Thesis (MSc)--Stellenbosch University, 2010. / ENGLISH ABSTRACT: Late blight caused by Phytophthora infestans on potato and tomato causes major economic losses worldwide. Until the 1980s, P. infestans populations outside its centre of origin (either central Mexico or the Andean region) only consisted of one mating type (A1), which prevented the pathogen from reproducing sexually. Pathogen populations outside the centre of origin most likely only consisted of a few genotypes prior to the 1980’s. Pan globally, these genotypes probably first consisted of genotype/s that had mitochondrial DNA (mtDNA) haplotype Ia, which was subsequently replaced by a mtDNA haplotype Ib genotype known as the US-1 lineage. This relative simple population structure of the pathogen changed almost worldwide in the late 1970s and early 1980s, when a second set of migrations took place from the centre of origin. These populations contained both A1 and A2 mating type isolates that consisted of several different genotypes, which were more virulent than the pre-1970s genotypes and resulted in the displacement of these genotypes almost worldwide. Some of the new genotypes were also resistant to metalaxyl, the fungicide that was most effective in controlling late blight. In Sub-Saharan Africa (SSA), the characteristics of P. infestans populations are not well documented in most countries except South Africa, Kenya and Uganda. Previous studies in SSA showed that populations were dominated by the US-1 lineage and its variants. The exceptions were reports of the presence of a few mtDNA haplotype Ia isolates in Rwanda and Ethiopia. The current study aimed to determine the population structure of P. infestans in eight selected SSA countries (Burundi, Kenya, Rwanda, Tanzania, Uganda, Malawi, Mozambique and South Africa), mainly on potato and on a limited scale on tomato and petunia, using ‘old’ markers (mating type determination, glucose-6-phosphate isomerase [Gpi] genotyping, mtDNA haplotyping, DNA fingerprinting with probe RG-57 and metalaxyl sensitivity). Populations were further also genotyped using seven recently published Simple Sequence Repeats (SSRs) markers. This information would help to define the population structure of P. infestans in SSA for the first time on a regional basis, and will also determine whether new migrations have taken place since the last characterization studies took place in 2001. A survey in the eight SSA countries yielded a total of 281 P. infestans isolates, mainly obtained from potato fields (Tanzania, Kenya, Uganda, Rwanda, Burundi, Malawi and South Africa), but also from tomato (Malawi, Mozambique and South Africa) and Petunia ´ hybrida (South Africa) that were characterized. Characterization of subsets of the isolates with the ‘old’ markers (176 isolates for mating type, 281 isolates for mtDNA, 70 isolates for [Gpi] and 49 isolates with restriction fragment length polymorphism analysis with probe RG-57), showed that most of the isolates belonged to the US-1 genotype or its variants (US-1.10 and US-1.11). The exception were isolates that belonged to genotype KE-1 (A1 mating type, mtDNA haplotype Ia, Gpi 90/100 and unique RG-57 genotype) that was identified in two fields in Kenya. Genotype KE-1, based on the ‘old’ marker data, is related to genotypes (RW-1 and RW-2) previously identified in Rwanda, and several Ecuadorean and European genotypes. Metalaxyl sensitivity testing of 64 isolates showed that metalaxyl resistant potato isolates were present in all the countries except Malawi, whereas all the tomato isolates were sensitive. Genotyping of 176 isolates with seven recently published simple sequence repeat (SSR) markers revealed a high number (79) of multi-locus genotypes (MLGs) in SSA. However, when locus D13, which was difficult to score, was excluded only 35 MLGs were