L’efficience d’utilisation de l’eau (WUE), rapport entre la biomasse produite et l’eau consommée, est un trait essentiel à étudier en agronomie pour améliorer la production des cultures soumises à la sécheresse. Cependant, mesurer l’eau consommée par un couvert végétal est difficile à réaliser. L’objectif général de ce travail de thèse est de répondre à trois principales questions : (i) peut-on déterminer WUE en utilisant la discrimination isotopique du carbone (CID) facile à mesurer? (ii) comment l’analyse de la variabilité de WUE et CID peut contribuer à sélectionner des génotypes de tournesol soumis à la sécheresse? (iii) et les variations de WUE peuvent-elles être révélées par les variations de relations hydriques? Quatre expériences ont été conduites en serre pendant deux années : (i) avec deux scenario de sécheresse, l’une progressive, l’autre stable, et (ii) avec cinq niveaux de contenu en eau du sol stables. Les principaux traits mesurés sont WUE, CID et d’autres traits représentant les relations hydriques tels que le contrôle de la transpiration (FTSWt), la capacité d’extraction de l’eau (TTSW) et la tolérance à la déshydratation (OA). Une forte corrélation négative a été mise en évidence entre WUE et CID. Une large variabilité a également été observée pour FTSWt, TTSW et OA. Ces résultats permettent de connaitre le contrôle génétique de WUE et CID, ainsi que des traits associés, lesquels n’ont jamais été relatés dans la littérature. De plus, l’analyse QTL pour FTSWt n’a pas non plus été réalisée, chez aucune plante. Des QTL pour WUE et CID ont été identifiés pour différents scenario de sécheresse. Les QTL pour CID sont considérés comme ‘constitutifs’, parce qu’ils sont détectés dans les différents scenarios. Les QTL pour CID co-localisent avec ceux pour WUE, pour la biomasse et pour la transpiration cumulée. Des co-localisations de QTL ont également été observées entre FTSWt et TTSW, entre TTSW et WUE-CID-Biomasse, et entre FTSWt-TTSW et Biomasse. Cette étude met en évidence que WUE est physiologiquement et génétiquement associée à CID. De plus, CID représente un excellent substitue à la mesure de WUE et permet d’améliorer l’efficience d’utilisation de l’eau par sélection assistée par marqueurs. / Water use efficiency (WUE), measured as the ratio of plant biomass to water consumption, is an essential agronomical trait for enhancing crop production under drought. Measuring water consumption is logistically difficult, especially in field conditions. The general objective of the present Thesis is to respond to three main questions: (i) can WUE be determined by using carbon isotope discrimination (CID), easy to measure?, (ii) how WUE and CID variation analysis can contribute to the genotypic selection of sunflower subjected to drought?, and (iii) can WUE variation be revealed by the variation of plant-water relation traits. Four experiments were carried out in greenhouse across two different years: (i) on two drought scenarios, progressive soil drying and stable water-stress, and (ii) on five levels of soil water content. The main traits that have been measured include WUE, CID, as well as plant-water relation traits, i.e. control of transpiration (FTSWt), water extraction capacity (TTSW), and dehydration tolerance (OA). A highly significant negative correlation was observed between WUE and CID, and a wide phenotypic variability was observed for both WUE and CID. A wide variability was also observed for FTSWt, TTSW and OA. The results provide new insight into the genetic control of WUE and CID related-traits, which, unlike to other crops, genetic control of WUE, CID, and TTSW in sunflower have never been reported in the literature. Further, quantitative trait loci (QTL) mapping for FTSWt was never reported in any plant species. The QTL for WUE and CID were identified across different drought scenarios. The QTL for CID is considered as a ‘‘constitutive’’ QTL, because it is consistently detected across different drought scenarios. The QTL for CID co-localized with the QTL for WUE, biomass and cumulative water transpired. Co-localization was also observed between the QTL for FTSWt and TTSW, between the QTL for TTSW and WUE-CID-biomass, as well as between the QTL for FTSWt-TTSW and biomass. This study highlights that WUE is physiologically and genetically associated with CID. CID is an excellent surrogate for WUE measurement, and can be used to improve WUE by using marker-assisted selection (MAS) to achieve the ultimate goal of plant breeding at genomic level.

