RESEARCH PROJECTS  //  Research Report 2018/2019

Continuation Research Projects

  1. Evaluation of commercially available sunflower cultivars

    Dr SH Ma'ali, MW Makgoga, JL Erasmus, MU Molale, SB Mahlatsi, PR Mogotlwane, SC Swanepoel and ZL Xhasa

    Data collected

    Twenty-four cultivars, of which six were new introductions, were successfully evaluated for seed yield, seed quality (oil and protein content) and some agronomical characteristics at 14 localities/planting dates during the 2018/19 growing season. Twelve of the 24 cultivars were Clearfield types on which the use of the post emergence broad leaf weed controlling herbicide mixture, imazapyr + imazamox (Euro-Lightning®), is possible. In the field trials these cultivars were treated in the same way as the regular cultivars and received no Euro-Lightning®. Two hybrids (PAN 7158 HO and SY3975 CLHO) of the 24 were high oleic acid. Yield probabilities were calculated for 18 cultivars that were evaluated in 27 trials during 2017/18 and 2018/19 and for 12 cultivars that were evaluated in 35 trials during 2016/17 until 2018/19 growing season.

    Value of project

    Dlamini et al. (2015) conducted a study to evaluate the economic impacts of national cultivar trials (A case for sorghum, sunflower, soybeans and dry beans) in South Africa over the period 1978-2012. They reported that the yield benefits are equivalent to 13.10kg for sunflower output per hectare per year, at the assumed plausible yield gain estimate attributable to the trials of 5 percent. In present value terms, the estimated total economic benefits that have accrued to South Africa over the period 1978-2012 amounted to R200 million in 2012 prices, which is equivalent to 4 percent of the total gross value of production for these crops in 2012. Of these benefits, about R23.2 million came from the evaluation of sunflower cultivars. The South African economy received more than R 40 of benefits for each R1 invested by the ARC-GC and OPSDT in the trials, with beneficiaries including consumers and food or feed processors (Dlamini 2014). Up to 30% of cultivars are replaced annually. If a new cultivar with an extreme seed yield or quality is introduced, it will be identified through this project.

    Previous results from cultivar evaluations showed that the mean yield of the five best cultivars is 0.25 t ha-1 higher than the overall mean yield of all cultivars. This relates to R1 125 ha-1 at a seed price of R4 500 t1. If only 10% of the national crop of 635 000ha could come from the five best cultivars as opposed to a random selection of cultivars, the increased yield will earn R71 437 500 per annum. Moreover, sunflower cultivars that combine genetics for high yield and quality are generally preferred by processors, sunflower oil concentration was reported to be a conservative genetic component (Ruiz and Maddonni 2006). However, recent studies highlighted differential responses of sunflower genotypes in different cropping conditions such as greater variability of oil concentration was whether linked to management and environmental conditions or to genotypic and environment interactions (Andrianasolo et al. 2012). Environmental factors, such as temperature, day length, intercepted solar radiation and precipitation, affect the growth of sunflowers differently (Van der Merwe et al. 2013; Pekcan et al. 2015; Robert et al. 2016). A variety of research pointed out that fluctuations in temperature and moisture availability affects the quantity and quality of oil accumulation in sunflower (Aguirrezábal et al. 2003; Hassan et al. 2011). Therefore, evaluating sunflower cultivars under different environments with different planting dates will give better indication about their performance and the best cultivar will be identified thorough this project. Information can be utilised by the producer and the industry to improve the sunflower seed productivity, quality and increase the profitability of sunflower.

  2. Oilseeds South African Soybean Crop Quality Survey

    Ms W Louw

    Soybeans are the most important oilseed crop produced in South Africa, driven mainly by the demand for protein feed (in this case soybean oilcake) from the animal feed industry.

    Of the 1,31 million tons of soybeans processed from March 2018 to February 2019, 1,9% was used for human consumption, 16,7% for animal feed as full fat soya and the bulk (1,06 million tons) was crushed to produce oil and oilcake. The quantity of soybeans crushed so far this season, is 19% more than the total quantity crushed during the previous season and double the ten-year average. Figures were obtained from the South African Grain Information Service (SAGIS).

    A number of factors affect soybean protein and oil content. Unfortunately, some of these factors are outside of a producer's control, such as the weather e.g. temperature, drought conditions as well as the timing of the drought conditions (early season or late season) and soil characteristics. Other factors include production practices and cultivar choice. Generally speaking, any factor affecting the protein content will inversely affect the oil content and vice versa.

    Crude protein content

    In the 2017/2018 production season, the weighted average crude protein content was 37,40% ('as is' basis), slightly higher than the 37,20% of the previous season. As in the 2016/2017 season, Limpopo had the highest weighted average crude protein content (38,38%). Gauteng (37,08%) and the Free State (37,18%) reported the lowest averages. See Graph 1 for a comparison of the average protein content per province per season ('as is' basis).

    Graph 1: Protein content per province per season ("as is" basis)
    Graph 1: Protein content per province per season ('as is' basis)
    Graph 2: Northern Cape oil content over seasons ("as is" basis)
    Graph 2: Northern Cape oil content over seasons ('as is' basis)

    Crude fat/oil percentage

    The weighted average crude fat/oil percentage of 18% ('as is' basis) was almost half a percentage point lower than the 18,4% in the previous season, but similar to the average values reported in 2014/2015 and 2015/2016. The samples from Limpopo had the highest weighted average crude oil content, namely 19,8%. The lowest average oil content was observed in the Free State with 17,6%.

    See Graph 2 to Graph 8 for a comparison of the average oil content per province compared to the national average over seasons. The minimum and maximum values are also indicated.

    150 samples representing the various production regions were taken and forwarded to the Southern African Grain Laboratory (SAGL) by the commercial grain silo owners during the third quarter of 2018. Full grading was done in accordance with the Regulations relating to the Grading, Packing and Marking of Soybeans intended for sale in South Africa (Government Notice No. R.370 of 21 April 2017). 87% (130) of the samples were graded as Grade SB1, while 13% (20) of the samples were downgraded to Class Other Soya Beans (COSB). During the previous two seasons, 12% (2016/2017) and 11% (2015/2016) of the samples were downgraded to COSB.

    The majority of samples downgraded this season was as a result of the percentage other grain present in the sample exceeding the maximum permissible deviation or as a result of the presence of poisonous seeds, namely Datura spp. or Convolvulus spp.

    Graph 3: North West oil content over seasons ("as is" basis)
    Graph 3: North West oil content over seasons ('as is' basis)
    Graph 4: Free State oil content over seasons ("as is" basis)
    Graph 4: Free State oil content over seasons ('as is' basis)
    Graph 5: Mpumalanga oil content over seasons ("as is" basis)
    Graph 5: Mpumalanga oil content over seasons ('as is' basis)

    Other deviations resulting in the downgrading of samples were the percentage foreign matter or soiled soybeans present or a combination of foreign matter, other grain, soiled soybeans, collective deviations as well as the presence of an undesirable odour. None of the samples graded contained wet pods.


    Samples containing sclerotia from the fungus Sclerotinia sclerotiorum decreased by 16% to the previous season (88 samples versus 105) The three highest percentages sclerotia observed (0,36%, 0,32% and 0,30%) were on samples from the Free State. However, these percentages are still well below the maximum permissible level of 4%.

    The national weighted average percentage this season was 0,06% compared to the 0,07% of the previous season.

    Foreign matter

    Gauteng (eleven samples) had the highest weighted average percentage foreign matter (1,43%). The percentage foreign matter in the rest of the samples ranged from 0,58% in Limpopo (four samples) to 1,38% in the Free State (45 samples).

    Other grading factors and test weight

    Gauteng also reported the highest weighted average percentage soybeans and parts of soybeans above the 1,8mm slotted sieve which pass through the 4,75mm round hole sieve, namely 2,02%. The samples from KwaZulu-Natal (nine samples) and the Northern Cape (two samples) were the lowest with 0,60% and 0,62% respectively. Mpumalanga (71 samples) averaged 1,57% and the Free State 1,74%.The national weighted average percentage increased from 0,88% the previous season to 1,54% this season.

    The lowest weighted average percentage defective soybeans on the 4,75 mm sieve was observed on the samples from Gauteng, namely 0,98%. The Northern Cape reported the highest percentage namely 2,85%, followed by Limpopo with 2,74%. The national weighted average decreased from 2,22% last season to 1,91% this season.

    Soybean leaves (soybean crop quality survey)

    This season the national weighted average percentage soiled soybeans were 1,53%, compared to the 2,87% of the previous season. Average weighted percentages per province ranged from 0,30% in the Northern Cape to 3,86% in North West Province (eight samples). Three samples exceeded the maximum permissible deviation of 10% according to the grading regulations. All three samples originated from Mpumalanga. Last season, eleven samples exceeded this grading limit.

    Test weight values were determined by following the standard working procedure of the Kern 222 instrument as described in ISO 7971-3:2009, dividing the g/1 litre filling mass by two and extrapolating the values by means of the formulas obtained from the Test Weight Conversion Chart for Soybean of the Canadian Grain Commission. The weighted average this season was 70,9kg/hl. Individual values ranged from 67,2kg/hl to 74kg/hl. The previous two seasons averaged the same as this season and ranged between 65,8kg/hl to 73,6kg/hl and 64,9kg/hl and 73kg/hl respectively.

    Graph 6: Gauteng oil content over seasons ("as is" basis)
    Graph 6: Gauteng oil content over seasons ('as is' basis)
    Graph 7: Limpopo oil content over seasons ("as is" basis)
    Graph 7: Limpopo oil content over seasons ('as is' basis)

    Genetically modified soybeans

    The EnviroLogix QuickComb kit for bulk soybeans was used to quantitatively determine the presence of genetically modified soybeans in 15 of the crop samples. The kit is designed to extract and detect the presence of certain proteins at the levels typically expressed in genetically modified bulk soybeans. The measurements made by using the above-mentioned kit is sensitive enough to detect one Roundup Ready soybean in 400 conventional soybeans (0,25%). All of the samples tested positive for the presence of the CP4 EPSPS (RR1/RR2) protein.

    Graph 8: Kwazulu-Natal oil content over seasons ("as is" basis)
    Graph 7: Kwazulu-Natal oil content over seasons ('as is' basis)
  3. The funding of the Supply and Demand Estimates Committee

    Dr A Balarane

    Purpose of the South African Supply & Demand Estimates (SASDE) Report

    The report provides an analysis of the fundamental conditions of the major grains and oilseeds in South Africa. The report is normally released within four to five working days after the Crop Estimates Committee (CEC) meeting. The report is released into the public domain by the approval of the South African competition authorities.