identified. When locus D13 was excluded from analyses of molecular variance (AMOVA), (i) there was no significant genetic differentiation between populations from central-east Africa (Burundi, Kenya, Rwanda, Tanzania and Uganda), south-east Africa (Malawi and Mozambique) and South Africa, (ii) the KE-1 population was genetically differentiated (Fst = 0.33; P = 0.001) from the US-1 and US-1.10 populations and (iii) genetic differentiation between populations from potato and tomato was low (Fst = 0.07; P = 0.004). The study has expanded the worldwide genotypic database of P. infestans for SSA. Previously, no populations were characterized from Burundi, Malawi and Mozambique. The characterization work showed that migrations seem unlikely to have taken place in SSA, or if these did occur, it was on a very limited scale. The more severe epidemics in some SSA countries could be due to the presence of metalaxyl resistance. Furthermore, the occurrence of mutations or mitotic recombination might have resulted in more aggressive and/or better adapted genotypes, for example the US-1.10 lineage that was only detected in the Western Cape Province of South Africa. The significance of the discovery of the KE-1 genotype in Kenya needs further investigation since it might (i) be an asexual descendent of genotypes (RW-1 and RW-2) that were previously reported in Rwanda in the 1980s, (ii) previously have gone undetected due to the small surveys that were conducted in SSA, (iii) be a new migrant from countries other than SSA or (iv) have been introduced in the very first introductions into Kenya prior to the 1970s. The SSR results from the survey will allow comparison of the SSA late blight populations with other populations worldwide through the EucaBlight database in future studies. / AFRIKAANSE OPSOMMING: Laatroes, veroorsaak deur Phytophthora infestans op aartappel en tamatie, veroorsaak groot ekonomiese verliese wêreldwyd. Phytophthora infestans populasies buite hul kern van oorsprong (óf sentraal Meksiko óf die Andes area), het tot die 1980’s slegs uit een paringstipe (A1) bestaan, wat verhoed het dat die patogeen geslagtelik vermeerder. Patogeenpopulasies buite die kern van oorsprong, het heel moontlik vóór die 1980’s slegs uit ‘n paar genotipes bestaan. Wêreldwyd, het hierdie genotipes moontlik aanvanklik uit genotipe(s) bestaan wat mitokondriale DNS (mtDNS) haplotipe Ia bevat het, wat later met ‘n mtDNS haplotipe Ib genotipe, bekend as die US-1 genotipe, vervang is. Hierdie relatiewe eenvoudige populasiestruktuur van die patogeen, het omtrent wêreldwyd in die láát 1970’s en vroeë 1980’s verander, toe ‘n tweede stel migrasies vanaf die patogeen se kern van oorsprong plaasgevind het. Hierdie populasies het beide A1 en A2 paringstipe isolate ingesluit, wat uit verskeie verskillende genotipes bestaan het, wat meer virulent as die vóór-1970’s genotipes was, en wat die verskuiwing van hierdie genotipes omtrent wêrelwyd tot gevolg gehad het. Sommige van die nuwe genotipes was ook weerstandbiedend teen metalaksiel, die fungisied wat mees effektief in die beheer van laatroes was. Die kenmerke van P. infestans populasies is nie goed in die meeste lande in Sub- Sahara Afrika (SSA) gedokumenteer nie, behalwe vir Suid-Afrika, Kenia en Uganda. Vorige studies in SSA het aangedui dat populasies deur die US-1 genotipe en sy variante gedomineer word. Die uitsonderings was aantekeninge oor die teenwoordigheid van ‘n paar mtDNS haplotipe Ia isolate in Rwanda en Etiopië. Die huidige studie was daarop gemik om die populasiestruktuur van P. infestans in agt geselekteerde SSA lande (Burundi, Kenia, Rwanda, Tanzanië, Uganda, Malawi, Mosambiek en Suid-Afrika), hoofsaaklik op aartappel en op ‘n beperkte skaal op tamatie en petunia, vas te stel, deur die gebruik van ‘ou’ merkers (paringstipe-bepaling, glukose-6-fosfaat