Thesis (MScAgric (Viticulture and Oenology))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: Within-vineyard variability in vigour and water status commonly occurs in South African vineyards. Different soil types found over short distances are probably the main cause of vigour variability, while differences in grapevine water status are commonly induced by lateral water flow in the vineyard, blocked irrigation emitters and differences in soil water-holding capacity. These factors can cause heterogeneous ripening and differences in fruit quality between different parts of the vineyard, an aspect that needs to be avoided as far as possible in order to produce quality wines. Measurements of carbon isotope discrimination (CID) have proved to be a tool to assess grapevine physiology in order to study the effects of environmental parameters on leaf carbon dioxide (CO2) gas exchange and stomatal conductance (gs). Grapevine water deficit stress/strain in reaction to these environmental conditions can then be determined by observing the amount of 13C absorbed by plant material after discrimination of 13C has taken place, and this is influenced by the grapevine stress condition and can indicate water-use efficiency. In this study, the variability of grapevine water status and vigour was determined in order to quantify these parameters in different parts of the vineyard. Two separate trials were conducted, the first at Wellington, South Africa, where different irrigation regimes resulted in variability in grapevine water status between plots. The second trial was at Stellenbosch, South Africa, where plots were divided among different vigour classes and irrigation was applied in different quantities for different irrigation treatments. Within-vineyard variability in water status (Wellington and Stellenbosch) and vigour (Stellenbosch) were then quantified and the effects on some grapevine physiological parameters and berry composition were measured. The treatments in the Wellington trial led to differences in grapevine water status, which could be quantified by measurements of stem water potential (SWP) and leaf water potential (LWP). Soil variability also led to differences in grapevine vigour, which were quantified by measurements of pruning mass, leaf area and shoot length. The effect of the variability in grapevine water status on grapevine physiology was assessed by measuring CID, which was the main focus of the study. Other physiological measurements, such as gs and leaf and canopy temperature, were also conducted. The effect of these conditions on grape berry composition was also studied. In the Stellenbosch trial, soil water content, plant water status measurements (SWP, predawn LWP and LWP), physiological measurements (CID and gs) and berry size measurements were used to classify plots into water status treatments (“wet” and “dry” treatments). The effect of vigour differences was analysed separately from these treatments by using pruning mass as a covariate in the statistical analyses. The effect of vigour variability on the measurements was studied by looking at the effect of the covariate on the measurements, while shoot growth rate, shoot length and leaf area measurements were conducted as vegetative growth measurements. Differences in measurements were then studied between the treatments and between the vigour levels of the different plots. In the Wellington trial, plant water status was determined by irrigation, showing increased stress for treatments that received less irrigation. The differences in plant water status then caused differences in grapevine physiology between the treatments, leading to increased gs for increased irrigation. This of course influenced leaf internal CO2 and therefore CID, although CID was also clearly influenced by berry development. Berry size was influenced by irrigation, with larger berries found in wetter treatments, while berry chemical composition was influenced by the irrigation regime, with increased irrigation leading to increased pH and leading to trends showing increased total soluble solids and malic acid, and reduced total and tartaric acid and colour intensity. In the Stellenbosch trial, plots with higher vigour had increased shoot growth rate, longer shoots and increased leaf area, although topping influenced this. Wet treatment vines also showed slightly longer shoots and larger leaf areas. There were differences in soil water content between the wet and dry treatments, and this led to differences in plant water status. Vigour also influenced pre-dawn LWP, especially in the 2007 season, as higher-vigour vines struggled more to rehydrate through the night. Differences in plant water potential led to differences in grapevine physiology, with increased gs for vines from the wet treatment, while higher-vigour vines had slightly increased gs. The differences in gs led to gas exchange differences and therefore differences in CID, meaning that water status and vigour influenced CID. CID measurements illustrated the long term effect of water status on plant physiology, while measurements such as SWP illustrated the short term effects. CID measurements therefore proved to be accumulative over the season, in contrast to SWP measurements that were much more dependent on the current state of grapevine water status. Other physiological measurements showed that wet-treatment vines had higher photosynthetic rates and evapotranspiration and lower leaf temperatures, while higher-vigour vines had slightly increased evapotranspiration and decreased leaf temperatures. Wet-treatment vines had larger berries, while a higher vigour also led to slightly larger berries. Berry composition was influenced by treatment, where wet-treatment vines had increased pH and total soluble solids, while higher-vigour vines had increased juice pH and, in the 2008 season, decreased total soluble solids. Extremely stressed conditions did not show significant effects on plant water potential, but SWP measurements indicated slightly higher stress for the extremely stressed vines and LWP showed slightly less stressed conditions for these vines. Measurements of gs showed slightly lower values for the extremely stressed vines, while measurements of CID showed large significant differences, with the extremely stressed vines having measurements showing high stress. The measurement therefore indicated highly stressed conditions accurately, while other physiological measurements, such as photosynthetic rate, evapotranspiration and leaf temperatures, only showed trends and no significant differences. Measurements of stomatal conductance reacted to plant water status measurements throughout the diurnal measurement days, while CID only reacted slightly with gs changes during these days and was perhaps influenced more by berry chemical composition and development at this early stage of the season. Vigour and water status therefore influenced grapevine