Wes-Kaap, Suid Afrika het gelei tot die algemene gebruik van swamdoders. Bo en behalwe chemiese beheer benodig effekiewe genetiese beheer van roessiektes ‘n gekoördineerde poging insluitende patogeniese monitering, verkryging en karakterisering van weerstandsbronne en weerstandsteling. Data verkry vanaf patogeniese waarneming vorm ‘n essensiële komponent van ‘n teelprogram en word uitgevoer deur die meeste koringproduserende lande. Om stamroespatotipes betyds te identifiseer, veral ten opsigte van nuwe virulensie, is belangrik vir die Suid- Afrikaanse koringbedryf. Stamroesopnames word jaarliks uitgevoer om die patogeniese variasie van P. graminis f. sp. tritici in die belangrike koring en triticale produserende areas van Suid Afrika te moniteer. Meer as 70% van die isolate verkry gedurende 2002-2004 is gekarakteriseer as patotipe 2SA88, wat, ongewoon vir plaaslike patotipes, virulent is vir Sr8b en Sr38. Gedurende ‘n sesjaar periode is stamroespatotipes 2SA4, 2SA55, 2SA88, 2SA99 en 2SA100 geïdentifiseer op broodkoring terwyl patotipes 2SA102, 2SA102K en 2SA103 geïdentifiseer is op triticale. Om in vitro toetse te optimiseer is die effek van temperatuur op ontkieming en kiembuislengte van P. graminis f. sp. tritici, sowel as inkubasieperiode geëvalueer. Agt en sewentig persent van urediniospore het ontkiem op wateragar petribakkies na 6 h van inkubasie by 20 en 25°C. Lae ontkiemingswaardes is verkry by 30°C. Nege- en-twintig swamdoders in nege chemiese groepe is in vitro getoets vir toksisiteit teenoor stamroesurediniospore. Swamdoders teen konsentrasies van 10 tot 0.0001 μl aktiewe bestanddeel per ml medium is by petribakkies met wateragar gevoeg. Lae EK50 waardes (effektiewe konsentrasie wat lei tot 50% inhibisie) is verkry met azoxystrobin, trifloxystrobin, kresoxim-methyl, mancozeb, azoxystrobin/ difenoconazole, iprodione, chlorothalonil en hexoconazole. ‘n Korter nawerking is Opsomming 111 verkry met trifloxystrobin op koring geïnfekteer met stamroes in glashuistoetse, terwyl azoxystrobin/difenoconazole en tebuconazole langer nawerkingsperiodes getoon het. Swamdoderveldproewe is uitgevoer by Greytown, Suid Afrika om die effek van roessiektes op opbrengs en 1000-korrel massa te bepaal. Inligting oor die effektiwiteit van verskillende blaarbespuitings in die beheer van die siektes is verkry. Al was die primere fokus gerig op stamroes, kon daar nie verhoed word dat blaarroes die proewe infekteer nie. In 2005 is 92%, en in 2006, 32% van die opbrengsverliese veroorsaak deur blaarroesinfeksies. Stamroes het bygedra tot 8% van verliese in 2005 en 68% in 2006. Opbrengsverhoging verkry deur swamdodertoediening het gewissel van 36% tot 45% oor die tweejaarperiode. In 2005 is die gemiddelde opbrengs per plant verhoog met 21% deur ‘n vlagblaarbespuiting, vergeleke met 43% met herhaalde bespuitings. Die hoogste opbrengs verhoging is verkry met tebuconazole (58%). Duisendkorrelmassa het verhoog met 29% vergeleke met die kontrole. Verdere navorsing is nodig om chemiese beheer te optimaliseer.