Thesis (PhD(Agric))--Stellenbosch University, 2012. / ENGLISH ABSTRACT: A lack of defects is required for successful table grape marketing, which pre-suppose optimal vine performance, berry development and post-harvest quality. The supply of mineral nutrients affects vine development, physiology and berry quality. Despite a vast amount of research conducted over decades, there remain many unresolved issues regarding table grape vine nutrition to ensure optimal table grape quality and shelve-life. Unjustified fertilisation practices often include excessive applications of nitrogen (N), potassium (K) and calcium (Ca). A four-year field trial was therefore conducted on a sandy soil in the Paarl district of South Africa, using grafted on Ramsey, and trained to a gable trellis system. Nitrogen, potassium and calcium were applied, singular or in combination, at rates up to 300% the calculated annual nutritional requirement. The effect of these excessive applications on table grape performance under typical South African cultivation conditions was investigated for Vitis vinifera L. cv. Prime Seedless, a very early seedless table cultivar that is produced with minimum berry diameter of 18mm, with special reference to 1) vegetative growth, 2) expression of grapevine nutrient availability through foliar analyses, 3) berry nutrient accumulation patterns of this early cultivar, 4) manipulation of berry nutrient content through soil and bunch directed applications and 5) the effect of berry nutrient content on its quality. No definite vegetative growth responses (expressed as shoot length, leaf surface area and shoot mass) and leaf chlorophyll content differences were obtained for all the treatments. These results were obtained in a vineyard on a sandy soil where excessive N fertilisation caused a reduction of soil pH to detrimentally low levels and where the excessive N, K and Ca applications reduced mutual concentrations and that of Mg, in the soil. A lack of stimulation in vegetative growth may therefore be ascribed to the combined negative effect of these excessive applications on soil pH and vine nutrition. Although the N content of petioles was higher for treatments where N was applied, consistent significant increases in petiole N with N fertilisation were not observed. Petiole N concentration showed a decreasing trend throughout the season. Petiole K concentrations were significantly increased by the K fertilisation at all phenological stages. None of the K fertilisation treatments, however, succeeded to raise petiole K concentrations above the accepted maximum norms and petiole K concentration at a specific sampling stage varied significantly between the four seasons. A general decrease in petiole K concentration was found for all seasons. Calcium fertilisation did not increase soil Ca content, resulting in a lack of differences in petiole Ca concentrations between treatments. An increase in petiole Ca concentration towards harvest was obtained. Correlations between petiole nutrient concentration and berry mineral content at harvest were poor. The only way of knowing the mineral content of berries would seem to be by measuring it directly instead of deducing it from the results of leaf or petiole analyses. The dynamics of berry growth impacted on berry nutrient concentration. Early rapid berry growth, predominantly due to cell division and cell growth, was associated with the most rapid decreases in N, P and Ca concentration. Due to mobility of K and Mg in the plant, that exceeds other nutrients, the decrease in concentration of these two mineral elements was not as pronounced as that of the others. Nutrient accumulation was most rapid during the pre-véraison period, but only Ca showed a definite termination during the early ripening period. The continued inflow of N, P, K and Mg, albeit at slower rates immediately after véraison, should be taken into consideration when fertilisation is applied. As a table grape, total accumulation of each nutrient in Prime Seedless berries also far exceeded that of other cultivars studied thus far. A particular difference is that the berry flesh:skin ratio is much higher than that of previously studied cultivars, leading to higher levels of nutrient accumulation in the flesh. Slightly larger berry size was obtained for N applications and is ascribed to slight increases in early vegetative growth, allowing a better response to GA3 treatments. The use of GA3 for berry enlargement is also considered the reason why K fertilisation, resulting in increased berry K levels, did not affect berry size, as is often found for wine grapes. Higher available NO3 - in the soil on account of excessive N applications resulted in higher levels of berry N, despite sub-optimal soil