physiology, with a more direct effect by water status and an indirect effect by vigour due to microclimatic differences. This also influenced berry composition and therefore quality. In future studies, CID measurements should be done on juice from which organic acids have been removed in order to eliminate the effect of seasonal berry composition on the measurement. Measurements of CID proved to be an integrative, but sensitive, indicator of grapevine stress, especially at the end of the season. It might at best be useful as a post-harvest management tool for producers or grape buyers, especially for irrigation control, as has also been stated by Van Leeuwen et al. (2007). / AFRIKAANSE OPSOMMING: Binne-wingerd variasie in groeikrag en waterstatus is algemeen in Suid-Afrikaanse wingerde. Verskillende grondsoorte wat na aan mekaar voorkom, is seker een van die vernaamste oorsake van variasie in groeikrag, terwyl verskille in wingerdwaterstatus algemeen deur laterale watervloei in die wingerd, verstopte besproeiingspuite en verskille in grond waterhouvermoë geïnduseer word. Hierdie faktore kan aanleiding gee tot heterogene rypwording en verskille in vrugkwaliteit tussen verskillende dele van die wingerd, ‘n aspek wat so ver moontlik vermy moet word om kwaliteitwyne te kan produseer. Die meting van koolstof-isotoopdiskriminasie (KID) is bewys om as gereedskap te kan dien vir die assessering van wingerdfisiologie om die effekte van omgewingsparameters op blaar koolstofdioksied (CO2) - gasuitruiling en stomatale geleiding (gs) te bestudeer. Die stres/stremming as gevolg van ‘n watertekort in die wingerd in reaksie op hierdie omgewingstoestande kan dan bepaal word deur te kyk na hoeveel 13C deur die plantmateriaal geabsorbeer word ná 13C-diskriminasie plaasgevind het, en dít word deur die wingerdstrestoestande beïnvloed en kan ‘n aanduiding verskaf van die doeltreffendheid van waterverbruik. In hierdie studie is die variasie in wingerdwaterstatus en groeikrag bepaal om hierdie parameters in verskillende dele van die wingerd te kwantifiseer. Twee afsonderlike proewe is uitgevoer, die eerste by Wellington, Suid-Afrika, waar verskillende besproeiingsregimes gelei het tot verskille in die wingerdwaterstatus tussen persele. Die tweede proef was by Stellenbosch, Suid-Afrika, waar persele tussen verskillende groeikragklasse verdeel is en besproeiing in verskillende hoeveelhede vir verskillende besproeiingsbehandelings toegepas is. Binne-wingerd variasie in waterstatus (Wellington en Stellenbosch) en groeikrag (Stellenbosch) is toe gekwantifiseer en die effekte op sekere wingerd-fisiologiese parameters en korrelsamestelling is gemeet. Die behandelings in die Wellington-proef het gelei tot verskille in wingerdwaterstatus, wat deur metings van stamwaterpotensiaal (SWP) en blaarwaterpotensiaal (BWP) gekwantifiseer kon word. Grondverskille het ook gelei tot verskille in wingerdgroeikrag, wat deur metings van snoeimassa, blaaroppervlak en lootlengte gekwantifiseer is. Die effek van die variasie in wingerdwaterstatus op wingerdfisiologie is deur metings van KID bepaal wat die hooffokus van hierdie studie was. Ander fisiologiese metings, soos gs en blaar- en lowertemperatuur, is ook gedoen. Die effekte van hierdie toestande op die samestelling van die druiwekorrels is ook bestudeer. In die Stellenbosch-proef is grondwaterinhoud, metings van plantwaterstatus (SWP, voorsonopgang SWP en BWP), fisiologiese metings (KID en gs) en metings van korrelgrootte gebruik om die persele in waterstatusbehandelings (“nat” en “droë” behandelings) te verdeel. Die effek van verskille in groeikrag is apart van hierdie behandelings geanaliseer deur snoeimassa as ‘n kovariaat in die statistiese analises te gebruik. Die effek van groeikragvariasie op die metings is bestudeer deur ondersoek in te stel na die effek van die kovariaat op die metings, terwyl lootgroeitempo-, lootlengte- en blaaroppervlakmetings as metings van vegetatiewe groei uitgevoer is. Verskille in metings tussen die behandelings en tussen die groeikragvlakke van die verskillende persele is toe bestudeer. In die Wellington-proef is plantwaterstatus deur besproeiing bepaal, met verhoogde stres in behandelings waar daar minder besproeiing toegedien is. Die verskille in plantwaterstatus het dan verskille in wingerdfisiologie tussen die behandelings veroorsaak, wat gelei het tot ‘n verhoogde gs in die geval van verhoogde besproeiing. Dit het natuurlik ‘n effek op die interne CO2 van die blaar en dus op KID gehad, hoewel KID ook duidelik deur korrelontwikkeling beïnvloed is. Korrelgrootte is deur besproeiing beïnvloed, met groter korrels in die natter behandelings, terwyl die chemiese samestelling van die korrel deur besproeiingsregime beïnvloed is. Verhoogde besproeiing het pH verhoog en gelei na tendense wat verhoogde totale oplosbare vaste stowwe en appelsuur, en verminderde totale suur, wynsteensuur en kleurintensiteit getoon het. In die Stellenbosch-proef het persele met hoër groeikrag ook verhoogde lootgroeitempo, langer lote en verhoogde blaaroppervlak getoon, hoewel dit deur top beïnvloed is. Wingerdstokke van die nat behandeling het ook effe langer lote en groter blaaroppervlakke getoon. Daar was verskille in grondwaterinhoud tussen die nat en droë behandelings en dit het verskille in plantwaterstatus veroorsaak. Groeikrag is ook deur voor-sonopgang BWP beïnvloed, veral in die 2007-seisoen, aangesien stokke met hoër groeikrag meer gesukkel het om in die nag te rehidreer. Verskille in plantwaterpotensiaal het gelei tot verskille in wingerdfisiologie, met ‘n verhoogde gs vir stokke in die nat behandeling, terwyl stokke met hoër groeikrag ‘n effens verhoogde gs getoon het. Die verskille in gs het gelei tot verskille in gasuitruiling en dus verskille in KID, wat beteken dat waterstatus en groeikrag ‘n invloed op KID het. KID was meer verteenwoordigend van die langtermyneffekte van water status op plantfisiologie, terwyl metings soos SWP die korttermyneffekte weerspieël het. KID metings was dus akkumalatief oor die seisoen, terwyl SWP metings meer ‘n weerspieëling was van die huidige toestand van plantwaterpotensiaal. Ander fisiologiese metings het getoon dat stokke in die nat behandeling ‘n hoër fotosintesetempo en evapotranspirasie sowel as laer blaartemperature ondervind het, terwyl die stokke met hoër groeikrag effe verhoogde evapotranspirasie en verminderde blaartemperature getoon het. Stokke in die nat behandeling het groter korrels gehad, terwyl hoër groeikrag ook effens groter korrels veroorsaak het. Korrelsamestelling is deur die behandelings beïnvloed, met stokke in die nat behandeling wat verhoogde pH en totale oplosbare vaste stowwe getoon het, terwyl stokke met hoër groeikrag verhoogde pH van die sap en verminderde totale oplosbare vaste stowwe (laasgenoemde in die 2008-seisoen) gehad het. Uitermate toestande van stres het geen beduidende effekte op plantwaterpotensiaal getoon nie, hoewel SWP-metings effens hoër stres vir die uitermate