Soybean rust (SBR) caused by Phakopsora pachyrhizi Syd. is an aggressive wind dispersed fungal disease which has spread around the world at an alarming rate in the last decade. The disease was first reported in South Africa (SA) in 2001. It has become well established in the province of KwaZulu-Natal. Reports are occasionally made from eastern Mpumalanga, late in the growing season, in years with good rainfall. Yield losses of 10 - 80% have been reported due to SBR infection. Literature was reviewed to better understand the pathogen in an attempt to find suitable disease management strategies. Many strategies involve delaying, rather than preventing, SBR infection. Of the two strategies to prevent infection, the use of fungicides was the only option for disease control in SA, as no resistant cultivars are available. Field trials were conducted to determine which fungicides are effective in controlling SBR. Further research was conducted to determine the timing, frequency and rate of fungicide applications for optimal control of SBR. Trials were evaluated for disease severity, seed yield and the effect of fungicides on seed quality. Fungicides from the triazole class of the sterol biosynthesis inhibiting group of fungicides were found to be the most effective in controlling SBR. A fungicide from the strobilurin group was found to be less effective than the triazoles at the suggested rate, but was found to be as effective when evaluated at higher dosage rates. Triazoles premixed with fungicides from the benzimidazole and strobilurin groups were also effective in controlling SBR. Timing of application was found to be critical for strobilurin fungicides, but not for triazole fungicides, which have a curative ability, unlike strobilurins. Strobilurin fungicides applied preventatively, before the appearance of disease symptoms were as effective as triazole fungicides applied after disease symptoms, but before infection levels had reached 10%. Across both wet and dry seasons two fungicide applications applied at 21d intervals at the R2 growth stage resulted in effective disease control. In wet seasons, a third fungicide application resulted in yields that were higher, albeit not statistically significant, than two fungicide applications. Assessments of individual fungicides for optimal dosage rate found that registered rates were already optimal for some fungicides, but for others it appeared as if alterations were necessary to the rate suggested for registration. This study was one of the first to extensively evaluate the efficacy of the new triazole and strobilurin fungicides on SBR control. The results have been shared globally, but particularly with newly affected countries in South and North America. Although this research has been groundbreaking, there are still many aspects of fungicide control which need to be studied in order to further optimise chemical control of SBR. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2005

Phakopsora pachyrhizi H. Syd and P. Syd, the causal organism of soybean rust (SBR) was first reported in Japan in 1902. In 1934 the pathogen was found in several other Asian countries and as far south as Australia. In India, SBR was first reported on soybeans in 1951. There have been several early reports of SBR in equatorial Africa but the first confirmed report of P. pachyrhizi on the African continent was in 1996 from Kenya, Rwanda and Uganda. Since then, the pathogen has spread south with reports from Zambia and Zimbabwe in 1998 and in Mozambique in 2000. In February 2001, P. pachyrhizi was first detected on soybeans near Vryheid, in Northern KwaZulu-Natal, South Africa (SA). As the season progressed, the disease was observed in other parts of the province, and epidemic levels were found in the Cedara, Greytown, Howick and Karkloof production regions. Soybean rust subsequently spread to Amsterdam and Ermelo in the Highveld region of SA. The disease reappeared in SA in March 2002. It is now established that the pathogen is a threat to soybean production in the country with yield losses in the region of 10-80%. A literature review on SBR investigating the taxonomy of the pathogen, its morphology, symptoms, host range, infection process, epidemiology, control options and the economic importance of P. pachyrhizi was complied to provide the necessary background information to conduct research under local conditions and to assist in interpretation of results of experiments. Epidemiological trials were conducted at the University of KwaZulu-Natal under controlled environmental conditions in a dew chamber and conviron. Development of P. pachyrhizi on the susceptible cultivar (LS5995) was quantified in combinations of seven temperatures (15,19,21,24,26,28 and 30°C) and five leaf wetness durations (LWD) (6,9,12,14 and 16hrs) at three relative humidities (RH) (75%, 85% and 95%). Studies indicate that optimum temperature for uredospore infection is 21-24°C with a LWD greater than 12hrs and RH 85-95%. The number of pustules as well as lesion size on the abaxial and adaxial leaf surface increased with increasing LWD at all the RH values tested. Infection did not occur on plants incubated at 15°C and 30°C at 85% or 95%RH whereas at 75%RH infection did not occur on plants incubated at 15°C, 19°C and 30°C regardless of LWD. Number of pustules per lesion produced at 75%, 85% and 95%RH was