    Composition of the S&DEC

    The S&DEC is a technical committee that communicates directly with industry role players through the Supply and Demand Estimates Liaison Committee (S&DELC). The S&DEC consists of a chairperson with two independent specialists appointed by the industry; the NAMC acts as the secretariat with four staff members, the South African Grain Information Service (SAGIS) and the secretariat of the CEC from the Department of Agriculture, Forestry and Fisheries (DAFF).

    Methodology used by the S&DEC

    Process before the meeting

    • The S&DEC meeting takes place three to five working days after the CEC meeting, at the end of each month with the exclusion of December.
    • A reminder is sent to all co-workers approximately seven days before the S&DEC meeting, requesting that they submit estimates on imports, exports and consumption for selected grains and oilseeds.
    • The S&DEC meeting takes place in a lockdown room with no cell phones allowed, except when there is uncertainty about any of the figures sent by co-workers.

    Factors taken into account during the meeting

    • Supply figures
    • CEC published figures
    • S&D historical figures
    • Demand figures
    • Historical figures (SAGIS)
    • Current deliveries published by SAGIS (extrapolations, regression modelling)
    • Co-workers' figures, without disclosing their identity
    • SAGIS figures
    • Figures published by the CEC
    • Figures and opinions of, and discussions with, co-workers
    • Baseline modelling figures on demand and consumption from the Bureau for Food and Agricultural Policy (BFAP)
    • Other modelling, including industry forecasting modelling and computable general equilibrium (CGE modelling), is also considered for use in the future, taking into consideration the expansion in capacity

    Distribution and users of the SASDE Report

    The NAMC retrieves the number of website hits on a monthly basis. The purpose of this exercise is to track the number of end-users of the SASDE report. It is noted from the NAMC monthly website hits that about 1000 hits are received for the SASDE report. The report is mainly utilised by the following:

    • Academia
    • Financial and investment institutions
    • Government officials
    • Grain Millers
    • Oilseed processors
    • Feed manufactures
    • Grain and oilseed traders
    • Grain and oilseed storage handlers
    • Baking industry
    • Research institutions and transport organisations
    • Others

    Expenditure for 2017/18 financial years

    The functions of the S&DEC rely entirely on the following stakeholders: NAMC, Maize Trust, Winter Cereals Trust, and Oilseeds Trust. The arrangement within these organisations is that about ±50-55% of the budget is contributed by the NAMC, while the other ±45-50% is contributed by the industry trust. The NAMC drafts a budget and request financial assistance from the Trust in the beginning of the new financial year and invoice the trusts after approval of actual for the previous year. The funds received are therefore utilised in accordance with the functions of the S&DEC.


    The efficacy of information plays a significant role in developing nations. Relevant and accurate information can ensure sustainability of the market and also that a nation's food security is well considered in accordance with private and government objectives. It is initiatives such as the SASDE report that could provide effective information to the market for a specific period. The publication of the SASDE report has taken the grain and oilseeds market in South Africa and the region to another level, as evident from the statistics on the end users of the report. The mutual understanding and collaboration of the NAMC and the trusts has also contributed significantly to the functioning of the S&DEC. Such support is what the committee requires in ensuring that its functions are fulfilled and continuously improved.

  4. Oilseeds South African Sunflower Crop Quality Survey

    Ms W Louw

    In the sixth annual sunflower crop quality survey that was performed by the Southern African Grain Laboratory (SAGL), 176 crop samples representing the various production regions were tested. Samples, provided by members of Agbiz Grain, were forwarded to SAGL by commercial grain silo owners and received from July to November 2018.

    Test weight

    The average test weight (a measure of the bulk density) of sunflower seed for this season was 40,1kg/hl, ranging between 33,2kg/hl and 45,9kg/hl. The previous two seasons averaged 42,1kg/hl and 42,5kg/hl respectively, with values varying between 34,2kg/hl and 45,5kg/hl and 35kg/hl and 48,1kg/hl.

    These values were determined by following the standard working procedure for the Kern 222 instrument, as described in ISO 7971-3:2009, to obtain the g/1 litre filling mass of the sunflower seed samples, which was then divided by two. The g/½ litre filling mass thus obtained, was then used to extrapolate the test weight by means of formulas obtained from the Test Weight Conversion Chart for Sunflower Seed Oil of the Canadian Grain Commission.

    Quality – grading

    Full grading was done in accordance with the regulations relating to the grading, packing and marking of sunflower seed intended for sale in the Republic of South Africa (Government Notice No. 45 of 22 January 2016). 143 (81%) of the samples were graded as Grade FH1, while 33 samples were downgraded to Class Other Sunflower Seed (COSF). During the previous two seasons, 15% (2016/2017) and 22% (2015/2016) of the samples were downgraded to COSF.

    The majority of samples (18) downgraded this season was as a result of the percentage of either the screenings or the collective deviations or a combination of both exceeding the maximum permissible deviations of 4% and 6% respectively. A further nine samples were downgraded due to the presence of poisonous seeds (either Datura spp., Crotalaria spp. or Xanthium strumarium) exceeding the maximum permissible number.

    North West (99 samples) reported the highest weighted average percentage screenings namely 2,18%, followed by Limpopo (five samples) and the Free State (64 samples) with 1,84% and 1,56% respectively. Mpumalanga (eight samples) reported the lowest average percentage screenings of 1,33%. The weighted national average was 1,91% compared to the 2,18% of the previous season. No samples were received from Gauteng this season.

    The highest weighted percentage foreign matter (1,4%) was reported on the samples from the Free State. The North West Province averaged 1,06% and the lowest percentages were found in Limpopo and Mpumalanga with 0,68% and 0,66% respectively. The South African average was 1,16% compared to the 1,06% and 1,41% of the previous two seasons.


    The number of samples received for this survey that contained sclerotia from the fungus Sclerotinia sclerotiorum, almost tripled from 28 samples (16%) in the previous season to 78 samples (44%) this season. 41 of these samples originated in the North West Province, 28 in the Free State, eight in Mpumalanga and one in Limpopo.

    However, none of these samples exceeded the maximum permissible deviation of 4%. Weighted average levels ranged from 0,02% in Limpopo to 0,36% in Mpumalanga. The national average of 0,17% was slightly higher than the 0,11% of the previous season.

    According to an article published by researchers of the Department of Plant Sciences: Plant Pathology Division of the University of the Free State, there are currently no commercially available sunflower or soybean cultivars worldwide that are resistant to the fungus (pathogen). The manner in which sunflower and soybean cultivars differ in their response to the pathogen under disease-favourable conditions, enables farmers to select cultivars that are more tolerant. This reduces the risk of infection, yield losses and inoculum build-up in the fields.

    Kern 222 instrument

    Quality – nutritional content

    The weighted average crude protein content (total nitrogen content x 6,25) was 16,61% – similar to the 16,63% of the previous season. The North West Province had the highest weighted average crude protein content of 17,12% and Mpumalanga the lowest with 15,15%.

    Mpumalanga has consistently reported the lowest average protein content since commencement of this survey in the 2012/2013 season. Limpopo's crude protein content averaged 16,95% and that of the Free State 15,97%. Please refer to Graph 1 for the average crude protein content per province over the last five seasons.

    The weighted average crude oil percentage of 37% was the lowest of the last six seasons and 1,6% lower than the previous season. Mpumalanga had the highest weighted average crude oil content of 40%. Last season Mpumalanga also reported the highest oil content. The lowest average oil content was the 36,1% of the North West Province (also the lowest in the previous season). Please see Graph 2.

    However, the weighted average percentage crude fibre was the highest of the six seasons at 21,9% (21% in 2016/2017). Average values varied between 20,2% in Limpopo to 22,2% in the Free State. The weighted average ash (mineral matter) content is also the highest over six seasons (2,69%), last season being 2,52%. The provincial averages ranged from 2,56% in Mpumalanga to 2,74% in Limpopo. The nutritional component analyses are reported as percentage (g/100 g) on an 'as received' or 'as is' basis.

    Graph 1: Average crude protein content per season
    Graph 1: Average crude protein content per season
    Graph 2: Average crude oil content per season
    Graph 2: Average crude oil content per season

    Production overview

    World sunflower seed production in the 2017/2018 season decreased by 1% year on year to 49,6 million tons. The local crop followed the same trend, decreasing by 1,4% (12 000 tons) to 862 000 tons.

    Globally, the Ukraine and Russia are the main sunflower producing countries, while in South Africa the Free State and the North West Province contributed 95% of the total crop. World sunflower seed production figures were obtained from the National Sunflower Association's website and national figures from the Crop Estimates Committee (CEC) of the Department of Agriculture, Forestry and Fisheries (now Agriculture, Land Reform and Rural Development).

    The area utilised for sunflower production decreased by 5,4% to 601 500 ha, compared to the 635 700 ha of the previous season. This season's area planted is in line with the five-year average of 606 780 ha.

    South Africa's national yield average increased by 4,4% to 1,43 t/ha, the highest national average to date. Less than 1,5% of the sunflower seed produced in the country this season was planted under irrigation.