gestresde wingerde getoon het en BWP effens minder gestresde toestande vir hierdie stokke getoon het. Metings van gs het effens laer waardes vir die uitermate gestresde stokke getoon, terwyl metings van KID groot noemenswaardige verskille getoon het, met die metings vir die uitermate gestresde wingerde wat hoër stres aangedui het. Dié meting het dus hoogs gestresde toestande akkuraat aangedui, terwyl ander fisiologiese metings, soos tempo van fotosintese, evapotranspirasie en blaartemperature net tendense en nie beduidende verskille aangedui het nie. Metings van stomatale geleiding het dwarsdeur die dae waarop daaglikse metings gedoen is op plantwaterstatusmetings gereageer, terwyl KID net effens met gs-veranderinge op hierdie dae gereageer het en moontlik meer deur die chemiese samestelling en ontwikkeling van die korrel in hierdie vroeë stadium van die seisoen beïnvloed is. Groeikrag en waterstatus het dus wingerdfisiologie beïnvloed, met ‘n meer direkte effek deur waterstatus en ‘n indirekte effek deur groeikrag as gevolg van mikroklimaatsverskille. Dit het ook korrelsamestelling en dus kwaliteit beïnvloed. In toekomstige studies moet KID-metings gedoen word op sap waarvan die organiese sure verwyder is om die effek van seisoenale korrelsamestelling op die meting uit te sluit. Metings van KID is getoon om ‘n integrerende, maar gevoelige, aanduider van wingerdstres te wees, veral aan die einde van die seisoen. Dit is ten beste miskien bruikbaar as naoesbestuursgereedskap vir produsente of druiwekopers, veral vir besproeiingsbeheer, soos ook reeds deur Van Leeuwen et al. (2007) aangedui is.

Sunflower is grown worldwide for oil production. In Australia, sunflower is under intense competition from canola as an oil crop. Confectionery sunflower aimed for the overseas snack food market is a potential opportunity for Australian farmers due to its ability to provide off-season supply to the major market located mainly in the northern hemisphere. To serve this aim, new confectionery sunflower hybrids adapted to Australian production system with drought tolerance, large achenes, high self-fertility and rust resistance are of paramount importance. The objectives of this project were to understand the genetic inheritance and genetic linkage relationships of the above traits and also to identify linked DNA markers with the potential for marker-assisted selection (MAS). The inheritance of achene-length was studied in a 12 x 12 diallel and a 14 x 12 North Carolina design II (NC II) mating design. Achene-length was evaluated under three environments for each mating design. Genetic correlation among all five achene-size related traits (achene-length, -width, - thickness and also kernel-length and -width) and 100-achene weight were moderate to high. Narrow-and broad-sense heritability for achene-length was moderate to high and similar for each design. The general combining ability (GCA) and maternal effects were found to be the main genetic factors controlling achene-length and each accounted for around 45% of the total genotypic variance. The inheritance of self-fertility (SF) was assessed with the same 12 x 12 diallel mating design evaluated in two environments. The 12 parents also displayed various levels of SF in addition to a wide range of achene-length. Negative genetic correlations between SF and all five achene-size related traits and 100-achene weight were observed. Modelling achene-length as a covariate revealed that certain pre-pollination floret characteristics represented by achene-length at maturity strongly affects SF in sunflower. When the effect of achene-length was removed statistically as a covariate, the specific combining ability (SCA) effect increased dramatically and became the major factor (other than pre-pollination floret characteristics) for SF determination and accounted for 65% of the total variance. Long achene genotypes tended to have low SF and a positive SCA effect was observed more frequently in F1 hybrids that involved small achene lines. The broad sense heritability was moderate (0.663 ± 0.017), while the estimate of narrow-sense heritability was nearly zero. A segregating F6:7 recombinant inbred line population with 239 progeny was developed from a cross between HaR4, a low delta (Δ), high transpiration efficiency (TE), sunflower line with rust resistance but relatively low SF, and SA52 with all the characteristics in the opposite direction. The population was evaluated in two environments for three TE-related traits including Δ, specific leaf weight (SLW) and chlorophyll content (CC) and four agronomic traits i.e. days to flowering (DTF), plant height (PlantH), SF and rust resistance (RustR). A linkage map 1074.6 cM in length with 138 SSR and 37 ISSR markers consisting of 19 linkage groups was constructed. Spatially adjusted BLUPs were used in QTL analysis for individual and combined environments. Altogether, 10 QTL were identified for Δ, SLW and CC at individual environments, and 7 were detected by a combined analysis. Sixteen QTL were identified for 4 agronomic traits and 14 of them were confirmed by a combined analysis. Significant linkage among all three TE related traits was discovered on linkage group 4 (LG4), where alleles from one parent had the same direction of influence on TE which suggested either coupling phase linkage or pleiotropy. QTL for SF and RustR were closely linked on LG13. An F2 population with 502 progeny developed from a cross between Jumbo, a large achene confectionery sunflower line, and RIL25, an oilseed small achene line, was evaluated for five achene-size related traits in a single field trial. Using a subset of 178 progeny a framework genetic linkage map, 899.7 cM in length, containing 128 SSR markers and 23 ISSR markers was constructed with 18 linkage groups. Based on the results of a simulated QTL analysis, a modified selective genotyping strategy was employed for QTL analysis with 178 F2 individuals (128 randomly selected, 25 shortest achene F2’s and 25 longest achene F2’s). Altogether 18 achene-size related QTL with small effects were identified. Achene-size related genes characteristically expressed additive gene action although several alleles were partially recessive towards the large achene parent. The 18 identified achene-size related QTL were validated in the same diallel mating experiment mentioned above. Significant marker-trait associations were detected for 15 QTL. In addition to 12 significant alleles detected from the F2 population, 12 additional alleles associated with achene-size traits were revealed after the analysis with the diallel. These markers are ready to be used in MAS for confectionery sunflower breeding. Great steps have been made in obtaining essential knowledge of inheritance and linkage of target traits for breeding confectionery sunflower adapted to Australian production environments. The identified markers can be used in MAS and further enhance the breeding process.

