Thesis (Ph. D. (Plant Production)) -- University of Limpopo, 2015 / Wild watermelon (Cucumis africanus LF.) and wild cucumber (Cucumis myriocarpus Naude.) are known for their ethnomedicine, ethnopesticide, ethnonematicide and nutritional properties, along with nematode resistance. The two Cucumis species were successfully used as inter-generic seedling rootstocks for watermelon (Citrullus lanatus Thunb.) cultivars, where nematode-resistant genotypes are not available. Also, the two Cucumis species are hardy and resilient to inland South Africa conditions, where temperatures are predicted to increase by 6°C in the year 2030. Seeds in the Cucurbitaceae Family contain high concentration of cucurbitacins, which induce auto-allelopathy that inherently inhibits plant growth and germination. Poor germination and non-uniform stands as a result of seed dormancy are a major challenge in sexual propagation of wild Cucumis species for various potential industries. Generally, true-to-type, uniform and disease-free plants in plant production are asexually-generated through in vitro propagation techniques. This study was therefore, initiated to address seed dormancy and related challenges of sexual propagation in the two wild Cucumis species by determining whether: (1) seed dormancy in C. africanus and C. myiocarpus would be ameliorated to allow for in vitro sexual propagation to establish pathogen-free parent stock, (2) the testa in C. africanus and C. myiocarpus seeds would possess structures, which interfere with imbibition and movement of water to the endosperm, (3) all organs of C. africanus and C. myriocarpus would be suitable for in vitro propagation, (4) suitable potting medium for in vitro propagated plantlets of C. africanus and C. myriocarpus would be available for acclimatisation of plantlets and (5) in vitro-produced xxviii plantlets from nematode-resistant C. africanus and C. myriocarpus would retain their resistance to Meloidogyne incognita race 2 under greenhouse conditions. In vitro and ex vitro experiments were conducted to achieve the stated objectives, with treatments in the laboratory and the greenhouse being arranged in completely randomised and randomised complete block designs, respectively. Validity was primarily ensured through the use of factorial trials, while the reliability of data was ensured by using appropriate levels of statistical significance. Leaching alone in C. africanus improved germination, while in C. myriocarpus this treatment had no effect on germination. The optimum leaching time in leached-control seeds of C. africanus was achieved at 7.1 h, with a 25-day mean germination time (MGT) and 52% optimum germination percentage (GP). In the two Cucumis species, the combined effect of leaching seeds in running tapwater and physical scarification of seeds at the chalaza region escalated germination in both Cucumis species, suggesting that both chemical and physical seed dormancies were involved. In C. africanus, cucurbitacin B (C32H48O8) was deposited exogenously to the testa, whereas in C. myriocarpus cucurbitacin A [cucumin (C27H4009) and leptodermin (C27H3808)], was deposited endogenously to the testa. The optimum leaching time in leached-scarified (LS) seeds of C. africanus was achieved at 5.7 h, with at least 40-day MGT and 89% optimum GP. In contrast, in C. myriocarpus LS seeds had the optimum leaching time of 6.3 h, with at least 41 days MGT and 93% optimum GP. Field emission SEM confirmed that there were two “water-gaps”, one at the micropylar region (hilum end) and the other at chalaza region (abaxial end) of seeds in both Cucumis species. Five distinct testa layers in seeds of C. myriocarpus were observed, namely, (i) epidermis, (ii) hypodermis, (iii) sclerenchyma, (iv) aerenchyma xxix and (v) chlorenchyma. In contrast, C. africanus seeds did not have the hypodermis between the micropylar and chalaza regions, but was present around both regions, which may provide some explanation of sporadic germination in non-leached and non-scarified seeds in this Cucumis species. The most suitable plant propagules for in vitro mass propagation of the two Cucumis species were nodal and apical buds. The optimum PGRs for shoot regeneration using both propagules in C. africanus and C. myriocarpus were at 0.80 and 0.35 μM 6-benzyladeninepurine (BAP), respectively. In contrast, the largest number of roots was regenerated at 0.31 and 0.44 μM indole-3-butyric acid (IBA) for C. africanus and C. myriocarpus, respectively. In vitro-produced plantlets were successfully acclimatised to ex vitro conditions, with sand + compost potting medium being the most suitable growing medium for weaning both Cucumis species. The in vitro-produced plantlets retained their resistance to M. incognita race 2. In conclusion, seeds of C. africanus and C. myriocarpus are structurally and chemically different, with strong evidence of chemical and physical dormancies. Structurally, C. myriocarpus seeds consist of five layers, four lignified and one non-lignified, whereas those of C. africanus have four layers, three lignified and one non-lignified. Evidence suggested that in C. africanus seeds, allelochemicals were primarily deposited outside the testa, whereas in C. myriocarpus they were deposited within the testa. The identified seed dormancies could successfully be ameliorated through combining leaching and scarification in both Cucumis species. The developed in vitro propagation protocols accord the two Cucumis species the potential for use as future crops in the context of climate-smart agriculture and research. / Flemish Interuniversity Council (VLIR)