  5. The role of seedling diseases in poor establishment of sunflower in South Africa

    Dr SC Lamprecht

    Poor establishment has been identified as one of the important constraints in sunflower production in South Africa. Although the contribution of other factors such as seedling vigour, seedbed preparation and soil temperature to poor establishment have been investigated, there is no information on the role of seedling diseases as a production constraint in sunflower production in South Africa. The main aim of this study is to determine the incidence of seedling diseases of sunflower and the major causal organisms associated with these diseases, as well as the efficacy of the standard seed treatment to control the most important pathogens. The first phase of the project involved surveys and sampling of diseased sunflower seedlings and morphological and preliminary molecular characterization of fungi associated with diseased tissues. The second phase included the final molecular characterization of the potential pathogens and the evaluation of their pathogenicity and virulence and the effect of the standard seed treatment on survival, growth, root, hypocotyl and cotyledon rot severity under glasshouse conditions. Since it was found that the standard seed treatment is not effective against many of the virulent pathogens identified in this study, the third phase of the project evaluated other seed treatments against these pathogens. Species or anastomosis groups of Alternaria, Bipolaris, Curvularia, Epicoccum / Phoma, Diaporthe, Fusarium, Macrophomina, Nothophoma, Pythium, Rhizoctonia and Setosphaeria were included in the seed treatment trials. Four chemical products [ST2 = Experimental Code BYF14182 from Bayer Crop Science with a.i. penflufen, prothioconazole and metalaxyl + Cruiser® 600 FS from Syngenta with a.i. thiamethoxam (insecticide); ST3 = Experimental Code BYF14182 from Bayer Crop Science with a.i. penflufen, prothioconazole and metalaxyl + Cruiser® 600 FS from Syngenta with a.i. thiamethoxam (higher dosage than ST2); ST4 = Cruiser® 600 FS from Syngenta with a.i. thiamethoxam (insecticide); ST5 = Celest® XL from Syngenta with a.i. fludioxonil and mefenoxam + Cruiser® 600 FS from Syngenta with a.i. thiamethoxam (insecticide) (standard registered seed treatment for sunflower); ST6 = Maxim Quattro® from Syngenta with a.i. thiabendazole, azoxystrobin, fludioxonil and mefenoxam + Cruiser® 600 FS from Syngenta with a.i thiamethoxam (insecticide)] and three biological products [ST7 = Tri-Cure® from MBFI with a.i. Trichoderma harzianum + Cruiser® 600 FS from Syngenta with a.i. thiamethoxam (insecticide); ST8 = TrichoPlus® from BASF with a.i. Trichoderma fertile; ST9 = Tri-Cure® from MBFi with a.i. Trichoderma harzianum + Rizofos® from MBFi with a.i. Pseudomonas fluorescens + Premax® (protector to increase the survival of P. fluorescens) also from MBFi] were included in the study. Seed treatment ST1 was untreated seed. The pathogens Alternaria spp., Bipolaris spp., Curvularia spp., Diaporthe spp., Epicoccum sorghina, Macrophomina phaseolina and Septopharia rostrata that were included in the study did not cause a significant reduction in survival or plant growth and did not cause significant root or hypocotyl rot. The most virulent pathogens where the differences in the efficacy of the different treatments were demonstrated in terms of an increase in survival, plant growth and reduction in root and hypocotyl rot severity were anastomosis groups within Rhizoctonia solani, Pythium irregulare, Pythium ultimum var. ultimum and FSSC (F. solani species complex). The most virulent anastomosis groups (AGs) within R. solani were AG2-2LP, AG4-HGI, AG4-HGIII and SB3 (unidentified AG) with AG4-HGI significantly more virulent than the other AGs. With regards to survival of seedlings in soil inoculated with the different AGs, ST2, ST3, ST5 and ST6 were equaly effective in improving survival in soil inoculated with AG2-2LP, AG4-HGIII and SB3. However, only ST2, ST3 and ST6 significantly improved survival in soil inoculated with AG4-HGI with ST6 significantly more effective than ST2 and ST3. Seed treatment ST3 significantly improved plant growth in soil inoculated with AG4-HGIII and ST2, ST3, ST6 and ST7 significantly improved growth of seedlings in soil inoculated with AG4-HGI. Seed treatment ST5 was significantly more effective in reducing root rot caused by AG2-2LP, ST6 against root rot caused by AG4-HGI and ST3 significantly improved growth in soil inoculated with AG4-HGIII. Seed treatments ST4, ST8 and ST9 significantly increased root rot severity caused by AG4-HGI and ST7 also significantly increased root rot severity caused by SB3. Pythium irregulare and P. ultimum var. ultimum did not cause a significant reduction in survival of seedlings, but both pathogens significantly reduced growth of seedlings and both pathogens caused root and hypocotyl rot. However, P. irregulare was significantly more virulent than P. ultimum var. ultimum. Seed treatments ST2, ST3, ST5 and ST6 significantly improved plant growth and significantly reduced root rot severity, but ST2, ST3 and ST6 were significantly more effective than ST5 in improving plant growth and ST2 and ST6 significantly more effective than ST5 in reducing root rot severity. Seed treatments ST2, ST3 and ST6 also significantly improved growth in soil inoculated with P. ultimum var. ultimum and ST2 and ST5 significantly reduced root rot severity. The only biological product that significantly reduced root rot in soil inoculated with P. irregulare was ST8, but root rot severity for this treatment was also significantly higher than for seed treatment ST5. The hypocotyl rot was more severe for seedlings in soil inoculated with P. irregulare than soil inoculated with P. ultimum var. ultimum. All treatments reduced hypocotyl rot caused by P. irregulare with the lowest hypocotyl rot reported for treatments ST2, ST3 and ST9. Fusarium solani species complex (FSSC) caused a significant reduction in survival, growth of seedlings and a significant increase in root rot severity. Considering all the parameters evaluated, seed treatments ST2 and ST6 were the most effective against FSSC. Survival was improved by seed treatments ST2, ST3, ST5 and ST6, but this improvement was only significant for treatments ST5 and ST6. Treatments ST2, ST6 and ST8 significantly improved plant growth and treatments ST2 and ST6 significantly reduced root rot severity. All treatments significantly reduced hypocotyl rot severity with the highest reduction recorded for ST6. Seed treatment ST3 had the same active ingredients than ST2, but the application dosage was just higher than for ST2. The results showed that ST2 and ST3 performed similar in their effect to improve survival and plant growth and also to reduce root and hypocotyl rot severity, and in many instances such as in soil inoculated with P. ultimum var. ultimum or FSSC, ST2 performed better than ST3. The effect of seed treatment ST4 (insecticide) for the different parameters recorded, did in most instances not differ significantly from the control (ST1) for the most virulent pathogens. Seed treatment ST4 also significantly increased root rot severity of seedlings in soil inoculated with R. solani AG4-HGI, P. irregulare and P. ultimum var. ultimum. Also for ST7 which was a combination of Trichoderma harzianum and thiamethoxam there was a significant increase in root rot severity in soil inoculated with R. solani AG4-HGI. In conclusion, this study showed that there are chemical products that performed as well and in certain instances better than the standard seed treatment (ST5). Also important is that these compounds were not phytotoxic to the seedlings. Proper establishment of seedlings is very important to improve yield and an essential component of sustainable production. Many of the pathogens affecting sunflower seedlings have a broad host range and cannot be controlled with crop rotation. Since crop rotation is such an important part of conservation agriculture, crops that are susceptible to some of the same pathogens such as maize, sunflower and soybean are often rotated in the same field. In order to protect seedlings against these pathogens with a broad host range, effective seed treatment can play a significant role and should be included in an integrated management strategy against soilborne diseases of sunflower. The current study demonstrated the ability of seed treatments which included combinations of active ingredients to effectively target a complex of pathogens associated with sunflower seedlings to significantly improve survival, growth and reduce root and hypocotyl rot severity. The results obtained under glasshouse conditions showed that it would be worth while to evaluate these treatments under field conditions in different production areas with different disease complexes and climatic conditions in order to confirm the positive results obtained in the glasshouse study, and also to motivate for registration of products that effectively controlled the most virulent pathogens of sunflower seedlings.

  6. National soybean cultivar trials

    AS de Beer, L Bonkhorst, HSJ Vermeulen, NN Mogapi, TC Ramatlotlo and S Seutlwadi
    ARC Grain Crops Institute, Potchefstroom

    A total of 32 commercially certified cultivars were evaluated for the moderate as well as for the warm areas and 28 cultivars for the cool area, for the 2018/19 season in 22 field trials. Only GMO cultivars were included in the trials and Roundup applications were used during the execution of the trials. A randomised latinised row/colum design with three replicates was used for all field trials. Date of flowering (50% flowering), date of harvest maturity, length of growing season, plant height, pod height, green stem, lodging, shattering, 100 seeds mass, undesirable seed and the yield probability of cultivars calculated. Yield probabilities served as guideline for cultivar selection.

    The mean number of days from planting to 50% flowering of cultivars for the cool-, moderate and warm areas were 75, 59 and 48 days respectively. The overall mean yield was 3 242kg ha-1 for the cooler areas, 3 397kg ha-1 for the moderate and 3 732kg ha-1 for the warm areas. Cultivars with a high yield probability are important in the selection of cultivars by producers due to the reliability of the expected future yield. DM 5302 RSF and P64T39 R for the cooler areas, PAN 1521 R for the moderate and warm areas as well as P61T38 R and P64T39R for the warmer areas.


    The National Soybean Cultivar Trials (project P05000002) were planted for the 41th successive year during the 2018/19 growing seasons. A total of 32 commercial certified cultivars were evaluated for the moderate as well as warm area and 28 cultivars for the cool area in 22 field trials, illustrated in locality list (Table 1).

    The aim of the project was primarily the following:

    • To compare cultivars for agronomic and economic performance;
    • To test the adaptability of cultivars and new releases for specific areas and cultivation practices.

    Materials and methods


    Field trial sites were selected by co-workers and the ARC-GC, depending on the availability of land at various localities. The localities represented a wide range of climatic conditions and included Departmental and ARC Research Stations as well as on privately owned farms. Only GMO cultivars were included in the national cultivar trials. A randomised latinised row/colum design with three replicates was used for the statistical layout. Procedures relating to the planting and care of trials were described in a manual that accompanied the seed for each trial. Planting time and cultivation practices were executed to correspond with that of commercial plantings in the specific areas.

    Planning of trial layout and packing of seed according to trial plans was done by ARC-GC. The results of all trials were analysed by ARC-GCI and only trials that passed a strict statistical test for reliability was included in a yield reliability analysis for cultivars. From these results cultivar recommendations were done.

    Each trial plot consisted of four, 5m rows. Four metres were harvested from each of the middle two rows, in order to avoid border effects. Soil form, fertilization and weed control together with row spacing as well as rainfall and irrigation was also recorded. All seeds were inoculated with Bradyrhizobium japonicumbacteria at planting.

    Observations were recorded by responsible officers of the ARC as well as collaborators from seed companies, Co-ops and Departmental staff members.


    The following agronomical observations in the trials were done:

    • Date of flowering (50% flowering)
    • Date of harvest maturity
    • Length of growing season
    • Plant height
    • Pod height
    • Green stem
    • Lodging
    • Shattering
    • 100 seed mass
    • Undesirable seed
    • Grain Yield (calculated on a basis of 12.5% moisture content)

    Only the most important agronomic data like days to flower and yield will be discussed in this report. Other factors mentioned above as well as planting dates, amount of fertiliser applied, soil analyses, rainfall, irrigation and other agronomic details from some field trials were reported in the annual progress reports. Data from the previous season has also been included for reference purposes.