Sunflower is grown worldwide for oil production. In Australia, sunflower is under intense competition from canola as an oil crop. Confectionery sunflower aimed for the overseas snack food market is a potential opportunity for Australian farmers due to its ability to provide off-season supply to the major market located mainly in the northern hemisphere. To serve this aim, new confectionery sunflower hybrids adapted to Australian production system with drought tolerance, large achenes, high self-fertility and rust resistance are of paramount importance. The objectives of this project were to understand the genetic inheritance and genetic linkage relationships of the above traits and also to identify linked DNA markers with the potential for marker-assisted selection (MAS). The inheritance of achene-length was studied in a 12 x 12 diallel and a 14 x 12 North Carolina design II (NC II) mating design. Achene-length was evaluated under three environments for each mating design. Genetic correlation among all five achene-size related traits (achene-length, -width, - thickness and also kernel-length and -width) and 100-achene weight were moderate to high. Narrow-and broad-sense heritability for achene-length was moderate to high and similar for each design. The general combining ability (GCA) and maternal effects were found to be the main genetic factors controlling achene-length and each accounted for around 45% of the total genotypic variance. The inheritance of self-fertility (SF) was assessed with the same 12 x 12 diallel mating design evaluated in two environments. The 12 parents also displayed various levels of SF in addition to a wide range of achene-length. Negative genetic correlations between SF and all five achene-size related traits and 100-achene weight were observed. Modelling achene-length as a covariate revealed that certain pre-pollination floret characteristics represented by achene-length at maturity strongly affects SF in sunflower. When the effect of achene-length was removed statistically as a covariate, the specific combining ability (SCA) effect increased dramatically and became the major factor (other than pre-pollination floret characteristics) for SF determination and accounted for 65% of the total variance. Long achene genotypes tended to have low SF and a positive SCA effect was observed more frequently in F1 hybrids that involved small achene lines. The broad sense heritability was moderate (0.663 ± 0.017), while the estimate of narrow-sense heritability was nearly zero. A segregating F6:7 recombinant inbred line population with 239 progeny was developed from a cross between HaR4, a low delta (Δ), high transpiration efficiency (TE), sunflower line with rust resistance but relatively low SF, and SA52 with all the characteristics in the opposite direction. The population was evaluated in two environments for three TE-related traits including Δ, specific leaf weight (SLW) and chlorophyll content (CC) and four agronomic traits i.e. days to flowering (DTF), plant height (PlantH), SF and rust resistance (RustR). A linkage map 1074.6 cM in length with 138 SSR and 37 ISSR markers consisting of 19 linkage groups was constructed. Spatially adjusted BLUPs were used in QTL analysis for individual and combined environments. Altogether, 10 QTL were identified for Δ, SLW and CC at individual environments, and 7 were detected by a combined analysis. Sixteen QTL were identified for 4 agronomic traits and 14 of them were confirmed by a combined analysis. Significant linkage among all three TE related traits was discovered on linkage group 4 (LG4), where alleles from one parent had the same direction of influence on TE which suggested either coupling phase linkage or pleiotropy. QTL for SF and RustR were closely linked on LG13. An F2 population with 502 progeny developed from a cross between Jumbo, a large achene confectionery sunflower line, and RIL25, an oilseed small achene line, was evaluated for five achene-size related traits in a single field trial. Using a subset of 178 progeny a framework genetic linkage map, 899.7 cM in length, containing 128 SSR markers and 23 ISSR markers was constructed with 18 linkage groups. Based on the results of a simulated QTL analysis, a modified selective genotyping strategy was employed for QTL analysis with 178 F2 individuals (128 randomly selected, 25 shortest achene F2’s and 25 longest achene F2’s). Altogether 18 achene-size related QTL with small effects were identified. Achene-size related genes characteristically expressed additive gene action although several alleles were partially recessive towards the large achene parent. The 18 identified achene-size related QTL were validated in the same diallel mating experiment mentioned above. Significant marker-trait associations were detected for 15 QTL. In addition to 12 significant alleles detected from the F2 population, 12 additional alleles associated with achene-size traits were revealed after the analysis with the diallel. These markers are ready to be used in MAS for confectionery sunflower breeding. Great steps have been made in obtaining essential knowledge of inheritance and linkage of target traits for breeding confectionery sunflower adapted to Australian production environments. The identified markers can be used in MAS and further enhance the breeding process.

