The challenge of food security requires that agricultural production is no longer based on a narrow genetic material present in conventional crops. Whereas conventional crops have been genetically improved to suit management practices of the modern farmer, the future farmer requires that there be access to a wide variety of genetic material for economic exploitation and to respond to the challenges of climate change in a sustainable fashion. This study was designed to learn about production of wild water melon [Citrullus lanatus (Thunb.) Matsumura and Nakai] from seed germination, seedling establishment and field crop production. The specific objectives of the study were (a) to determine the effect of water stress on three landraces of watermelon differing in seed colour and provenance during seedling establishment, (b) to determine the effect of planting date on crop growth and yield under field conditions, and (c) to relate proline accumulation to water stress in wild watermelon. Three seedlots, ‘B’, ‘DB’ and ‘VDB” were derived from seeds collected from subsistence farming communities of the Eastern Cape, and KwaZulu-Natal. Following one season of seed production in Pietermaritzburg, KwaZulu-Natal, seeds were tested for germination capacity, before seedlot response to water stress was determined in three substrates made of pine bark, a 1:1 mixture of fine sand and pine bark and fine sand only. The substrates were kept at 75% FC, 50% F.C and 25% F.C., to create varying levels of water regimes during 12 weeks of seedling growth in a glasshouse (16/21oC (day/night) and 60% RH). Leaf proline content was determined at seedling harvest. Crop production under field conditions occurred at one site with three planting dates late September 2008, November 2008 and January 2009, respectively. There were significant differences among seedlots with respect to seed quality and seedling yield, which consistently showed that B > VDB > DB. The differences in seedlots continued in the same order even in response to field conditions. Wild watermelon was responsive to water stress during seedling growth, but high water regimes compromised water use efficiency. Proline accumulation correlated with water stress. The best plant growth and yield under field conditions was obtained when planting occurred in September, followed by November and January plantings, respectively. Early planting was also associated with high crop growth rate and larger fruit size. It is concluded that despite being a desert crop, wild watermelon responds to water deficits during seedling growth. Results of field studies cannot be conclusively used to determine crop response to water stress, although they gave a good indication of crop response to different conditions of rainfall and temperature at the study site from September to March. / Thesis (M.Sc.Agric.)-University of KwaZulu-Natal, Pietermaritzburg, 2010.

Local maize landraces have evolved over hundreds of years of natural and farmer selection under varying conditions. These landraces may have developed tolerance to abiotic stresses such as water deficits during this cycle of selection. However, despite its continued existence and importance, little is known on their agronomy and responses to water stress. If indeed landraces have developed tolerance to water stress, they may prove a key genetic resource for future crop improvement in light of increasing water scarcity. The primary objective of this study was to evaluate the responses of a local maize landrace to water stress at different stages of growth in comparison to two known commercial hybrids, SC701 and SR52. Seed from a local maize landrace was multiplied and characterised according to kernel colour. Two distinct colours were selected for the purposes of this study, white (Land A) and dark red (Land B). In a holistic approach, the thesis consisted of four separate studies whose overall objective was to evaluate the responses of the maize landraces to water stress at different growth stages, up to and including yield and its components. These comprised three controlled environment studies (25°C; 60% RH) and a field trial. For the controlled environment, two water regimes were used, 25% field capacity (FC) (stress treatment) and 75% FC (non-stress). The first study investigated the effect of water stress on early establishment performance. Seed quality was evaluated using the standard germination test together with electrolyte leakage. Catalase activity and accumulation of proline were examined as seedling physiological response to water stress. The second study was conducted as a pot trial to investigate the effect of water stress on growth, photosynthesis and yield. Photosynthesis was measured as chlorophyll fluorescence (CF). In addition, a field study over three planting dates was conducted at Ukulinga Research Farm in Pietermaritzburg, under dryland conditions, during the period from August 2008 to June 2009. The objective was to evaluate the effect of planting dates and changing soil water content on growth, yield and yield components. Three planting dates were used, representative of early (28 August 2008), optimum (21 October 2008) and late planting (9 January 2009). Lastly, a study on hydro-priming was conducted, necessitated by observations made primarily in the first study. The study was carried out under controlled environment conditions. The objective was to evaluate whether hydropriming can improve germination, vigour and emergence under water stress. Seeds were soaked in water for 0 hours (Un-primed or control), 12 hours (P12) and 24 hours (P24). Results from the first study showed that maize landraces were slower to germinate and emerge, and produced less vigorous seedlings compared to the hybrids. The study showed that hybrids were more superior under optimum (75% FC) conditions than under stress conditions (25% FC). Physiological showed that both hybrids and landraces expressed catalase under water stress, with landraces showing slightly better expression compared to the hybrids. Proline accumulation was observed in both hybrids and landraces as a response to water stress, with hybrids being more sensitive to water stress. In the pot trial, results showed that the vegetative stage of both hybrids and landraces was less sensitive to water stress than the reproductive stage. Results showed no differences between field capacities, with respect to emergence, mean emergence time, leaf number, CF, ear prolificacy and ear length. Photosynthesis, as measured by CF, was shown to be desiccation tolerant. Water stress had a negative effect on cob mass, lines per cob, grains per cob and total grain mass, and resulted in barrenness in the landraces. The hybrids had superior yield compared to the landraces. Results for the field trials showed that planting date had highly significant effects on emergence, plant height, leaf number and days to tasseling (DTT). Landraces emerged better than hybrids in all plantings; highest emergence was in the early and late plantings. Optimum and late planting resulted in maximum plant height and leaf number, respectively, compared to early planting. Hybrids were superior, growing taller and with more leaves than landraces in all plantings. DTT decreased with successive plantings. Planting date had an effect on ear prolificacy (EP), kernels/ear (KNE) and 100 grain mass. Planting date had no effect on ear length and mass, kernel rows/cob, grain mass and yield. With the exception of EP, hybrids out-yielded the landraces in all three planting dates. Hydro-priming landraces for 12 hours and 24 hours, respectively, improved germination velocity index, reduced mean germination time and improved emergence and mean emergence time of maize landraces under water stress. Performance of hybrid seeds remained superior to that of landraces even after seed treatment to improve germination and vigour. Landraces were slower to germinate and emerge and produced less vigorous seedlings in controlled conditions only. Both hybrids and landraces expressed catalase activity and also accumulated proline in response to water stress, although hybrids were more sensitive to stress in the establishment phase. Results confirmed literature, showing that, for both hybrids and landraces, the vegetative stage is less sensitive to stress than the reproductive stage. Hybrids produced superior yields compared to landraces in both controlled environment and field conditions. However, the pattern of seedling establishment observed in the initial controlled environment study for hybrids and landraces was reversed in the field study. Lastly, hydro-priming is of some benefit to maize establishment. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2009.