    Results and discussion

    Table1: Localities for the 2018/19 planting season
    2018/2019 Irrigation / Dry land
    Bapsfontein I
    Bergville I
    Bethlehem D
    Bossies D (Not planted to dry)
    Brits K2 I
    Cedara D
    Clarens D
    Clocolan D (High CV%. Poor emergence due to severe drought just after planting)
    Delmas D (Data not reliable)
    Dundee D (Data not reliable)
    Greytown (PANNAR) D
    Groblersdal I
    Hertzogville D (Not planted to dry)
    Hoopstad D (Not planted to dry)
    (Hopetown) Skuinsdrift B (Insufficient data)
    Kinross D
    Kokstad D
    Kroonstad D
    Leeudoringstad D (Poor emergence due to severe drought just after planting. Trial terminated.)
    Marble Hall I
    Middelburg D (Poor emergence due to severe drought just after planting. Trial terminated.)
    Potchefstroom (I) I
    Potchefstroom Seed Co D (High CV%. Poor emergence and low yield due to severe drought.)
    Schweizer-Reneke D (Poor emergence due to severe drought. Trial terminated.)
    Stoffberg D
    • 25 trials on 24 localities planned.
    • 3 trials not planted due to drought.
    • 16 trials successful (3 terminated due to drought; 1 insufficient data; 2 high CV%).
    Table 2: Days from planting to 50% flowering of soybean cultivars evaluated in the cool production areas for 2017/18 and 2018/19
    Cultivar Cool area Moderate area Warm area
    7 Loc 5 Loc 7 Loc 5 Loc 2 Loc 1 Loc
    2017/2018 2018/2019 Mean 2017/2018 2018/2019 Mean 2017/2018 2018/2019 Mean
    P48T48 R 63 62 63 53 48 51 42 42
    LS 6248 R 83 69 76 69 59 64 48 42 45
    DM 5953 RSF 63 65 64 53 49 51 43 37 40
    SSS 5449 (tuc) 77 77 77 68 57 62 48 43 46
    NS 5009 R 65 65 65 55 49 52 44 40 42
    NS 5258 R 63 64 64 54 52 53 44 39 42
    PAN 1532 R 84 77 81 69 54 62 49 43 46
    LDC 5.3 76 76 54 54 44 44
    DM 5351 RSF 63 66 64 54 49 52 43 36 40
    Y 540 76 79 77 65 65 48 48
    SSS 5052 (tuc) 86 77 82 72 59 66 49 49 49
    NA 5509 R 85 81 83 72 62 67 51 49 50
    LS 6851 R 83 72 78 68 57 63 49 45 47
    PAN 1521 R 86 77 81 73 61 67 51 47 49
    DM 5302 RSF 80 76 78 69 53 61 49 45 47
    NS 5909 R 88 81 84 75 60 68 53 51 52
    LDC 5.9 76 76 60 60 50 50
    DM 5901 RSF 78 78 62 62 49 49
    LS 6860 R 93 79 86 76 64 70 53 48 51
    LS 6164 R 74 74 61 61 51 51
    P61T38 R 82 71 76 74 62 68 52 52 52
    Y 605 65 65
    LS 6161 R 86 77 82 74 61 67 52 51 52
    SSS 6560 (tuc) 82 82 73 73 51 50 51
    Y 627 84 84 72 62 67 52 49 51
    DM 6663 RSF 90 90 78 66 72 52 52 52
    NS 6448 R 86 79 83 76 65 70 50 52 51
    P64T39 R 89 82 86 76 64 70 51 52 52
    Y 657 90 90 77 65 71 53 54 54
    PAN 1644 R 80 80 63 63 54 54
    PAN 1653 R 82 82 65 65 49 49
    LS 6868 R 93 85 89 79 63 71 53 53 53
    DM 6.8i RR 89 89 77 64 71 54 54 54
    DM 6968 RSF 55 55
    P71T74 R 65 65 56 56
    PAN 1454 R 59 59 56 56 42 42
    PHB 94 Y 80 R 58 58 54 54 40 40
    Y 550 81 81 70 70 49 49
    DM 5609 RSF 84 84 73 73 49 49
    PHB 96 T 06 R 92 92 76 76 52 52
    PAN 1623 R 87 87 73 73 52 52
    LS 6862 R 87 87 74 74 50 50
    NS 6267 R 83 83 72 72 50 50
    DM 6402 RSF 93 93 78 78 52 52
    Gem / Mean 81 75 79 69 59 65 49 48 49

    The mean number of days from planting to 50% flowering of cultivars ranged from 62 to 85 days for the cool production areas, 48 to 66 and 36 to 56 respectively for the moderate and warm areas. Calculated across cultivars and seasons, this period was 75 days for the cool, 59 for the moderate and 48 days for the warm production areas.

    Table 3: The yield (t ha-1) of cultivars evaluated for 2017/18 and 2018/19
    Cultivar Cool area Moderate area Warm area
    7 Loc 5 Loc 8 Loc 6 Loc 3 Loc
    2017/2018 2018/2019 Mean 2017/2018 2018/2019 Mean 2017/2018 2018/2019 Mean
    P48T48 R 2979 2939 2959 3189 3099 3144 2996 2996
    LS 6248 R 2454 2997 2725 3139 3167 3153 3613 3502 3558
    DM 5953 RSF 3531 3367 3449 3512 3200 3356 3240 4157 3699
    SSS 5449 (tuc) 2792 3054 2923 3082 3217 3150 3484 3430 3457
    NS 5009 R 2593 2598 2596 2968 2845 2906 3605 3419 3512
    NS 5258 R 3082 2834 2958 3244 3256 3250 3334 3788 3561
    PAN 1532 R 3065 3242 3154 3486 3587 3537 3195 3323 3259
    LDC 5.3 3458 3458 3315 3315 3822 3822
    DM 5351 RSF 3499 3641 3570 3383 3535 3459 3624 3861 3743
    Y 540 3104 1854 2479 3629 3629 3270 3270
    SSS 5052 (tuc) 2383 3306 2844 3264 3450 3357 3526 4083 3805
    NA 5509 R 2985 3331 3158 3345 3661 3503 3989 3838 3914
    LS 6851 R 2294 3484 2889 3405 3525 3465 3258 3667 3463
    PAN 1521 R 3049 3429 3239 3569 3497 3533 4049 3979 4014
    DM 5302 RSF 3029 3546 3287 3401 3223 3312 3564 3495 3530
    NS 5909 R 2465 3262 2863 3431 3588 3510 3637 3822 3730
    LDC 5.9 3783 3783 3842 3842 3901 3901
    DM 5901 RSF 3400 3400 3664 3664 3627 3627
    LS 6860 R 2081 3157 2619 3195 3310 3253 3706 3222 3464
    LS 6164 R 3275 3257 3425 3425 3751 3751
    P61T38 R 2797 3153 2975 3621 3585 3603 4386 3455 3920
    Y 605 1832 1832
    LS 6161 R 2592 3352 2972 3133 3402 3268 3623 3614 3619
    SSS 6560 (tuc) 2592 2592 3360 3360 3962 3811 3886
    Y 627 2884 2884 3642 3667 3655 3393 3838 3615
    DM 6663 RSF 2615 2615 3401 3315 3358 3930 3952 3941
    NS 6448 R 2805 3409 3107 3420 3772 3596 3943 3651 3797
    P64T39 R 2881 3922 3402 3558 3813 3686 4360 3991 4175
    Y 657 3020 3020 3651 3739 3695 4277 3948 4113
    PAN 1644 R 3425 3425 3492 3492 3668 3668
    PAN 1653 R 3178 3178 3566 3566 3865 3865
    LS 6868 R 2143 3146 2645 2804 2936 2870 3455 3279 3367
    DM 6.8i RR 2437 2437 3571 3479 3525 3855 3688 3771
    DM 6968 RSF 3763 3763
    P71T74 R 3709 3709 4211 4211
    PAN 1454 R 2704 2704 2864 2864 3200 3200
    PHB 94 Y 80 R 3234 3234 3066 3066 3024 3024
    Y 550 2790 2790 2996 2996 3232 3232
    DM 5609 RSF 2861 2861 3454 3454 3347 3347
    PHB 96 T 06 R 2750 2750 3247 3247 3551 3551
    PAN 1623 R 3134 3134 3227 3227 4090 4090
    LS 6862 R 2564 2564 3602 3602 3907 3907
    NS 6267 R 3013 3013 3529 3529 3514 3514
    DM 6402 RSF 2053 2053 3263 3263 3629 3629
    Gem / Mean 2779 3242 2975 3333 3397 3354 3622 3732 3658

    The mean yield of cultivars within climate areas varied from 1854 (Y 540) to 399kg ha-1 (P64T39 R), 1832 (Y 605) to 3 842kg ha-1 (LDC 5.9) and 3222 (LS 6860 R) to 4 211kg ha-1 (P71T74 R) respectively for the cool, moderate and warm production areas. The overall mean yield was 3 242kg ha-1, 3 397kg ha-1 and 3 732kg ha-1 (cool, moderate, warm).

    Figure 1: Average yield of the different cultivars for the cool dryland areas for the 2018/2019 season
    Figure 1: Average yield of the different cultivars for the cool dryland areas for the 2018/2019 season
    Figure 2: Average yield of the different cultivars for the moderate dryland areas for the 2018/2019 season
    Figure 2: Average yield of the different cultivars for the moderate dryland areas for the 2018/2019 season
    Figure 3: Average yield of the different cultivars for the warm irrigation areas for the 2018/2019 season
    Figure 3: Average yield of the different cultivars for the warm irrigation areas for the 2018/2019 season

    Relatively large differences in annual mean yields existed among cultivars. The mean difference between the highest and lowest yielding cultivars varied from 2 068kg ha-1, 2 010kg ha-1 and 989kg ha-1 respectively for the cool, moderate and warm areas (Fig: 1, 2 and 3).

    Yield probability

    The yield probability is being indicated in Tables 4 to 6. Some cultivars showed a high probability at the lower yield potentials (DM 5953 RSF) for the cool and warm areas and PAN 1623 R for the moderate areas, while the same cultivar PAN 1623 R tend to have a high yield probability at the higher yield potentials in the cooler areas. The cultivars DM 5302 RSF and P64T39 R can be regarded as all-rounder cultivars over all the yield potentials in the cooler, PAN 1521 R for the moderate and warm area as well as P61T38 R and P64T39 R also for the warm area.