isomerase [Gpi] genotipering, mtDNS haplotipering, DNS fingerafdrukke met RG-57 en metalaksielsensitiwiteit). Die genotipe van populasies is verder ook bepaal deur gebruik te maak van sewe onlangs-gepubliseerde “Simple Sequence Repeats (SSRs)” merkers. Hierdie inligting sal help om die populasiestruktuur van P. infestans in SSA vir die eerste keer op ‘n streeksbasis vas te stel, en sal ook bepaal of nuwe migrasies sedert die laaste karakteriseringstudies wat in 2001 uitgevoer is, plaasgevind het. ‘n Opname in die agt SSA lande, het ‘n totaal van 281 P. infestans isolate opgelewer, hoofsaaklik vanaf aartappellande (Tanzanië, Kenia, Uganda, Rwanda, Burundi, Malawi en Suid-Afrika), maar ook vanaf tamatie (Malawi, Mosambiek en Suid- Afrika) en Petunia ´ hybrida (Suid-Afrika) wat gekarakteriseer is. Karakterisering van geselekteerde isolate met die ‘ou’ merkers (176 isolate vir paringstipe, 281 isolate vir mtDNS, 70 isolate vir Gpi en 49 isolate met restriksiefragment-lengte-polimorfismeanalise met RG-57), het aangetoon dat die meeste van die isolate aan die US-1 genotipe of sy variante (US-1.10 en US-1.11) behoort het. Die uitsondering was isolate wat tot die genotipe KE-1 behoort het (A1 paringstipe, mtDNS haplotipe Ia, Gpi 90/100 en unieke RG-57 genotipe) wat in twee velde in Kenia geïdentifiseer is. Genotipe KE-1, gebaseer op die ‘ou’ merkerdata, is aan genotipes (RW-1 en RW-2) verwant, wat voorheen in Rwanda, en verskeie Ekwadoreaanse en Europese lande geïdentifiseer is. Metalaksielsensitiwiteitstoetsing van 64 isolate het aangetoon dat metalaksiel-weerstandbiedende aartappel-isolate in al die lande teenwoordig was, behalwe vir Malawi, terwyl al die tamatie-isolate sensitief was. Genotipering van 176 isolate met sewe onlangs gepubliseerde “Simple Sequence Repeat” (SSR) merkers, het ‘n hoë aantal (79) multilokus genotipes (MLGs) in SSA aangedui. Met die uitsluiting van lokus D13, wat moeilik was om te evalueer, is slegs 35 MLGs egter geïdentifiseer. Met die uitsluiting van lokus D13 uit die analise van molekulêre variansie (AMOVA), was (i) daar geen betekenisvolle genetiese differensiasie tussen populasies van sentraal-oos Afrika (Burundi, Kenia, Rwanda, Tanzanië en Uganda), suid-oos Afrika (Malawi en Mosambiek) en Suid-Afrika nie, (ii) die KE-1 populasie geneties (Fst = 0.33; P = 0.001) van die US-1 en US-1.10 populasies gedifferensieerd en (iii) genetiese differensiasie tussen populasies vanaf aartappel en tamatie laag (Fst = 0.07; P = 0.004). Die studie het die wêreldwye genotipe-databasis van P. infestans vir SSA uitgebrei. Voorheen is geen populasies vanuit Burundi, Malawi en Mosambiek gekarakteriseer nie. Die karakteriseringswerk het aangetoon dat die waarskynlikheid klein is dat migrasies in SSA plaasgevind het, of indien dit wel plaasgevind het, dit op ‘n baie beperkte skaal plaasgevind. Die meer ernstige epidemies in sommige SSA lande kan die gevolg wees van die teenwoordigheid van metalaksiel-weerstand. Die voorkoms van mutasies of mitotiese rekombinasie kon verder meer aggressiewe en/of beter aangepaste genotipes tot gevolg gehad het, byvoorbeeld die US-1.10 genotipe wat slegs in die Westelike Kaapprovinsie van Suid-Afrika waargeneem is. Die betekenis van die ontdekking van die KE-1 genotipe in Kenia benodig verdere ondersoek aangesien dit (i) ‘n ongeslagtelike afstammeling van genotipes (RW-1 en RW-2) mag wees wat voorheen in die 1980’s in Rwanda aangeteken is, (ii) voorheen nie waargeneem is nie weens die klein opnames wat in SSA uitgevoer is, (iii) ‘n nuwe genotipe van lande buite die SSA kan wees of (iv) ingebring is tydens die heel eerste inkoms in Kenia vóór die 1970’s. Die SSR resultate van die opname sal vergelykings tussen die SSA laatroespopulasies en ander populasies wêreldwyd toelaat, deur gebruik te maak van die EucaBlight databasis in toekomstige studies.