highest at 24°C and showed a gradual increase with increasing LWD. Lesion size on both leaf surfaces increased after 12hrs LWD at 24°C at 75% and 85%RH whereas at 95%RH lesion size increased after 14hrs LWD at 24°C. Exposure of uredospores to ultraviolet light which is equivalent to ultraviolet C (sunlight) which is < 280nm, shows a decrease in germination (7%). Under continuous darkness, the germination percentage was found to range from 58% after 48 hrs. Germination was found to peak at 16hrs in darkness with a gradual decrease as time increased whereas germination under ultraviolet light was highest after 6hrs with a gradual decrease with increased exposure to light. Germ tube lengths were found to be shorter when exposed to ultraviolet light (107µm) compared to controls kept in the dark (181µm). Results obtained clearly show a negative effect of ultraviolet light on the germination and germ tube length of uredospores. A 0.1 ml suspension of uredospores on 1.25% water agar Petri dishes was exposed to cycles of 14h ultraviolet light and 10h darkness for 48h. Results indicate an increase in germination percentage of uredospores when exposed to 10h of darkness following a 14h period under ultraviolet light. Controlled environmental studies were conducted to determine alternative hosts of P. pachyrhizi in SA. The control used in this experiment was Prima 2000, a susceptible cultivar to soybean rust. Seven legume plants [Cajanus cajan (L.) Huth, Glycine max (L.) Merr, Lablab purpureus (L.) Sweet, Lupinus angustifolius (L.) Finnish, Phaseolus vulgaris (L.), Pueraria lobata (M&S) Wild and Vigna unguiculata (L.) Walp] and three dry bean lines (Bonus; OPS-RS2 and PAN 159) showed typical SBR symptoms when rated after 21 days post inoculation with uredospores for percentage disease severity. Disease severity was significantly different within the alternative hosts, but G. max, P. vulgaris and P. lobata were not significantly different from Prima 2000 (control). A uredospore suspension of 2.5 x 10(5) uredospores ml(-1) from plants that showed typical SBR symptoms was made and inoculated on to Prima 2000, a susceptible soybean cultivar. Uredospores from pustules on G. max, L. purpureus, L. angustifolius, P. vulgaris, P. lobata, V. unguiculata, Bonus and PAN 159 produced viable uredospores on PRIMA 2000. These plants are considered alternative hosts of P. pachyrhizi. Effect of leaf age on susceptibility of soybean to SBR was tested under controlled environmental conditions. Mean number of lesions as well as lesion size were greater on younger leaves than on older leaves of plants at the same physiological age. Plants at the early vegetative and reproductive stages had a significantly lower number of lesions as well as a smaller lesion size. Plants at the V6 and R1 growth stages were significantly more susceptible to P. pachyrhizi than plants at other developmental stages. Trichoderma harzianum Rifai, Eco-77® a commercial biological control product, was evaluated for its efficacy as a biological control agent of P. pachyrhizi. Trichoderma harzianum sprayed at the standard concentration on infected soybean plants was significantly more effective in controlling P. pachyrhizi than plants sprayed at 1/2X and 2x the standard concentration. This was noted in both Trial 1 and 2. Data indicate that spraying the filtrate two days after inoculation produces less disease. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2006.

Soybean rust (SBR), caused by the fungus Phakopsora pachyrhizi Syd., is a real threat to soybean crops in South Africa. Its ability to spread rapidly and its potential to severely reduce yields have earned it the reputation as the most destructive foliar disease of soybeans. SBR has been reported in South Africa every year since its arrival in 2001. While extensive research had been done on the epidemiology and fungicide application requirements in South Africa, no work into the long term climatic vulnerability of soybean production areas to SBR had been done. This meant soybean producers do not know whether SBR is a threat in their areas. Through this research a SBR algorithm was developed using readily available climate data, viz. temperature and rainfall, to create a daily index specifying the climatic vulnerability of SBR infection. The algorithm was applied to a 50 year historical climate database, and a series of maps was created illustrating the long term vulnerability of different areas to SBR infection. These maps allow soybean producers to understand the climatic vulnerability of their area to SBR infection. Time series graphs were created for selected key soybean production areas to allow soybean producers to distinguish periods of high and low climatic risk during the season. This may help with decisions regarding the planting times, the maturation rate of different cultivars as well as the timing and application of fungicides. The framework for a near real time forecasting system was created outlining how the system could amalgamate recently recorded and forecasted weather data, run it through the SBR algorithm and provide a near real time, as well as forecasted vulnerability, based on the climatic conductivity for SBR infection. Anticipated limitations and difficulties on developing the forecasting system are also outlined. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2009.