    Table 4: Yield probability at different yield targets for the cooler dryland production areas, 2016/17, 2017/18 and 2018/19
    Cultivar Yield potential (t/ha)
    1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
    LS 6248 R 49 44 36 28 21 16 13 13
    DM 5953 RSF 39 46 55 66 75 83 87 89
    SSS 5449 (tuc) 15 16 19 24 32 42 52 61
    PAN 1532 R 80 77 72 64 53 41 32 25
    SSS 5052 (tuc) 27 22 18 14 11 10 10 12
    PAN 1521 R 40 46 53 62 71 78 83 85
    DM 5302 RSF 54 56 58 61 63 64 64 65
    NS 5909 R 17 18 20 24 31 40 49 57
    P 61T38 R 48 46 43 40 36 34 32 32
    LS 6161 R 56 53 48 44 38 33 30 28
    NS 6448 R 92 90 86 78 65 49 34 24
    P 64T39 R 60 64 67 71 74 76 76 75
    Table 5: Yield probability at different yield targets for the moderate dryland production areas, 2016/17, 2017/18 and 2018/19
    Cultivar Yield potential (t/ha)
    1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
    LS 6248 R 49 44 38 33 28 23 20 17
    DM 5953 RSF 56 54 52 49 47 45 42 40
    SSS 5449 (tuc) 49 43 37 30 25 20 16 13
    PAN 1532 R 64 61 57 53 48 44 40 36
    SSS 5052 (tuc) 74 69 62 54 46 38 30 24
    PAN 1521 R 94 92 89 86 81 75 68 60
    DM 5302 RSF 81 76 68 59 49 38 29 22
    NS 5909 R 22 28 35 44 53 63 71 78
    P 61T38 R 27 34 43 52 62 71 79 85
    LS 6161 R 64 60 54 49 42 37 31 28
    DM 6663 RSF 25 29 33 38 43 49 55 60
    NS 6448 R 17 23 30 39 49 60 68 77
    P 64T39 R 37 45 54 63 71 79 84 89
    DM 6.8i RR 44 47 51 55 58 62 65 68
    Table 6: Yield probability at different yield targets for the warm irrigation production areas, 2016/17, 2017/18 and 2018/19
    Cultivar Yield potential (t/ha)
    1.5 2.0 2.5 3.0 3.5 4.0 4.5
    LS 6248 R 43 42 43 44 45 47 49
    DM 5953 RSF 91 88 83 73 58 40 26
    SSS 5449 (tuc) 44 43 41 40 39 39 39
    PAN 1532 R 9 8 7 6 7 9 14
    SSS 5052 (tuc) 14 16 21 29 42 55 68
    PAN 1521 R 94 94 93 89 83 71 57
    DM 5302 RSF 37 37 38 39 42 44 48
    NS 5909 R 23 23 23 23 26 30 36
    P 61T38 R 78 78 78 76 73 67 61
    LS 6161 R 2 3 6 15 39 68 87
    SSS 6560 (tuc) 7 10 18 32 56 77 89
    DM 6663 RSF 57 57 56 55 53 51 49
    NS 6448 R 38 37 36 35 36 37 40
    P 64T39 R 51 54 60 65 71 74 77
    DM 6.8i RR 87 85 82 75 66 52 41

    The yield probability of a cultivar is the chance to get an above average yield at a particular yield potential. For instance, if the yield probability of a cultivar, at a particular yield potential equals 60%, the chance to get a yield above the mean of all cultivars is 60% with a 40% chance of obtaining a yield below the mean.

    Yield probability values of the 12, 14 and 15 cultivars for the three production areas (cool, moderate and warm) are presented in Tables 4, 5 and 6. Tables 4, 5 and 6 contain information regarding cultivars included in the trials for three years. Since new cultivars are introduced and some removed annually, a multi-season yield probability is only possible.

    Table 7: Summary of the trials planted over the reporting period 20012/13 to 2018/19
    Season Trials planted Trials failed Success rate Cultivars
    2007/08 27 5 81.5% 20
    2008/09 27 8 70.4% 18
    2009/10 40 19 52.5% 21
    2010/11 24 7 70.8% 19
    2011/12 28 9 68.0% 23
    2012/13 24 9 62.5% 27
    2013/14 21 5 76.2% 31
    2014/15 22 3 86.4% 29
    2015/16 19 5 73.7% 26 + 2 retained seed
    2016/17 21 2 90.5% 32
    2017/18 21 3 85.7% 35
    2018/19 22 8 63.6% 28 (cool) 32 (moderate and warm)

    Six of the 35 (16.7%) cultivars evaluated during the 2018/19 season were new cultivars. This results emphasise the importance of the constant evaluation of cultivars in an aggressively growing soybean industry. The average success rate of the trials decreased (22.4%) from the 2017/18 to the 2018/19 growing season, due to the severe drought experienced during the crop production season.


    The best cultivar selection for a specific climatic region can have a significant financial impact for the producer and it is thus from utmost importance to continue with unbiased cultivar trials.

  7. Website

    Ms M du Preez and Ms Y Papadimitropoulos

    Since its creation in 2009, the OPDT/OAC website has mainly focused on providing information related to sunflower, groundnuts, and other oilseeds. This changed when soybeans and canola were added to the research database in 2016, which almost quadrupled the amount of information now available on the website.

    New main sections were added for the Canolafokus and Income and Cost Budgets with subcategories added to the Crops, Statistics and Estimates, Gallery and Video Gallery sections.

    This expansion of knowledge coincided with rapid advances in technology application that also led to major updates of the programming languages used to build the OPDT/OAC website. As a result, it was proposed that the base programming should be brought into line with the latest developments and to also update the look and feel of the website at the same time.

    In July 2018 a visual proposal was presented in PDF format to the OPDT/OAC and – upon approval of this presentation – we unveiled a demo site in November 2018 for final approval before work is started on converting all the content files.

    Visitor statistics
    Reporting Year Unique Visitors
    Raw values* Google values
    Visitors Pages Pages per visit
    2007 11
    2008 74
    2009 752
    2010 2964
    2011 4037 788 3284 3.21
    2012 4052 720 3775 4.25
    2013 4342 674 3296 3.94
    2014 4503 1086 4600 3.62
    2015 4800 1340 6993 4.14
    2016 4329 927 5042 3.83
    2017 6384 909 5406 4.39
    2018 5428 1841 8885 3.59

    Google values show an increase in page views and a significant increase in unique visitors. Pages per visit decreased moderately. The most page views came from the following pages in order of percentage share:

    • Homepage, 15.63%
    • Oilseed Focus Magazine, 3.07%
    • Contact Us, 2.98%
    • Bursary Application, 2.77%
    • Crops 2.27%

    The future of the OPDT / OAC website

    Work has commenced on the new website content and its expected for it to go live later in 2019.

  8. Oilseeds information

    MR N Hawkins
    SAGIS Forums and Trusts

    Information to follow.

  9. Evaluation of shortened canola production periods and the use of alternative crops on the sustainability of winter grain production under conservation agricultural practices in the Riversdale Flats

    JA Strauss
    Western Cape Department of Agriculture

    2018 was the 7th year of production on the new trial. Six cash crop systems were tested including shortened canola rotations and cover crops. A total of 60 plots were planted. The 6 systems tested are replicated 3 times and all crops within each system are represented on the field each year.

    All protocols developed during the annual technical committee meeting in February 2018 were followed and the integrity of the trial layout was upheld.

    Riversdale received very little summer rainfall, which resulted in a very dry start to the 2017 production season. Only 65mm fell from January to the end of April. In 2017 a new weather station was installed at the research site which is managed by the Department of Agriculture. A total of 128mm was received from April to the end of September.

    Wheat production

    SST0127 was planted at Riversdale at 65 kg/ha. A total of 38kg N/ha was applied to each plot (8kg N/ha at planting and 30kg N/ha top-dressing). Wheat yields at Riversdale averaged 1 885ha. This was 441 kg/ha more than in 2017.

    Canola production

    Diamond was planted at Riversdale at 3,5kg/ha. A total of 38kg N/ha was applied to each plot (8kg N/ha at planting and 30kg N/ha top-dressing). Canola yields at Riversdale averaged 1 662kg/ha with all plots showing oil yield of over 40%.

    Barley production

    Hessekwa was planted at Riversdale at 65kg/ha. Barley yields at Riversdale averaged 2 515kg/ha. This average yield was 1 304kg/ha more than in 2017. All plots were classified as malting grade.

    Lupin production

    A bitter lupine mixture was planted at a rate of 100kg/ha. No lupine plots were harvested due to poor germination and weed problems in the very low rainfall year.

    Cover crops

    Saia oats and field peas were planted at Riversdale at 25kg/ha and 80kg/ha, respectively. No other input cost was incurred during the season except the herbicide cost to kill the cover crop following the information day.


    Although 2018 proved to be a poor production year, all systems tested show a positive gross margin above directly allocated production costs.

  10. Biological control of sclerotinia of sunflower

    Prof M Laing
    University of KwaZulu-Natal

    No progress was made with the project during the reporting year.

  11. The evaluation of locally produced full-fat canola and canola oilcake meal as protein sources in diets for slaughter ostriches

    Prof TS Brand
    Department of Agriculture: Western Cape

    The project is composed out of two trials that investigates different traits to optimize the commercial benefits of the results obtained from the research.

    Trial 1 – Growth trial

    In this investigation 150-180 day old chicks was randomly divided into 15 groups of 10 to 12 chicks per group and remained in their respective group for each production stage. The growth trial only started at onset of the starter phase (± 84 days old on the 3rd of February 2016) and was concluded after the finisher phases (± 300 days old on the 15th of September 2016).

    During the lifetime of slaughter ostriches they are fed four different diets, namely:

    – Pre-starter (0-83 days of age)
    – Starter (84-147 days of age)
    – Grower (148-231 days of age)
    – Finisher (232-294 days of age)
    – Quarantine (245-308 days of age) still receiving finisher diet

    During the pre-starter phase the chicks received a standard diet. From the starter phase, two iso–nutrient diets were formulated respectively for the three remaining production phases (starter, grower and finisher) to contain soybean oilcake meal (control diet) or full-fat canola (treatment/alternative protein source). The full-fat canola incrementally replaced (0%, 25%, 50%, 75% and 100% of main protein source) the soybean oilcake meal. Five experimental diets with three groups of birds per diet were used in the study. Diet formulation was based on nutrient values presented in the Elsenburg Ostrich Feed Database. The water and feed supply was available ad libitum for each treatment.

    The production trial started off with a 160 birds for the starter phase on 3 February 2016. At the end of this phase there were 158 birds due to the fact that we had to cull birds with leg abnormalities. When the grower phase started on 6 April 2016 we took some birds out of the trial that were too light weight, the grower phase started with a 153 birds. The trial ran as expected, the birds were in a healthy condition with good growth and feed intake. The slaughter date was on the 15th of September 2016 where 146 birds were slaughtered. Samples of the meat, organs, skin and feathers were collected and most of the laboratory analysis have been completed.

    Most of the data for trial 1 have been statistically analysed, we are still waiting on certain results from laboratories, after which all statistical analysis will be completed. The following preliminary results have been obtained.

    Data that has been analysed showed that there were only differences between diets for dry matter intake during the grower phase. During this phase the 100% full-fat canola inclusion diet had the lowest feed intake compared to the rest of the diets that had no difference among them.

    Differences regarding growth was only observed during the grower phase and over the whole trial period. During the grower phase the ostriches receiving the 100% full-fat canola diet had the slowest growth (89.2g/bird/day vs. the 50% inclusion diet that showed 204.5g/bird/day) compared to the rest of the groups that again did not differ from one another. Over the whole period the 100% inclusion diet did not differ from the 25% or 75% inclusion diets and only differed from the 50% and 0% inclusion diets.

    Again with feed conversion ratio during the grower phase the 100% inclusion diet was the lowest and all the other treatment diets did not have significant differences.

    After all the data has been analysed the final results will be presented in the thesis.

    Trial 2 – Feed preference trial

    The preference of ostriches to the diets from a palatability point of view was investigated in a free-choice system to determine the effect of different inclusion levels of full-fat canola on feed preference of grower ostriches. Ten groups of six birds per group were used. After the data were analysed the investigation gave an indication as to the preferred inclusion level of the full-fat canola as protein source (0%, 25%, 50%, 75% and 100%) of the same iso-nutrient diets. Each group received the five experimental diets ad libitum in five different feed troughs for five consecutive days. The trial was replicated for another five days with a brief period in between. The water supply was also available ad libitum for each treatment.