A requeima, causada pelo oomiceto Phytophthora infestans e a pinta preta, causada pelo fungo Alternaria solani, estão entre as doenças mais importantes e destrutivas da cultura da batata no Brasil e no mundo. Considerando a importância do controle químico em sistemas integrados e a necessidade de se conhecer detalhadamente o desempenho de fungicidas e indutores de resistência visando à sustentabilidade da produção, o presente estudo objetivou caracterizar e comparar a ação preventiva, residual, curativa, anti-esporulante e resistência à chuva de fungicidas em condições controladas, bem como, avaliar em campo o potencial de controle de fungicidas e indutores de resistência e seus reflexos sobre a produtividade total e comercial de tubérculos. Todos os fungicidas promoveram elevada ação preventiva contra a requeima e a pinta preta. Os fungicidas sistêmicos ou com alta tenacidade proporcionaram controle das duas doenças até os 12 DAP, enquanto que os fungicidas translaminares até os 9 DAP. Quanto à ação curativa e anti-esporulante destacaram-se principalmente os fungicidas sistêmicos aplicados até as 24 horas após a inoculação (HAI). Os fungicidas translaminares foram capazes de inibir a requeima quando aplicados até 12 HAI e os de contato destacaram-se apenas para ação preventiva. Semelhantemente, os fungicidas sistêmicos, translaminares ou com alta tenacidade foram os menos afetados pela chuva simulada. O aumento do tempo de secagem promoveu uma maior retenção ou absorção dos produtos reduzindo o impacto negativo da precipitação. Os melhores níveis de controle, produtividade e qualidade de tubérculos foram obtidos com os fungicidas mandipropamida+clorotalonil, fluopicolida+propamocarbe, dimetomorfe+ametoctradina, mandipropamida, fenamidona+propamocarbe, bentiavalicarbe+ fluazinam, seguidos de dimetomorfe+clorotalonil mefenoxam+clorotalonil e famoxadona+ cimoxanil+mancozebe para requeima e azoxistrobina+difenoconazol, picoxistrobina, piraclostrobina+metconazol, trifloxistrobina+tebuconazol, azoxistrobina, boscalida+ piraclostrobina, iprodiona+pirimetanil e ciprodinil para pinta preta. Acibenzolar-s-metílico (ASM) reduziu a severidade da requeima e da pinta preta, porém, promoveu aumento da produtividade, apenas no campo de requeima. O fosfito de potássio (FP) também reduziu a severidade da requeima, no entanto, não influenciou na produtividade. Com exceção de mandipropamida, a adição de ASM à mefenoxam+mancozebe, cimoxanil+mancozebe e mancozebe promoveu aumento do controle da requeima, no entanto, apenas quando em mistura com mancozebe proporcionou aumento na produtividade. A adição de FP à mandipropamida, mefenoxam+mancozebe, cimoxanil+mancozebe, mancozebe e de ASM à azoxistrobina e difenoconazol não refletiu no controle da requeima, da pinta preta e na produtividade, respectivamente. Mancozebe e ASM não diferiram quanto severidade, progresso da pinta preta e produtividade, porém mancozebe+ASM foi superior ao ASM. A adição de ASM a programas de aplicação reduziu a requeima e a pinta preta e incrementou a produtividade apenas quando adicionados a programas onde prevaleceram fungicidas translaminares e de contato. O FP não influenciou nenhum dos programas testados para requeima. / Late