pH regimes that were created by these treatments. Berry K concentration and content were increased by K fertilisation. Rapid vine K uptake and translocation to the berries seem to negate the reduced vine nutritional status as observed in petioles for situations of over-fertilisation with N. Berry Ca levels were not increased by Ca fertilisation or by bunch applied Ca. The rapid rates of berry growth, together with low rates of berry Ca uptake and Ca uptake that terminates at the onset of ripening, are assumed to be the main reasons for this result. Low levels of decay as well as a lack of consistently increased decay were obtained for N containing treatments. Nitrogen levels in the berries above which their susceptibility to fungal infection is increased, should be established. Information on specific N compounds that may lead to more susceptibility is required. Potentially increased berry browning on account of high rates of K fertilisation needs to be further investigated; indications that this may occur were observed. Neither soil applied Ca nor bunch applied Ca improved berry quality, although Ca treatments seemed to reduce decay during the only season that significant differences were obtained. The negative effect of excessive fertilisation on soil chemistry of sandy soils has again been highlighted by this study. This annuls the fertilisation, leading to inefficient fertilisation and a lack of the desired responses. As indicator of vine nutrient availability, petiole analysis, was proven unreliable and should be evaluated in parallel with soil analyses, taking seasonal variation into consideration. The danger of being only guided by published norms for leaf nutrient concentrations when establishing fertilisation practices has again been highlighted by this study. This research indicated that for a very early cultivar like Prime Seedless, nutrient accumulation dynamics can already start to change during the pre-véraison period in some seasons. This is due to different edaphic and climatic conditions as well as berry size, which leads to much higher flesh:skin ratios. Future research on table grapes would need to develop an understanding of the various factors and dynamics that determine berry nutrient concentration and accumulation of early ripening, large berry sized, seedless table grape cultivars. / AFRIKAANSE OPSOMMING: Suksesvolle bemarking van tafeldruiwe is ten nouste afhanklik van die beskikbaarheid van druiwe sonder defekte, wat ‘n direkte verband met optimale wingerdprestasie, korrelontwikkeling en na-oes kwaliteit inhou. Voorsiening van minerale voedingstowwe beïnvloed die stok se groei, fisiologie en korrelgehalte. Ten spyte van ‘n oorweldigende hoeveelheid navorsing wat oor dekades reeds gedoen is, is daar steeds onopgeloste kwessies aangaande bemesting van tafeldruiwe vir optimale druifgehalte en houvermoë. Die gevolg is onoordeelkundige bemestingspraktyke wat o.a. aanleiding gee tot oorbemesting met stikstof (N), kalium (K) en kalsium (Ca). ‘n Vier-jaar-lange veldproef is gevolglik op ‘n sandgrond in die Paarl distrik (Suid-Afrika) onderneem deur gebruik te maak van Vitis vinifera L. cv. Prime Seedless geënt op Ramsey en op ‘n dubbel-gewel prieelstelsel opgelei is. Stikstof, K en Ca is alleen, of in kombinasie, toegedien teen hoeveelhede gelykstaande aan 300% van die wingerd se jaarlikse behoefte. Die effek van hierdie oormatige toedienings op tafeldruif prestasie onder Suid-Afrikaanse verbouingstoestande is ondersoek, met spesiale verwysing na 1) vegetatiewe groei, 2) uitdrukking van voedingstofbeskikbaarheid deur blaarontledings, 3) die voedingstof akkumulasie patrone van korrels van hierdie vroeë kultivar, 4) manipulasie van korrel voedingstofinhoud deur grond en trosgerigte toedienings en 5) die effek van korrel voedingstofinhoud op kwaliteit. Die doel van die proef was om bemestinspraktyke van Prime Seedless, ‘n baie vroeë pitlose tafeldruifkultivar met ‘n minimum korrelgrootte van 18 mm, te verfyn. Deur die akkumulasie patrone van die druiwe uit te klaar is daar ook ondersoek ingestel of oestyd en na-oes gehalte deur oormatige toediening van voedingstowwe affekteer word. Geen duidelike verskille betreffende vegetatiewe groeireaksies (uitgedruk as lootlengte, blaaroppervlaktes en lootmassas) asook verskille in blaar chlorofilinhoud is vir die behandelings verkry nie. Hierdie resultate is verkry in ‘n wingerd op ‘n sandgrond, waar oormatige N-bemesting aanleiding gegee het tot grond pH verlagings tot die peil van nadelige vlakke. Verder het die oormatige N, K en Ca toedienings wederkerige verlagings in konsentrasies, asook op dié van Mg, in die grond teweeggebring. Die tekort aan vegetatiewe groeiresponse op die behandelings kon dus toegeskryf word aan ‘n gekombineerde