    The data collected for trial 2 were statistically analysed and showed the following. There was in fact a difference in feed intake between diets (P<0.05). The birds had a statistically significant preference towards the 25% canola inclusion level in the feed and consumed 41.5% more of the 25% diet than all the other diets which did not differ from each other.

    At this stage we can come to the conclusion that birds favour the 25% inclusion level and will have a higher voluntary feed intake when receiving this diet compared to the other four diets, thus resulting in better growth.

    Time Frame

    The trials are being conducted at the Oudtshoorn Research Farm.

    Uncontrollable factors associated with ostrich farming such as high mortalities, disease outbreaks and hatching problems result in time frames acting as a guide and not definite certainties. Time frames will also depend on the availability of infrastructure as well as birds.


    • The growth trial on five different canola inclusion levels in the diet was carried out from November 2015 (Chicks hatching in November 2015 and Growth trial starting in February 2016) and ran until September 2016 when the animals were slaughtered at a registered abattoir and post slaughter data collected.
    • The preference trials with five canola inclusion levels in the diet, was conducted in May 2016 and June 2016.
    • From September 2016, data analyses and lab work started on the collected samples and was completed in June 2017.
    • Statistical analyses on data took place from January until June 2017.
    • Thesis writing started in February 2017 and will continue until September 2017.
  12. Developing soybeans as a modern food commodity

    Prof H Schönfeld
    University of Pretoria


    A report entitled "Developing soy beans as a modern food commodity – Consumer trends in the soy market of South Africa" was requested by Oil & Protein Seeds Development Trust (OPDT) and the Oilseeds Advisory Committee (OAC) in 2017. This report focused on general soy production, consumption and the contribution of soy to the South African diet. This addendum acts as an extension of this report, taking an in-depth look at import, export and production values of soy in the South African market. This addendum also further explores soy as a part of the National School Nutrition Programme.


    Soy bean consumption for human use in South Africa is predicted to be on the rise (Global Soybean Production, 2017). However, very limited data is available on the exact amounts that are used and the origin of these products with local data even contradicting this statement. In order to quantify and validate the role of soy in the South African diet a desktop review was launched. The review will attempt to better understand the supply and demand of soy for human consumption within the South African value chain.

    The soy that is used in food stuffs comes in varying forms that are processed using vastly different techniques from all over the world. In order to evaluated the source of the different derivatives of soy the base products had to be identified and traced back to origin.

    Main stream data capturing channels proved to have limited information on the import and export of soy products for human use. This may be either due to the low usage rate or that these products are mainly imported by private sector companies. In order to make future decisions and explore new market ventures it is important to have background knowledge of the origin and uses of a product within a value chain.

    The National Agricultural Marketing Council (NAMC) was chosen as a point of departure for data gathering for this review. The NAMC was established in terms of the Marketing of Agricultural Products Act No. 47 of 1996, as amended by Act No 59 of 1997 and Act No. 52 of 2001. This council acts as a statutory body reporting to the Minister of Agriculture, Forestry and Fisheries with the core mandate to capture information in four core divisions namely; Agribusiness Development, Agricultural Trusts, Statutory Measures and the Markets and Economic Research Centre (MERC). The goal of this council is to increase market access for all market participants; to promote the efficiency of the marketing of agricultural products; to optimise export earnings from agricultural products, and to enhance the viability of the agricultural sector. This is done in order to obtain valuable data on production of food stuffs in South Africa.

    Table 1 shows projections as made by the National Agricultural Research Council of South Africa on the consumption of soy in the production year 2016/2017 and the projected values for 2017/2018. In this table it can be seen that soy for human consumption is predicted to rise from 23 875 tons in 2016/2017 to 27 000 in 2017/2018. It is believed that these values are the most accurate values available on product supply and demand within South African borders (NAMC, 2018).

    Table 1: Detailed supply and demand for soybeans for January 2018 (NAMC, 2018)
    Marketing Season Final for 2016/2017 Projection for 2017/2018
    CEC (Crop Estimate) 742 000 1 316 370
    Retention 0 32 000
    Opening stock (1 March) 89 128 84 792
    Prod deliveries 713 660 1 284 370
    Imports for South Africa 271 098 28 000
    Surplus 1 122 3 000
    Total Supply 1 075 008 1 400 162
    Processed 974 901 1 047 000
    – human 23 875 27 000
    – animal feed (full fat soy) 98 718 140 000
    – crush (oil/oilcake) 852 308 880 000
    Withdrawn by producers 367 1 500
    Released to end‐consumers 1 098 1 000
    Seed for planting purposes 5 678 8 800
    Net receipts(‐)/disp(+) 1 427 500
    Deficit 0 0
    Exports 6 745 500
    Total Demand 990 216 1 059 300
    Closing Stock (28/29 Feb) 84 792 340 862
    – processed p/month 81 242 87 250
    – months' stock 1,0 3,9
    – days stock 32 119

    The NAMC data provided guidance of further paths to explore to delve deeper into the soy bean market of South Africa. The data used in the rest of the study was found using various key word searches over numerous platforms. Even though in some cases data from the varying sources delivered inconsistent and contradicting data all the data obtained are provided in this report.

    Soy consumption

    As seen in Figure 1 soy beans processed for human consumption comprises a very small part of total soy bean production in South Africa. In Figure 2 a change in the percentage soy beans used for human consumption can be seen fluctuating over the past few years. The average for the past 6 years is stated at 7% whilst the average for the past 3 years is at 9% indicating at a slight increase in local production for human consumption.

    On a 5 year average, 28 640 tons of soybeans was locally processed for human consumption. However, this volume does not satisfy the total demand needed in the human food sector. It is estimated that more than half of the soy market for human consumption relies on imports (SAGIS, 2018).

    Figure 3 shows the tonnage of soy beans processed for human consumption in South Africa. This does however not state which percentage originated in South Africa. In this figure slight increases and decrease can be seen over the period 2005-2010 with the highest amount being achieved in 2010 (29 600 tons).

    However, this information as received from the NAMC and South African Grain Information Services (SAGIS) does not clearly stipulate the exact amounts that are locally grown crops for human consumption (NAMC, 2018; SAGIS, 2018).

    Figure 1: Breakdown of soybean for domestic use from 2005-2010 (SAGIS, 2018)
    Figure 1: Breakdown of soybean for domestic use from 2005-2010 (SAGIS, 2018)
    Figure 2: Domestic use of soybeans from domestically produced soybeans (SAGIS, 2018)
    Figure 2: Domestic use of soybeans from domestically produced soybeans (SAGIS, 2018)
    Figure 3: Soybeans processed for human consumption in South Africa
    Figure 3: Soybeans processed for human consumption in South Africa

    In Table 2 the fiscal import values of different soy derivatives is shown. The total rand value of imported soy products was estimated at R344 million showing enormous potential for the production of soy derivatives within South African borders.

    Table 2: Import values of soy products to South Africa
    (Van der Merwe, et al., 2014; Dlamini, et al., 2014)
    Soy protein product type Import (ton) Value (Rand)
    Soy concentrate and TSP (<65% protein) 3168 28 million
    Soy concentrate and TSP (>65% protein) 3390 45 million
    Soy isolate 11774 242 million
    Lecithin 1698 29 million
    Total 20030 344 million

    Soy is used in a wide variety of forms for varying purposes. Figure 4 shows the share of different end user products which contain soy and its derivatives. Soy is most commonly used as a source of protein (52%) followed by uses in bakery items (15%), health supplements (11%), other markets (11%), cereals (10%) and beverages (1%).

    Soy based beverages experienced a 200% increase in demand around the turn of the century. The boom in production and consumption of these products was correlated with an increase awareness on protein health claims on products, more baby-boomers seeking longevity and good health, a rise in lactose-intolerant minorities, as well as technological improvements in processing and flavouring these products (Nguyen, et al., 2016; Food Ingredients Online, 2003).

    Figure 4: End user consumption of soy products as found in food stuffs
    (Van der Merwe, et al., 2014; Dlamini, et al., 2014)
    Figure 4: End user consumption of soy products as found in food stuffs

    Soy containing products for human consumption

    Soy derivatives such as lecithin, oil, reconstructed vegetable protein, isolate, flour and hydrogenated vegetable protein are commonly used in varying sectors of the foods industry to achieve numerous outcomes. Upon investigation of some of the most popular South African retail outlets it was found soy plays a fundamental role in food production and can be found in its various forms in every isle of the supermarket. Table 3 shows a summary of manufacturer's and the products they produce that contains soy derivatives. From this table it becomes evident that soy derivatives are used by some of the largest food manufacturing companies in South Africa to produce commonly consumed products.

    To gain broader knowledge of the origin of these products an investigation into the different forms of soy was launched. The relevant info will be discussed separately in each section below.

    Company Product Ingredient
    Fry's Meat Free Soy bean
    Eskort Polony Soy bean
    I&J Beefers Reconstructed vegetable protein, soy bean
    Knorrox Soy mince Soy bean, reconstructed vegetable protein
    Top Class Soy mince Soy bean, reconstructed vegetable protein
    Coatings Toasted soy flour
    Immana Soy mince Soy bean, reconstructed vegetable protein
    Robertsons Jikelele – Sishebo mix Hydrolysed vegetable protein
    Royco Soup mix Hydrolysed vegetable protein
    High Noon Fresh mince Soy bean
    Albany Tinkies Soy flour
    Tasty treats Waffer cookies Soy lecithin
    Simba Doritos Soy, hydrolysed vegetable protein
    Magic Moments Hard candies, toffees Hydrogenated vegetable oil, soy bean
    Nestle Klim, milk powder Soy lecithin
    Clover Baby milk powder Soy lecithin
    Bokomo ProNutro Soy bean meal
    Future Life Smart food cereal Soy bean flour, soy lecithin
    No name brand Peanut butter Hydrogenated vegetable fat, soy bean oil
    Yum Yum Peanut butter Hydrogenated vegetable fat, soy bean oil
    Wyeth Baby milk powder Soy lecithin
    Aspen Baby milk powder Soy bean oil
    Purity Baby food Soy isolate powder
    Pioneer Foods Bovril Hydrogenated vegetable protein
    Redro Fish paste Soy protein
    Nutella Chocolate spread Soy lecithin
    Cadbury Astros Soy emulsifier
    Woolworths brand Soy beans Soy bean, whole
    Soy milk Soy bean
    Sport supplements – Various companies Protein supplement Soy protein
    Soy isolate

    Imports and exports of soybeans and derivatives

    It was found that soy derivatives such as lecithin, isolate and flour are most commonly imported by private companies who place tenders out to international suppliers which complicates sourcing to origin through official channels. As seen in Figure 5 some of these import request can be very casual. Soy in various forms are sourced in this manner for use or further processing in South Africa.