blight, caused by the oomycete Phytophthora infestans and early blight, caused by the fungus Alternaria solani, are among the most destructive diseases of potato crops in Brazil and worldwide. Considering the importance of the chemical control on integrated systems and the need for detailed knowledge of the performance of resistance inducers and fungicides targeting the sustainability of production, the present study aimed to:1- characterize and compare the preemptive, residual, curative, antisporulative action and rain resistance of fungicides under controlled conditions; 2-: evaluate, under field conditions, the control potential of fungicides and resistance inductors and their effects on the total and commercial yield of potato tubers. All fungicides tested provided a high pre-emptive action against late and early blight. The systemic fungicides or high tenacity ones provided control of both diseases until 12 days after application, while translaminar ones until 9 days after application. As for the curative and anti-sporulative action, the systemic fungicides stand out when applied until 24 hours postinoculation, while the translaminar ones inhibited late blight when applied until 12 hours postinoculation. The contact fungicides stand out concerning pre-emptive action only. Similarly, systemic, translaminar and high tenacity fungicides were less affected by the simulated rain. An increase in drying time promoted higher retention and absorption of the products, so decreasing the negative precipitation impact. The better levels of control, yield and tubers quality were reached with the use of the following fungicides in the control of late blight: mandipropamid+chlorothalonil, fluopicolide+propamocarb, dimetomorph+ametoctradin, mandipropamid, fenamidone+ propamocarb, bentiavalicarb+fluazinam, followed by dimetomorph+chlorothalonil, mefenoxam+ chlorothalonil and famoxadone+cymoxanil+mancozeb. In the control of early blight the most efficient were: azoxystrobin+difenoconazole, pycoxystrobin, pyraclostrobin+metconazole, trifloxystrobin+tebuconazole, azoxystrobin, boscalid+pyraclostrobin, iprodione+pyrimethanil and cyprodinil. Acibenzolar-s-methyl (ASM) reduced the severity of late and early blight, but promoted an increase in the tuber yield only in late blight affected field. Potassium phosphite (Pp) also reduced the severity of late blight, although it did not have influenced the yield. Except for mandipropamid, the addition of ASM to mefenoxam+mancozebe, cymoxanil+mancozeb or mancozeb led to a better control of late blight, but only when mixed with mancozeb it promoted increase in the yield. Addition of Pp to mandipropamid, mefenoxam+mancozeb, cymoxanil+mancozeb, mancozeb and ASM to azoxystrobin, difenoconazole did not result either in the control of late and early blight or in an increase of yield. Mancozeb and ASM did not differ as for severity, progress of late blight and increasing yield, but however, mancozeb+ASM had a superior efficiency than ASM. Addition of ASM to application programs reduced late and early blight and increased yield only when added to programs in which prevailed translaminar and contact fungicides. Pp did not influenced any of the tested programs against late blight.