The forest product industry is keenly interested in extending the normal planting season, as well as in the comparative field performance of standard nursery bare-root seedlings and containerized rooted cuttings. The effect of seasonal planting dates on survival, above and belowground biomass allocation, water relations, gas exchange attributes and foliar carbon isotope composition (δ13C) of two stock types of slash pine (Pinus elliottii Engelm.) were examined. Slash pine bare-root seedlings (BRS) and containerized rooted cuttings (CRC) were hand planted in September, November, January, March and April in three consecutive planting seasons (2000-2001, 2001-2002 and 2002-2003) on three sites with silt loam topsoils in southwestern Louisiana. First-year mean survival of CRC across all planting dates and sites was consistently high at 96 to 98%, whereas BRS survival was significantly (P < 0.0001) lower at 59 to 81% and highly variable among study sites and dates through three planting seasons. Generally, there was a negative relationship between soil moisture at the time of planting and first-year survival of BRS planted September through March in 2001-2002 and 2002-2003 planting seasons, whereas the opposite was observed only for BRS planted in April 2002 and 2003. Survival of CRC was affected very little by the variation in soil moisture. Containerized rooted cuttings had higher early above and belowground biomass, and height and diameter than did BRS. However, three years after planting the size differences between stock types disappeared or became negligible. Early size differences among trees planted September through March also decreased after three years, although September trees were tallest. Growth of the April-planted trees was poor compared to trees planted in other months. Late-planted April trees had higher δ13C values, and higher water-use efficiency in the first growing season compared to earlier planted trees. Differences in δ13C values among the planting dates disappeared in the second growing season. Net photosynthesis rates did not differ considerably between stock types or among planting dates in the second and third growing seasons. This study indicates that it is possible to extend the planting season to as early as September and as late as March by using CRC.

The forest product industry is keenly interested in extending the normal planting season, as well as in the comparative field performance of standard nursery bare-root seedlings and containerized rooted cuttings. The effect of seasonal planting dates on survival, above and belowground biomass allocation, water relations, gas exchange attributes and foliar carbon isotope composition (&#948;13C) of two stock types of slash pine (Pinus elliottii Engelm.) were examined. Slash pine bare-root seedlings (BRS) and containerized rooted cuttings (CRC) were hand planted in September, November, January, March and April in three consecutive planting seasons (2000-2001, 2001-2002 and 2002-2003) on three sites with silt loam topsoils in southwestern Louisiana. First-year mean survival of CRC across all planting dates and sites was consistently high at 96 to 98%, whereas BRS survival was significantly (P < 0.0001) lower at 59 to 81% and highly variable among study sites and dates through three planting seasons. Generally, there was a negative relationship between soil moisture at the time of planting and first-year survival of BRS planted September through March in 2001-2002 and 2002-2003 planting seasons, whereas the opposite was observed only for BRS planted in April 2002 and 2003. Survival of CRC was affected very little by the variation in soil moisture. Containerized rooted cuttings had higher early above and belowground biomass, and height and diameter than did BRS. However, three years after planting the size differences between stock types disappeared or became negligible. Early size differences among trees planted September through March also decreased after three years, although September trees were tallest. Growth of the April-planted trees was poor compared to trees planted in other months. Late-planted April trees had higher &#948;13C values, and higher water-use efficiency in the first growing season compared to earlier planted trees. Differences in &#948;13C values among the planting dates disappeared in the second growing season. Net photosynthesis rates did not differ considerably between stock types or among planting dates in the second and third growing seasons. This study indicates that it is possible to extend the planting season to as early as September and as late as March by using CRC.

Sunflower is grown worldwide for oil production. In Australia, sunflower is under intense competition from canola as an oil crop. Confectionery sunflower aimed for the overseas snack food market is a potential opportunity for Australian farmers due to its ability to provide off-season supply to the major market located mainly in the northern hemisphere. To serve this aim, new confectionery sunflower hybrids adapted to Australian production system with drought tolerance, large achenes, high self-fertility and rust resistance are of paramount importance. The objectives of this project were to understand the genetic inheritance and genetic linkage relationships of the above traits and also to identify linked DNA markers with the potential for marker-assisted selection (MAS). The inheritance of achene-length was studied in a 12 x 12 diallel and a 14 x 12 North Carolina design II (NC II) mating design. Achene-length was evaluated under three environments for each mating design. Genetic correlation among all five achene-size related traits (achene-length, -width, - thickness and also kernel-length and -width) and 100-achene weight were moderate to high. Narrow-and broad-sense heritability for achene-length was moderate to high and similar for each design. The general combining ability (GCA) and maternal effects were found to be the main genetic factors controlling achene-length and each accounted for around 45% of the total genotypic variance. The inheritance of self-fertility (SF) was assessed with the same 12 x 12 diallel mating design evaluated in two environments. The 12 parents also displayed various levels of SF in addition to a wide range of achene-length. Negative genetic correlations between SF and all five achene-size related traits and 100-achene weight were observed. Modelling achene-length as a covariate revealed that certain pre-pollination floret characteristics represented by achene-length at maturity strongly affects SF in sunflower. When the effect of achene-length was removed statistically as a covariate, the specific combining ability (SCA) effect increased dramatically and became the major factor (other than pre-pollination floret characteristics) for SF determination and accounted for 65% of the total variance. Long achene genotypes tended to have low SF and a positive SCA effect was observed more frequently in F1 hybrids that involved small achene lines. The broad sense heritability was moderate (0.663 ± 0.017), while the estimate of narrow-sense heritability was nearly zero. A segregating F6:7 recombinant inbred line population with 239 progeny was developed from a cross between HaR4, a low delta (Δ), high transpiration efficiency (TE), sunflower line with rust resistance but relatively low SF, and SA52 with all the characteristics in the opposite direction. The population was evaluated in two environments for three TE-related traits including Δ, specific leaf weight (SLW) and chlorophyll content (CC) and four agronomic traits i.e. days to flowering (DTF), plant