effek van die oormatige toedienings op grond pH en voedingstofbalanse. Hoewel die N-inhoud van bladstele hoër was vir behandelings wat N toediening ingesluit het, was daar nie konstante toenames in die vlakke verkry nie. Bladskyf N-konsentrasie het afgeneem deur die loop van die groeiseisoen. Vir alle fenologiese stadiums was bladskyf Kkonsentrasies betekenisvol verhoog deur K-bemesting. Nie een van die Kbemestingsbehandelings het egter daarin geslaag om bladskyf K inhoud vir enige monstertyd bo die algemeen aanvaarde maksimum norms te lig nie. Verder het bladskyf K inhoud by ‘n spesifieke fenologiese stadium ook betekenisvol tussen seisoene verskil. Die K-inhoud van bladskywe het afgeneem met verloop van die seisoen. Kalsiumbemesting het nie die grond se Ca inhoud deurgans verhoog nie, wat dus die tekort aan verskille in Ca konsentrasies tussen die behandelings verklaar. ‘n Toename in Ca konsentrasie en korrel Ca inhoud is vanaf set tot oes waargeneem. Swak korrelasies tussen bladskywe se voedingstofinhoude en korrels se voedingstofinhoude is verkry. Die enigste manier waarop korrels se voedingstofinhoude dus afgelei kan word, blyk te wees deur direkte bepaling daarvan. Voedingstofinhoude van korrels is deur groeipatrone daarvan beïnvloed. Vroeë korrelgroei, hoofsaaklik a.g.v. seldeling en selgroei, het met die vinnigste afnametempo van N, P en Ca gepaard gegaan. As gevolg van die hoër beweeglikheid van K en Mg in die plant in vergelyking met ander voedingstowwe, was die afname in konsentrasie van hierdie twee elemente nie so groot soos vir die ander nie. Voedingstofakkumulasie was die vinnigste in die periode voor deurslaan. Slegs Ca het ‘n beeïndiging van opname aan die einde van hierdie periode getoon. Die voortgesette opname van N, P, K en Mg, alhoewel stadiger kort na deurslaan, moet in ag geneem word wanneer bemesting toegedien word. Vir hierdie kultivar het die totale opname van elke bemestingstof dié van die ander kultivars wat tot hede bestudeer is, ver oorskry. ‘n Spesifieke verskil is ‘n baie hoër vleis:dop verhouding as wat vir ander kultivars verkry is. Dit gee aanleiding tot baie hoër vlakke van voedingstofakkumulasie in die vleis. Effens groter korrelgroottes is verkry waar N toedienings gemaak is. Dit word toegeskryf aan klein toenames in vroeë vegetatiewe groei, wat dus beter reaksie op GA3 behandelings tot gevolg gehad het. Die gebruik van GA3 vir korrelvergroting word ook beskou as die rede waarom K-bemesting, wat tot hoër vlakke van K in die korrels aanleiding gegee het, nie korrelgrootte, soos by wyndruiwe, bevorder het nie. Hoër NO3 - in die grond (water), na aanleiding van N toedienings, het aanleiding gegee tot hoer vlakke van N in die korrels. Dit het plaasgevind ten spyte van sub-optimale grond pH wat deur die oormatige N toedienings veroorsaak is. Korrel K konsentrasie en -inhoud is deur K-bemesting verhoog. Vinnige opname en translokasie van K na die korrels het ook geblyk die rede te wees waarom die verlaagde voedingstatus van die stokke a.g.v. oorbemesting met N nie die korrels se K inhoud geaffekteer het nie. Die vinnige groeitempo van die korrels, tesame met lae vlakke van Ca opname, asook korrels se Ca opname wat tydens rypwording ophou, word as die redes vir die tekorte aan behandelingseffekte beskou. Lae vlakke van bederf, asook ‘n tekort aan betroubare tendense dat bederf deur Nbemesting verhoog word, is verkry. Daar moet vasgestel word of daar N vlakke in die korrels is waarbo hul vatbaarheid vir swaminfeksies verhoog word, en of daar spesifieke N verbindings is wat die korrels meer vatbaar maak vir bederf. Indikasies dat K-bemesting interne verbruiningsvlakke verhoog het, regverdig verdere ondersoek. Korrelkwaliteit is nie deur grond- of trosgerigte toedienings bevoordeel nie. Die negatiewe effek van oormatige bemesting op die chemiese samestelling van sandgronde is weer deur hierdie navorsing uitgelig. Dit lei tot oneffektiewe bemesting en ‘n tekort aan die verlangde effekte. Blaarontledings blyk onbetroubaar te wees as aanduiding van voedingstof beskikbaarheid. Dit moet evalueer word saam met grondontledings en ook seisoenale variasie in ag neem. Die gevaar om slegs deur gepubliseerde norme gelei te word wanneer bemestingspraktyke bepaal word, is weer deur hierdie navorsing uitgelig. Voorst is daar in hierdie navorsing gevind dat voedingstof akkumulasiepatrone van ‘n baie vroeë kultivar soos Prime Seedless alreeds voor deurslaan begin verander a.g.v. omgewingstoestande en korrelgroei wat tot ‘n veel hoër vleis:dop verhouding aanleiding gee. Toekomstige navorsing op tafeldruiwe behoort die faktore en dinamika wat voedingstofkonsentrasie en -akkumulasie in korrels van vroeë, groot korrel, pitlose tafeldruifkultivars beïnvloed verder te ondersoek.