La gestion de Phytophthora infestans, agent du mildiou de la pomme de terre, nécessite l’application d’une quinzaine de traitements par saison culturale. Pour réduire l’usage des fongicides, combiner résistance induite et résistance quantitative pourrait être une bonne stratégie. Elle nécessite cependant une meilleure connaissance des interactions entre les réponses physiologiques de Solanum tuberosum et l’écologie de P. infestans. Dans cet objectif, les réponses de défense induites chez la pomme de terre ont été confrontées aux traits d’histoire de vie et effecteurs de P. infestans. Quatre génotypes présentant différents niveaux de résistance ont été traités avec un filtrat de culture concentré (CCF) de P. infestans induisant la PAMP-triggered immunity (PTI). Des folioles détachées ont ensuite été inoculées avec des souches rapides ou lentes de P. infestans. Les expressions de 14 gènes de défense et de 6 effecteurs ont été analysée simultanément par qRT-PCR.Les symptômes de la maladie ont été mesurés classiquement ou par analyse d’images dans le visible et en fluorescence. Les résultats obtenus montrent que la réduction des symptômes après induction de la PTI est fonction du couple génotype-souche. En effet, l’efficacité des défenses induites par le CCF dépend des stratégies d’échappement (vitesse de croissance) ou d’adaptation (effecteurs) de P. infestans et du potentiel d’inductibilité du génotype (expression des protéines PR). Ainsi, pour optimiser l’utilisation de la résistante induite il serait nécessaire de sélectionner des génotypes inductibles et capables de modu / The management of Phytophthora infestans, responsible for potato late blight, requires the application of about 15 fungicide treatments per cropping season. To reduce the use of pesticides, combining induced resistance and quantitative resistance could be a positive strategy. However, this method requires a better understanding of the interactions between Solanum tuberosum physiological responses and P. infestans ecology. To this end, defense responses induced in potato have been opposed to the pathogen life history traits and effectors. Four potato genotypes with different resistance levels were treated with a concentrated culture filtrate (CCF) of P. infestans inducing PAMPtriggered immunity (PTI). Then, detached leaflets were inoculated with fast- or slow-growing strains of P. infestans.The expression of 14 defense genes and the expression of 6 effectors were analyzed simultaneously by qRT-PCR. Disease symptoms were measured either conventionally or by visible and fluorescence image analysis. The results obtained show that the reduction of symptoms after induction of PTI was specific to the genotype-strain pair. Indeed, the effectiveness of induced defenses by CCF depends on either the escape (growth rate) or adaptation (effectors) strategies of P. infestans and on the genotype inductibility potential (expression of PR proteins). Thus, to optimize the use of induced resistance, it would be necessary to breed inducible genotypes that are able to modulate strains growth rate.

A requeima, causada pelo oomiceto Phytophthora infestans e a pinta preta, causada pelo fungo Alternaria solani, estão entre as doenças mais importantes e destrutivas da cultura da batata no Brasil e no mundo. Considerando a importância do controle químico em sistemas integrados e a necessidade de se conhecer detalhadamente o desempenho de fungicidas e indutores de resistência visando à sustentabilidade da produção, o presente estudo objetivou caracterizar e comparar a ação preventiva, residual, curativa, anti-esporulante e resistência à chuva de fungicidas em condições controladas, bem como, avaliar em campo o potencial de controle de fungicidas e indutores de resistência e seus reflexos sobre a produtividade total e comercial de tubérculos. Todos os fungicidas promoveram elevada ação preventiva contra a requeima e a pinta preta. Os fungicidas sistêmicos ou com alta tenacidade proporcionaram controle das duas doenças até os 12 DAP, enquanto que os fungicidas translaminares até os 9 DAP. Quanto à ação curativa e anti-esporulante destacaram-se principalmente os fungicidas sistêmicos aplicados até as 24 horas após a inoculação (HAI). Os fungicidas translaminares foram capazes de inibir a requeima quando aplicados até 12 HAI e os de contato destacaram-se apenas para ação preventiva. Semelhantemente, os fungicidas sistêmicos, translaminares ou com alta tenacidade foram os menos afetados pela chuva simulada. O aumento do tempo de secagem promoveu uma maior retenção ou absorção dos produtos reduzindo o impacto negativo da precipitação. Os melhores níveis de controle, produtividade e qualidade de tubérculos foram obtidos com os fungicidas