Thesis (M.Sc. (Plant Production)) -- University of Limpopo, 2013 / Two phytonematicides were researched and developed from fermented crude extracts of wild watermelon (Cucumis africanus) and wild cucumber (Cucumis myriocarpus) fruits for use as alternatives to methyl bromide in managing root-knot (Meloidogyne species) nematodes in tomato (Solanum lycopersicum) production. Fruits of C. africanus contain cucurbitacin B (C32H48O8), while those of C. myriocarpus contain cucurbitacin A, which comprises cucumin (C27H40O9) and leptodermin (C27H38O8). Phytonematicides from C. africanus and C. myriocarpus fruits are referred to as nemafric-B and nemarioc-A, respectively. The two phytonematicides, due to their origin from plant species with allelochemicals, have high potential of being phytotoxic to crops. The use of the Curve-fitting Allelochemical Response Dosage (CARD) computer-based model assisted in the establishment of concentrations which were stimulatory to growth of tomato (Solanum lycopersicum) plants, while exhibiting nematoxic properties to Meloidogyne species. The two phytonematicides were developed from crude extracts of fruits dried at 52˚C in air-forced ovens and ground in a Wiley mill through 1-mm-opening sieves. However, equipment for drying and grinding fruits would not be accessible to smallholder farmers who wished to prepare their own products on-farm. The objective of this study therefore, was to determine whether nemafric-BL and nemarioc-AL produced from fresh fruit of the two Cucumis species would be suitable for use (i.e. non phytotoxic) in tomato production for managing population densities of M. incognita race 2. In order to distinguish the products of fresh (F) fruits from those of dried (D) fruits, they were code-named nemafricF-BL or nemariocF-BL and nemafricD-BL or nemariocD AL, respectively, where G and L denoted granular and liquid formulations, respectively. Tomato cv. ‘Floradade’ seedlings were infested with 3 000 eggs and second-stage xv juveniles of M. incognita race 2. An equivalent of 40 g and 80 g dried fruit mass of nemafric-B and nemarioc-A, namely, 284 g and 411 g fresh fruit mass for nemafric-B and nemarioc-A, respectively, were separately fermented using EMROSA effective micro-organisms mixed with 16 L chlorine-free tapwater in 20 L container for 14 days at ± 25˚C, allowing pH to gradually decline to ± 3.7. Separate experiments for each product run concurrently. Treatments, namely, 0, 2, 4, 8, 16, 32 and 64% concentrations, where for instance, 2% = 20 ml/1000 ml x 100, were arranged in a randomised complete block design, with 10 replications. Blocking in the greenhouse was done for wind direction which was regularly erected by fans for cooling down the greenhouse. At 56 days after weekly application of each treatment, flower number, fruit number, dry shoot mass, dry root mass, dry fruit mass, plant height, stem diameter and nematode numbers were each subjected to analysis of variance. Nematode data were, prior to analysis, transformed using log10(x + 1), but untransformed data were reported. Using the sum of squares, nemafric-BL and nemarioc-AL treatments affected dry root mass, dry shoot mass, flowers number, fruit number, plant height and stem diameter. Nemafric-BL contributed 67%, 78%, 58%, 43%, 60% and 26%, while nemarioc-AL contributed 71%, 61%, 19%, 35%, 34% and 24% to total treatment variation of the six respective variables. Plant variables with significant (P ≤ 0.05) treatment effects were further subjected to the CARD model to generate seven biological indices, with three distinct phases, namely, stimulation, neutral and inhibition phases. Using the quantified stimulation phase, the mean concentration stimulation range (MCSR) was computed for each variable using two biological indices, namely, threshold stimulation point (Dm) and saturation point (Rh). The CARD model explained 98%, 99%, 98% and 98% of the quadratic models of dry root mass, dry shoot mass, plant height and stem diameter, xvi respectively, against increasing concentrations of nemarioc-AL. Similarly, the CARD model explained 99%, 96%, 84% and 93% of total treatment variation in the respective plant variables. The integrated MCSR [MSCR = Dm + (Rh/2)] for nemafric-BL on tomato plants was 7%, while that for nemarioc-AL was 4%. In the CARD model, the overall sensitivities (∑k) of tomato plants exposed to nemafric-BL and nemarioc-AL were 3 units and 5 units, respectively. Tomato plants were therefore, less sensitive to nemarioc-AL since it had higher ∑k value than nemafric-BL. At 4% nemarioc-AL and at 7% nemafric-BL, the two phytonematicides were each highly suppressive to population densities of M. incognita race 2. In conclusion, on the basis of non-phytotoxicity of the computed MCSR values and their suppressive effects on population densities of M. incognita race 2, the smallholder farmers could produce nemafric-BL and nemarioc-AL phytonematicides on-farm. However, the production of the two products from fresh fruits would not be sustainable since fruits of the two Cucumis species are highly seasonal due to the high incidence of post-harvest decays. / The Land Bank Chair of Agriculture – University of Limpopo, Limpopo Agro-processing Technology Station,and the Flemish Interuniversity Council of Belgium