    Figure 5: An example of a selling a buying site used for soy imports into South Africa
    Figure 5: An example of a selling a buying site used for soy imports into South Africa

    Exports and imports of soybeans and derivatives takes place through state (Table 4 and 5) and private sector value chains (Figure 6). Soybean oil and meal are two of the main imported products through national import channels with imports streaming from Argentina, Netherlands, Spain, Germany, Brazil and Malaysia (Table 3). Soy bean oil is most commonly exported from South Africa to Zimbabwe, Zambia and the Republic of Congo (Table 4).

    Table 4: South African exports of soybean oil (NAMC, 2018)
    Product Export value Export value Export quantity Growth value Main destinations of South African agricultural exports
    (Billion rand) (Billion rand) (Tons) (%) (Share in total South African agricultural exports)
    Year 2007 2012 2012 2007-2012
    Soybean oil 0.004 1.0 71 769 23 778 Zimbabwe (95%), Zambia (3%), Republic of Congo (2%)
    Product Import value Import value Import quantity Growth in value Share of total South African agricultural imports
    (Billion rand) (Billion rand) (Tons) (%)
    Year 2012 2007-2012 2012 2007-2012 2012
    Soybean meal 1.5 2.8 767 412 87 Argentina (100%)
    Soybean oil 0.7 1.9 167 259 171 Netherlands (39%), Spain (35%), Germany (16%), Brazil (6%), Malaysia (2%)

    Figure 6 shows the area planted, total production and demand of soy in South Africa. All of these values have experienced linear growth over the past ten years. Figure 7 shows the demand for various forms of soy. From this figure it can been seen that there is minimum to no growth in this sector. This may be due to importation of soy derivatives from other countries due to the lack of produce for processing and processing knowledge and equipment within South Africa.

    Figure 6: Area planted, total production and demand of soy in South Africa
    Figure 6: Area planted, total production and demand of soy in South Africa
    Figure 7: Demand from various sectors for soy use in South Africa
    (NAMC, 2012)
    Figure 7: Demand from various sectors for soy use in South Africa (NAMC, 2012)

    Table 6 shows the list of suppliers of soy beans imported to South Africa from various countries in 2015. This data refers to whole soy beans and not derivatives such as lecithin and isolate. Contrary to popular believe China is not the main importer of soy to South Africa. Furthermore, a 37% decrease in importation between 2011 and 2015 was seen from China. There are various factors that have an effect on imports from China, such as prices, import sanctions and supply. However recently the largest influencing factor is the significant rise in pork production in China for which soy feed is used and therefore and overall decline in soy exports from China was experienced (Share Net, 2018). Brazil is the largest importer of soy to South Africa with 120 318 tons imported in 2015 (DAFF, 2016). Long term forecasts predict that imports from Brazil to various counties may be influenced by increased import moratoriums applied by the USA (Share Net, 2018).

    Exporters Imported value 2015 (thousand US$) Share in South Africa's imports (%) Imported quantity in 2015 (tons) Unit value (US$/unit) Imported growth in value between 2011 and 2015 (% p.a.) Imported growth in quantity between 2011 and 2015 (% p.a.) Imported growth in value between 2014 and 2015 (% p.a.)
    World 71 342 100 182 151 392 253 314 36
    Brazil 47 130 66.1 120 318 392 743 632 67
    Paraguay 19 570 27.4 54 005 362
    Zambia 2 470 3.5 4 168 593 -84
    Ethopia 1 050 4.5 2 281 460
    USA 551 0.8 479 1 150 46 56 -83
    Malawi 320 0.4 660 485 258 56 -91
    United States 234 0.3 218 1 073 42 -19
    Argentina 12 10 1 200 6 28 -82
    China 3 0.1 6 500 -28 -37 -80

    The soy bean

    Among the legumes, the soybean is valued for its high (38-45%) protein content as well as its high (approximately 20%) oil content. These legumes are broadly classified as "vegetable" that can be processed into numerous secondary products (Guillon & Champ, 2002).

    Soy Lecithin

    Soy and its derivatives are widely used in a variety of sectors throughout the food industry. One of the most common forms of soy seen on ingredient lists is soy lecithin. Lecithin has long been an important component of a myriad of both food and non-food products and is one of the most versatile and valuable by-products of the oilseed industry. In foods, lecithin provides about a dozen functions, as an emulsifier, a wetting agent, for viscosity reduction, release agents, and for crystallization control and is therefore used in a wide array of products.

    Soy lecithin is a by-product of soy oil which is extracted during the degumming process. Lecithin can easily be extracted chemically using solvents such as hexane, ethanol, acetone, petroleum ether, benzene, etc., or extraction can be done mechanically. Once these process are completed lecithin is separated and dried in preparation for further use (Scholfield, 1981).

    The production of soy lecithin requires continues supply of palatable water in order to conduct the degumming process. This may be one of the reasons why soy lecithin is not commonly produced in South Africa (Ceci, et al., 2008).

    The lecithin that is imported and available for purchase on the South African market is most commonly imported by private companies who do further bulk sales to downstream companies for secondary uses. Most of these companies import various non GMO soy products, oils, textured vegetable proteins, powder, mince and chunks, flours, soy lecithin-powder, granulate and liquid. Upon request of importation origin most companies either declined to answer, did not reply or stated various sources of import depending on price and availability.

    Figure 8 shows imports of soy lecithin from India which seems to be a major supplier of soy lecithin to South Africa. The two main ports of entry for soy lecithin imports in this manner is through Durban, which is bulk imports of as much as 18 tons through the harbour whilst imports through Johannesburg (OR Tambo International Airport is smaller amounts between 10-50kg bags at a time.

    Figure 8: Main origins of order for soy lecithin imports from India
    (Info Drive India, 2018)
    Figure 8: Main origins of order for soy lecithin imports from India (Info Drive India, 2018)

    Soy oil

    Soybean oil is a vegetable oil extracted from the seeds of the soybean and is one of the most widely consumed cooking oils. To produce soybean oil, the soybeans are cracked, adjusted for moisture content, heated to between 60 and 88°C, rolled into flakes, and solvent-extracted with hexanes. The oil is then refined, blended for different applications, and sometimes hydrogenated. Soybean oils, both liquid and partially hydrogenated are sold as vegetable oil, or are ingredients in a wide variety of processed foods. Once this process is completed soy meal remains which can be used for animal feed or further processed (Gunston, 2011).

    In South Africa soy oil is most commonly imported from Spain (44%), Argentina (27%), the Netherlands (21%), Romania (4%) and Brazil (3%) (Sihlobo & Kapuya, 2015). Unlike most other countries soy oil is only used for human consumption in South Africa and not for bio fuels as is the case in some of the countries South Africa imports from (Haas, 2005).

    The South African processing capacity for sunflower and soybean crushing is estimated at 1 100 000 ton per annum of which approximately 364 000 ton was utilized for processing soybeans in recent years. The processing capacity for full fat soybeans for animal feed is estimated at 534 000 ton with an additional 33 000 ton expected in the near future. Processing capacity for high protein soybean meal for animal feed is currently 127 000 ton and it is expected to increase to 327 000 ton in the near future. Processing capacity for high protein soybean meal, for flour in human consumption is currently 104 000 ton. This directly affects and informs the capacity for soy oil production (SAGIS, 2018).

    On average, 94% of the soybean oil consumed domestically is imported (NAMC, 2018). Sunflower oil dominates the South African vegetable oil market. The South African consumer prefers sunflower oil to cook with and therefore soybean-sunflower blended oils are more common than pure soy bean oil in supermarkets (Golden Fry, 2017).

    Soy isolate

    Food-grade soy protein isolate first became available on October 2, 1959 with the dedication of Central Soy's edible soy isolate, Promine D, production facility on the Glidden Company industrial site in Chicago (Shurtleff & Aoyagi, 2008).

    Soy protein isolate is a protein that is isolated from the soybean which is made from soybean meal that has been dehulled and defatted. The soy bean meal is them further processed into three kinds of high protein commercial products: soy flour, concentrates, and isolates. Once soy reaches this stage it has most of the non-protein components, fats and carbohydrates removed. These processes result in a product that has a neutral flavour and will cause less flatulence than soy flours. This form of highly refined, or sometimes referred to as a purified form of soy protein, has a minimum protein content of 90%. Pure soy protein isolate is mainly used by the food industry to improve the texture of meat products, but are also used to increase protein content, to enhance moisture retention, and as an emulsifier. It is also available in health stores and pharmacies and commonly used in high protein sport supplements (Singh, et al., 2008).

    No substantial data could be gathered in soy isolate imports to South Africa. International studies show significant trade of soy isolate from India (Zauba, 2018).

    Soy flour/meal

    Soybean meal is used primarily as a protein source in animal feeds for the production of poultry, beef, pork, milk, butter, and eggs. A small proportion of the meal is used to make defatted soy flour, soy protein concentrates and isolates, and textured soy protein products which is used for human consumption.

    Soy flour is made by roasting the soybean, removing the coat, and grinding into a flour finely enough to pass through a 100-mesh or smaller screen. It comes in three forms: whole or full-fat (contains natural oils); defatted (oils removed) with 50% protein content and with either high water solubility or low water solubility, and lecithinated (lecithin added). As soy flour is gluten free it is commonly seen in gluten free baked products and delivers a dense product (Singh, et al., 2008).

    Figure 9 shows the import of soy meal over the past 45 years. During this time steady growth occurred up until 2009 after which a decline was experienced. The decline in imports are stated to be due to anincrease in crushing capacity within South Africa (Index Mundi, 2018). However, it is not clear of the imported meal what percentage is used for human consumption but it can be expected that the vast majority was used for animal feed in South Africa. Global figures estimate that about 2% of soy meal is further processed into soy flour for human consumption (Soya Tech, 2017).

    Figure 9: Imports of soy bean meal imported to South Africa
    (Index Mundi, 2018)
    Figure 9: Imports of soy bean meal imported to South Africa (Index Mundi, 2018)

    Processing and crushing capacity

    Ideally South Africa should do its own processing of soybeans given that the potential to do so exists. This is especially important in light of the fact that soybean meal is currently one of South Africa's largest agricultural import products. Domestic crushing capacity in South Africa mostly consists of crushing for animal feed.

    Crushing capacity of domestic produced products for the past 6 years are as follows: full fat (53% of domestically produced soybeans), meal and oil represent 37% of the domestic use of locally produced soybeans and 7% of the soybeans produced is used for human consumption. Processing capacity for high protein soybean meal for animal feed is currently 127 000 ton and it is expected to increase to 327 000 ton in the near future. Processing capacity for high protein soybean meal for human consumption is 104 000 ton currently. However these values refer to crushing capacity and not necessarily real crushing (SAGIS, 2018).