mandipropamida+clorotalonil, fluopicolida+propamocarbe, dimetomorfe+ametoctradina, mandipropamida, fenamidona+propamocarbe, bentiavalicarbe+ fluazinam, seguidos de dimetomorfe+clorotalonil mefenoxam+clorotalonil e famoxadona+ cimoxanil+mancozebe para requeima e azoxistrobina+difenoconazol, picoxistrobina, piraclostrobina+metconazol, trifloxistrobina+tebuconazol, azoxistrobina, boscalida+ piraclostrobina, iprodiona+pirimetanil e ciprodinil para pinta preta. Acibenzolar-s-metílico (ASM) reduziu a severidade da requeima e da pinta preta, porém, promoveu aumento da produtividade, apenas no campo de requeima. O fosfito de potássio (FP) também reduziu a severidade da requeima, no entanto, não influenciou na produtividade. Com exceção de mandipropamida, a adição de ASM à mefenoxam+mancozebe, cimoxanil+mancozebe e mancozebe promoveu aumento do controle da requeima, no entanto, apenas quando em mistura com mancozebe proporcionou aumento na produtividade. A adição de FP à mandipropamida, mefenoxam+mancozebe, cimoxanil+mancozebe, mancozebe e de ASM à azoxistrobina e difenoconazol não refletiu no controle da requeima, da pinta preta e na produtividade, respectivamente. Mancozebe e ASM não diferiram quanto severidade, progresso da pinta preta e produtividade, porém mancozebe+ASM foi superior ao ASM. A adição de ASM a programas de aplicação reduziu a requeima e a pinta preta e incrementou a produtividade apenas quando adicionados a programas onde prevaleceram fungicidas translaminares e de contato. O FP não influenciou nenhum dos programas testados para requeima. / Late blight, caused by the oomycete Phytophthora infestans and early blight, caused by the fungus Alternaria solani, are among the most destructive diseases of potato crops in Brazil and worldwide. Considering the importance of the chemical control on integrated systems and the need for detailed knowledge of the performance of resistance inducers and fungicides targeting the sustainability of production, the present study aimed to:1- characterize and compare the preemptive, residual, curative, antisporulative action and rain resistance of fungicides under controlled conditions; 2-: evaluate, under field conditions, the control potential of fungicides and resistance inductors and their effects on the total and commercial yield of potato tubers. All fungicides tested provided a high pre-emptive action against late and early blight. The systemic fungicides or high tenacity ones provided control of both diseases until 12 days after application, while translaminar ones until 9 days after application. As for the curative and anti-sporulative action, the systemic fungicides stand out when applied until 24 hours postinoculation, while the translaminar ones inhibited late blight when applied until 12 hours postinoculation. The contact fungicides stand out concerning pre-emptive action only. Similarly, systemic, translaminar and high tenacity fungicides were less affected by the simulated rain. An increase in drying time promoted higher retention and absorption of the products, so decreasing the negative precipitation impact. The better levels of control, yield and tubers quality were reached with the use of the following fungicides in the control of late blight: mandipropamid+chlorothalonil, fluopicolide+propamocarb, dimetomorph+ametoctradin, mandipropamid, fenamidone+ propamocarb, bentiavalicarb+fluazinam, followed by dimetomorph+chlorothalonil, mefenoxam+ chlorothalonil and famoxadone+cymoxanil+mancozeb. In the control of early blight the most efficient were: azoxystrobin+difenoconazole, pycoxystrobin, pyraclostrobin+metconazole, trifloxystrobin+tebuconazole, azoxystrobin, boscalid+pyraclostrobin, iprodione+pyrimethanil and cyprodinil. Acibenzolar-s-methyl (ASM) reduced the severity of late and early blight, but promoted an increase in the tuber yield only in late blight affected field. Potassium phosphite (Pp) also reduced the severity of late blight, although it did not have influenced the yield. Except for mandipropamid, the addition of ASM to mefenoxam+mancozebe, cymoxanil+mancozeb or mancozeb led to a better control of late blight, but only when mixed with mancozeb it promoted increase in the yield. Addition of Pp to mandipropamid, mefenoxam+mancozeb, cymoxanil+mancozeb, mancozeb and ASM to azoxystrobin, difenoconazole did not result either in the control of late and early blight or in an increase of yield. Mancozeb and ASM did not differ as for severity, progress of late blight and increasing yield, but however, mancozeb+ASM had a superior efficiency than ASM. Addition of ASM to application programs reduced late and early blight and increased yield only when added to programs in which prevailed translaminar and contact fungicides. Pp did not influenced any of the tested programs against late blight.