height (PlantH), SF and rust resistance (RustR). A linkage map 1074.6 cM in length with 138 SSR and 37 ISSR markers consisting of 19 linkage groups was constructed. Spatially adjusted BLUPs were used in QTL analysis for individual and combined environments. Altogether, 10 QTL were identified for Δ, SLW and CC at individual environments, and 7 were detected by a combined analysis. Sixteen QTL were identified for 4 agronomic traits and 14 of them were confirmed by a combined analysis. Significant linkage among all three TE related traits was discovered on linkage group 4 (LG4), where alleles from one parent had the same direction of influence on TE which suggested either coupling phase linkage or pleiotropy. QTL for SF and RustR were closely linked on LG13. An F2 population with 502 progeny developed from a cross between Jumbo, a large achene confectionery sunflower line, and RIL25, an oilseed small achene line, was evaluated for five achene-size related traits in a single field trial. Using a subset of 178 progeny a framework genetic linkage map, 899.7 cM in length, containing 128 SSR markers and 23 ISSR markers was constructed with 18 linkage groups. Based on the results of a simulated QTL analysis, a modified selective genotyping strategy was employed for QTL analysis with 178 F2 individuals (128 randomly selected, 25 shortest achene F2’s and 25 longest achene F2’s). Altogether 18 achene-size related QTL with small effects were identified. Achene-size related genes characteristically expressed additive gene action although several alleles were partially recessive towards the large achene parent. The 18 identified achene-size related QTL were validated in the same diallel mating experiment mentioned above. Significant marker-trait associations were detected for 15 QTL. In addition to 12 significant alleles detected from the F2 population, 12 additional alleles associated with achene-size traits were revealed after the analysis with the diallel. These markers are ready to be used in MAS for confectionery sunflower breeding. Great steps have been made in obtaining essential knowledge of inheritance and linkage of target traits for breeding confectionery sunflower adapted to Australian production environments. The identified markers can be used in MAS and further enhance the breeding process.

Sunflower is grown worldwide for oil production. In Australia, sunflower is under intense competition from canola as an oil crop. Confectionery sunflower aimed for the overseas snack food market is a potential opportunity for Australian farmers due to its ability to provide off-season supply to the major market located mainly in the northern hemisphere. To serve this aim, new confectionery sunflower hybrids adapted to Australian production system with drought tolerance, large achenes, high self-fertility and rust resistance are of paramount importance. The objectives of this project were to understand the genetic inheritance and genetic linkage relationships of the above traits and also to identify linked DNA markers with the potential for marker-assisted selection (MAS). The inheritance of achene-length was studied in a 12 x 12 diallel and a 14 x 12 North Carolina design II (NC II) mating design. Achene-length was evaluated under three environments for each mating design. Genetic correlation among all five achene-size related traits (achene-length, -width, - thickness and also kernel-length and -width) and 100-achene weight were moderate to high. Narrow-and broad-sense heritability for achene-length was moderate to high and similar for each design. The general combining ability (GCA) and maternal effects were found to be the main genetic factors controlling achene-length and each accounted for around 45% of the total genotypic variance. The inheritance of self-fertility (SF) was assessed with the same 12 x 12 diallel mating design evaluated in two environments. The 12 parents also displayed various levels of SF in addition to a wide range of achene-length. Negative genetic correlations between SF and all five achene-size related traits and 100-achene weight were observed. Modelling achene-length as a covariate revealed that certain pre-pollination floret characteristics represented by achene-length at maturity strongly affects SF in sunflower. When the effect of achene-length was removed statistically as a covariate, the specific combining ability (SCA) effect increased dramatically and became the major factor (other than pre-pollination floret characteristics) for SF determination and accounted for 65% of the total variance. Long achene genotypes tended to have low SF and a positive SCA effect was observed more frequently in F1 hybrids that involved small achene lines. The broad sense heritability was moderate (0.663 ± 0.017), while the estimate of narrow-sense heritability was nearly zero. A segregating F6:7 recombinant inbred line population with 239 progeny was developed from a cross between HaR4, a low delta (Δ), high transpiration efficiency (TE), sunflower line with rust resistance but relatively low SF, and SA52 with all the characteristics in the opposite direction. The population was evaluated in two environments for three TE-related traits including Δ, specific leaf weight (SLW) and chlorophyll content (CC) and four agronomic traits i.e. days to flowering (DTF), plant height (PlantH), SF and rust resistance (RustR). A linkage map 1074.6 cM in length with 138 SSR and 37 ISSR markers consisting of 19 linkage groups was constructed. Spatially adjusted BLUPs were used in QTL analysis for individual and combined environments. Altogether, 10 QTL were identified for Δ, SLW and CC at individual environments, and 7 were detected by a combined analysis. Sixteen QTL were identified for 4 agronomic traits and 14 of them were confirmed by a combined analysis. Significant linkage among all three TE related traits was discovered on linkage group 4 (LG4), where alleles from one parent had the same direction of influence on TE which suggested either coupling phase linkage or pleiotropy. QTL for SF and RustR were closely linked on LG13. An F2 population with 502 progeny developed from a cross between Jumbo, a large achene confectionery sunflower line, and RIL25, an oilseed small achene line, was evaluated for five achene-size related traits in a single field trial. Using a subset of 178 progeny a framework genetic linkage map, 899.7 cM in length, containing 128 SSR markers and 23 ISSR markers was constructed with 18 linkage groups. Based on the results of a simulated QTL analysis, a modified selective genotyping strategy was employed for QTL analysis with 178 F2 individuals (128 randomly selected, 25 shortest achene F2’s and 25 longest achene F2’s). Altogether 18 achene-size related QTL with small effects were identified. Achene-size related genes characteristically expressed additive gene action although several alleles were partially recessive towards the large achene parent. The 18 identified achene-size related QTL were validated in the same diallel mating experiment mentioned above. Significant marker-trait associations were detected for 15 QTL. In addition to 12 significant alleles detected from the F2 population, 12 additional alleles associated with achene-size traits were revealed after the analysis with the diallel. These markers are ready to be used in MAS for confectionery sunflower breeding. Great steps have been made in obtaining essential knowledge of inheritance and linkage of target traits for breeding confectionery sunflower adapted to Australian production environments. The identified markers can be used in MAS and further enhance the breeding process.