    Figure 10 shows the supply chain with the role players in the various sectors. Figure 11 shows the South African Grain Laboratories values for soy use and consumption within South Africa. As in Figure 11 the stagnant growth of soy for human consumption can be seen.

    Figure 10: In-depth analysis of the South African soybean industry
    (SAGIS, 2018)
    Figure 10: In-depth analysis of the South African soybean industry (SAGIS, 2018)
    Figure 11: Local sale of soybeans as declared by the South African Grain Laboratory
    (SAGIS, 2018)
    Figure 11: Local sale of soybeans as declared by the South African Grain Laboratory (SAGIS, 2018)

    National School Nutrition Programme (NSNP)

    All learners have the right to reach their full potential regardless of their socio-economic background. The National School Nutrition Programme, funded by a grant from National Treasury, provides one meal per day for learners attending the poorest primary and secondary schools. Secondary schools are currently being added to the programme. The Department of Basic Education, through its provincial offices, prepares weekly menus to suit the culinary and cultural diversity of learners in each province. Ingredients are carefully selected to meet the nutritional needs of growing bodies and developing minds. The daily doses of vitamins and minerals in the meals help ensure good health, reduce absenteeism and optimise learning ability. The Department of Basic Education has made aneffort in the last few years to change from a cold menu (bread) to cooked meals. This attempt resulted in the Mnandi 4 Sure recipe book (Figure 13, 14 and 15). The recipe book is aimed at improving the quality and palatability of the meals. It is primarily intended for volunteer food handlers, dedicated mothers, fathers and community members, who offer their time and skills to prepare and serve meals to learners at schools.

    Soy in the NSNP

    Texturized vegetable proteins plays a role in the South African school nutrition programme. In order to act a supplier for the NSNP stringent quality specifications has to be passed. The extensive list of these specifications can be found on the Department of Basic Education's website.

    These specifications refer to technical details on product specifications, such as size, composition and flavour as well as packaging, delivery and storage. These specifications do however not specify the origin of the soy. Most of the companies that are registered suppliers of the NSNP are based in the metropolitan areas of South Africa and more commonly Johannesburg. Upon investigation of these company's websites very little sourcing specific information could be gathered. It does however appear that these companies conduct at source milling and processing on the soy products indicating that soy is delivered to them in a fresh/whole form. Labelling details such as "Product of South Africa" on various companies supplying to the NSNP also supports this statement.

    The approved list of providers is added as an annex to this report – "Status of NSNP compliant protein processors as at April 2018".

    The NSNP menus, as developed by the various provinces, consist of a protein, starch and vegetable/fruit portion. The protein portion entails beans, soy, pilchards, peas, lentils, sour milk and chicken which are served alternating throughout the week. Menus in the different provinces are based on availability and preference. Soy mince is included at least once in a week except in the Eastern Cape where soy mince does not from part of the menu. In Gauteng soy mince meals are served twice a week.

    Figure 16 is a newspaper snippet from June 2018 where it is reported that soy is refused by school children in the NSNP. This article states that children refuse to eat on days when soya is served at lunch which leads to children not performing up to standard on those days as well as going home hungry. However, this article states that children are given curry soy mince and not the recipes as provided in the Mnandi 4 Sure recipe book.

    Figure 12: Menu example of the NSNP in Gauteng

    Gauteng Province: Secondary Schools Menu
    Days Meal plan Menu (food item) Dry portion size
    1 – Monday LUNCH
    Protein Soya Mince Stew 40g
    Starch Maize Pap 50g
    Vegetable/fruit Pumpkin/Butternut 70g
    2 – Tuesday LUNCH
    Protein Pilchard Stew 40g
    Starch Cooked Rice 40g
    Vegetable/fruit Spinach/Cabbage 70g
    3 – Wednesday LUNCH
    Protein UHT Milk 250ml
    Starch Maize Pap 50g
    Vegetable/fruit Whole Fruit 1 medium sized
    4 – Thursday LUNCH
    Protein Cooked Sugar Beans 40g
    Starch Cooked Samp 40g
    Vegetable/fruit Cabbage/Peas/Green Beans 70g
    5 – Friday LUNCH
    Protein Soya Mince Stew 40g
    Starch Maize Pap 50g
    Vegetable/fruit Pumpkin/Butternut 70g
    Figure 13: Soya mince stew recipe
    Figure 13: Soya mince stew recipe
    Figure 14: Soya mince balls recipe
    Figure 14: Soya mince balls recipe
    Figure 15: Soya burger recipe
    Figure 15: Soya burger recipe
    Figure 16: News article from 28 June on the NSNP
    Figure 16: News article from 28 June on the NSNP

    Conclusions and recommendations

    Determining exact values of import and use of soy products can be complex due to all the different chains that are used. This report attempted to explore and address as many of these chains as possible to get an idea of the usage and fiscal value of this sector.

    Even though there is no official data available, the various sources and chains explored prove that there is a thriving soy derivatives market in South Africa used by various sectors contributing to an assumed under reported estimated monetary value of R 344 million per annum. This provides an opportunity for the local soy industry to fulfil these needs that are currently met by imports of various products from all over the world.

    It is proposed that South Africa is capable of producing and crushing/processing enough soy to comply with the required amounts.

    As soy products and derivatives are increasingly being used by the food production sector consumer education is proposed to inform consumers on the versatility and uses of soy.

    The NSNP is a large consumer or soy in South Africa, however resistance to eat the soy is commonly seen in schools. It is proposed that a study be conducted on the acceptability of different soy products and flavours under school to identify better soy options in the programme.

  13. Project to determine groundnut quality at moisture levels between 7% and 10% and the foreign matter weigh loss from 18% to 7%

    Mr N Wegner,


    Traditionally groundnuts were dried on the field to levels that will relate to the acceptable maximum moisture level of 7% for shelled kernels at intake. This trials will extend the data to a 6% level). In existing guidelines, generally acceptable allowances and deductions are made for shells and foreign matter at time of delivery to selection facilities of unshelled groundnuts – thus arriving at a final quality and weight determination which is directly linked to the monetary compensation due to the producer and the expected output value of the selection facility.

    Since mechanical harvesting, bulk intakes and the need for artificial drying has been rapidly increasing over recent years, it is necessary to expand the existing standards to allow for a generally acceptable guideline for farmers and selection facilities to implement with regards to intake grading of material above 7% (and > 6%) moisture (kernel base) – indicating the levels of deductions that may reasonably be made, not only for the physical shells and foreign material, but now also for the additional moisture weight carried in these shells and foreign material.

    In additional hereto, an average scale is required to indicate a realistically expected weight adjustment in kernels that can either be made at intake of pods with a kernel moisture of more than 7% (and > 6%) or used as a measuring tool after drying to establish if reported mass loss due to moisture, fall within the expected norm. Again, these calculations are directly relevant to the compensation to the farmer, drying cost and the expected value of input product for the selection plant.

    In the interest of the industry and to foster mutual understanding and trust, such a generally acceptable and well researched standard or guideline is of critical importance. This project has been requested by industry during 2015 and is supported by the SA Groundnut Forum as well as Grain SA.

    Due to their particular knowledge of the industry and the commodity, specific expertise regarding grading in terms of the Agricultural Product Standards Act as well as impartiality and objectivity, PPECB is well positioned and equipped to render this service to the industry. Due to the nature of legislation and their specific role and mandate, there is no further value or benefit to PPECB other than of being to service to the industry.

    Aim and methodology

    To register a project that will focus on two aspects related to the Producer Intake Grading Guidelines for Groundnuts:

    • The refinement of physical grading parameters for unshelled groundnut (pods) deliveries from producers with the specific aim to expand the current guideline to include stock with moisture levels of:
      • between10% and 7% (Including to 6%);
      • 7% and below

      (Note – it is accepted in the industry that a physical grading on stock with a moisture level >10% is not practical nor accurate. Product with a moisture level higher than 10% at time of delivery will therefore have to be dried to at least 10% or lower before a physical grading parameters can be applied.)

    • A scale for physical weight loss (due to moisture loss) determination of foreign matter, shells and kernels during the drying process from 18% down to 7% (including to 6%); (at which point the physical grading can be done according to existing standards/guidelines).

    The methodology will include the following

    • To identify groundnut samples (in pods) with moisture levels of between 10%-7% (including to 6%); (kernel basis). To conduct trials at three representative selection plants which are involved in the drying of groundnut pods.

      Perform continuous physical grading tests on such samples of groundnut pods (below 10% but higher than 7% (including to 6%); during the drying process, with aiming intervals of about 0,2% moisture loss, until the moisture reaches 7% and then 6%. All these grading test results will be recorded in order to have a complete set of data of each moisture percentage point level.

      In total the number of samples to be included in the statistic data, would need to be at least 100 and inclusive of all industry variants like cultivar, harvesting methods, irrigation and dry land production and different geographic areas, in order to have sufficient available data.

      As explained earlier, the project team will also accumulate data on the moisture loss during drying – from 18% – 7% (and 6%) (kernels), and the correlating weight loss of foreign matter (sticks and parts of plant material), groundnut kernels and shells during this moisture loss period.

      The data when properly tabled as an attachment to the producer grading guidelines for groundnuts should enable the buyers and sellers of groundnut producer stock to determine a derivative grade for "wet" groundnut pods with moisture levels between 6% to 10% as well as provide an indicative weight loss deduction calculation.

    Results envisaged

    • The available data of different grading particulars at different moisture percentage levels (10-6%), will make it possible to do recommendations as to what percentage deductions should be applicable in a new producer grading guideline with regards to
      • quality deterioration as well as
      • foreign matter weight loss.
    • Data will supply an average scale that can be applied in the case of expected moisture/mass loss due to drying product from levels from 18% down to 6% – either as an agreed deduction at intake or as a control after drying.

    Value for the industry

    • The value of this information and recommended new producer guidelines would be that it would be possible to do a producer grading on a consignment where the moisture percentage is still not regarded as dry. (Above 6% but not higher than 10%) So with the necessary recommended deductions to be made, it would become possible to determine the grade by simply doing the necessary deductions from a table according to the measured moisture.
    • Weight related deductions on foreign material will also be possible from the table indicating the weight loss deductions from for pods delivered between 18% to 6%.


    Limitations could be the number of available lots where the moisture levels are not yet at the respective 18% and 10% levels required for the project. It could be necessary to travel distances between selection plants, in order to find suitable moisture levels to do the necessary tests. The project success will also require the cooperation of at least 3 selection plants who are involved in drying of groundnut pods. It is also possible that the data of one season would not be sufficient to provide accurate information.


    These grading results will be made available as statistic tables that will be included in the Groundnut Producer Grading Guidelines. These tables will then be used to determine the applicable deductions that are to be made when grading groundnut pods with moisture levels above 7%, but the above 6% will also be included as part of the new table.

    The published tables would be made available to industry through the distribution channels of the Groundnut Forum and members.