Carbon isotope discrimination (?13C) may be an appropriate measure for determination of the water use efficiency (WUE) of plants. This work confirms the negative correlation of ?13C and WUE in selected genotypes of barley and wheat. It also deals with the suitability of using the observed relationship in breeding of drought resistant crops.

Made available in DSpace on 2014-06-11T19:32:25Z (GMT). No. of bitstreams: 0 Previous issue date: 2006-02-03Bitstream added on 2014-06-13T19:22:04Z : No. of bitstreams: 1 marchese_ja_dr_botfca.pdf: 1209124 bytes, checksum: 3852bccf50e18f3028e21b50674aba78 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Artemisia annua L. (Asteraceae) é uma planta herbácea altamente aromática, nativa da Ásia e aclimatada no Brasil. As folhas são fonte abundante de artemisinina, uma lactona sesquiterpênica que, conjuntamente aos seus derivados semi-sintéticos, apresentam ação efetiva contra as cepas resistentes das espécies de Plasmodium causadoras da malária. Os objetivos deste trabalho foram identificar o mecanismo fotossintético e avaliar o efeito de diferentes condições de temperatura e fotoperíodo no crescimento e desenvolvimento do acesso CPQBA 2/39x1V de A. annua. Para identificar o mecanismo fotossintético foram realizados experimentos para determinar composição dos isótopos do carbono (d 13C) e a anatomia foliar associada a testes histoquímicos. Para avaliar o efeito da temperatura e fotoperíodo no crescimento e desenvolvimento de A. annua foram realizados dois experimentos em câmaras fotoperiódicas (o primeiro com temperaturas médias máx. de 37ºC e mín. de 19ºC e o segundo com máx. de 29ºC e mín. de 13ºC) e um experimento com 6 épocas de plantio em campo no município de Pato Branco-PR (26º11' S, 52º36' W e 760 m de altitude), sul do Brasil. Um último experimento realizado, foi a aplicação exógena de artemisinina em plantas de A. annua para verificar o papel da molécula na indução do florescimento. Como resultados, A. annua apresentou uma d 13C - 31.76 l 0.07, valor típico de espécies com mecanismo fotossintético C3, que apresentam em média valores de d 13C - 28. Os estudos da anatomia foliar e testes histoquímicos confirmaram os resultados encontrados para a d 13C, onde, a despeito da existência de células parenquimáticas formando um anel ao redor do feixe vascular, estas não apresentaram cloroplastos e amido, confirmando ser A. annua uma espécie de mecanismo... / Leaves of Artemisia annua L. are a plentiful source of artemisinin, a drug with proven effectiveness against malaria. One of the objectives of this work was to identify the photosynthetic type of A. annua through studies of the carbon isotope composition (d 13C) and the leaf anatomy. We also verified the growth and development of the CPQBA 2/39x1V accession under moist subtropical climate. Two experiments were carried out in greenhouse using photoperiodic chambers. The plants were submitted to photoperiods of 7, 9, 11, 13, 15 and 17 h, in two natural conditions of temperature: spring/summer (maximum of 37ºC and minimum of 19ºC) and autumn/winter (maximum of 29ºC and minimum of 13ºC). Another experiment with different planting dates at field was carried out in Pato Branco, PR (26º11' S, 52º36' W and 760 m), Southern part of Brazil. The last experiment was the application of artemisinin (0, 500, 5000, and 10000 mg L-1) in A. annua plants to verify the role of the molecule on the flowering induction. A. annua presented a d 13C - 31.76 l 0.07, what characterizes it as typical species with a C3 photosynthetic mechanism, with an average of d 13C- 28, while C4 species possess an average of d 13C - 14. The study of A. annua leaf anatomy confirms the results of d 13C, where, in spite of the existence of parenchymatic cells forming a different sheath surrounding of the vascular tissue, these cells do not have chloroplast or starch. These features do not describe the Kranz anatomy typical of C4 species. The application of artemisinin did not induce the flowering of A. annua at any of the concentrations used. The results suggest that the accumulation of the artemisinin in the pre-flowering and flowering phases in A. annua is not the physiological signal of floral induction in this species...