One of the most frequently encountered types of data is where the response variables are measured on an ordinal scale. Although there have been substantial developments in the statistical techniques for the analysis of ordinal data, methods appropriate for repeatedly assessed ordinal data collected from field experiments are limited. A series of biennial field screening trials for evaluating cultivar resistance of potato to the disease, late blight, caused by the fungus Phytophthora infestans (Mont.) de Bary has been conducted by the New Zealand Institute of Crop and Food Research since 1983. In each trial, the progression of late blight was visually assessed several times during the planting season using a nine-point ordinal scale based on the percentage of necrotic tissues. As for many other agricultural field experiments, spatial differences between the experimental units is one of the major concerns in the analysis of data from the potato late blight trial. The aim of this thesis is to construct a statistical model which can be used to analyse the data collected from the series of potato late blight trials. We review existing methodologies for analysing ordinal data with mixed effects particularly those methods in the Bayesian framework. Using data collected from the potato late blight trials we develop a Bayesian hierarchical model for the analyses of repeatedly assessed ordinal scores with spatial effects, in particular the time dependence of the scores assessed on the same experimental units was modelled by a sigmoid logistic curve. Data collected from the potato late blight trials demonstrated the importance of spatial effects in agricultural field trials. These effects cannot be neglected when analysing such data. Although statistical methods can be refined to account for the complexity of the data, appropriate trial design still plays a central role in field experiments. / Accompanying dataset is at http://hdl.handle.net/2292/5240

Phytophthora infestans (Mont) de Bary is a devastating pathogen in potato producing regions around the world. Populations of the organism in Canada shifted during the mid-1990’s as the US-1 strain (A1, metalaxyl-sensitive) was displaced by the highly aggressive, US-8 strain (A2, metalaxyl-insensitive). An increase in the incidence and severity of late blight has followed. Late blight is controlled by cultural practices aimed at eliminating disease sources and by the application of foliar fungicides. Tubers can become infected at harvest from contact with blighted vines leading to severe losses in storage. In many production areas, growers desiccate vines two to three weeks prior to harvest to reduce late blight tuber rot. However, in Manitoba, because of the loss of potential yield that results from vine killing prior to harvest in a late maturing cultivar such as Russet Burbank, growers are reluctant to adopt this practice. The objective of this study was to develop recommendations for preharvest management practices that reduce storage losses due to late blight. Field trials were conducted in 1997 and 1998 to investigate the effect of vine desiccation with diquat and/or a late season application of chlorothalonil and copper hydroxide on yield, processing quality, and disease development in storage. Desiccating vines with diquat two weeks prior to harvest reduced yield and tuber size. Compared to the untreated control, the largest reductions in marketable yield were observed for the early September harvest. By the late September harvest, however, the effect of vine killing in reducing marketable yield was less apparent. Specific gravity was lower in the vine killed treatment for all harvest dates in 1997 and in the early and mid September harvests in 1998. Vine killing did not contribute to elevated levels of reducing sugars or consistently darker fry colour at harvest or during storage. Skin-set was improved when vines were desiccated for all harvest dates in 1997 and at the early September harvest date in 1998. Vine killing reduced tuber rot in storage caused by Fusarium dry rot and Pythium leak for the early and mid September harvest dates in 1997. The incidence of late blight tuber rot was reduced in storage for the early September harvest in 1998 when vines were desiccated. The late-season application of chlorothalonil and copper hydroxide did not reduce tuber rot in storage in storage either year. Results from this study indicate that vine killing two weeks before an early to mid- September harvest is not recommended in Manitoba because of reductions in yield and specific gravity. Alternative management practices to reduce late blight tuber rot in storage should be investigated. / February 2005

Phytophthora infestans (Mont) de Bary is a devastating pathogen in potato producing regions around the world. Populations of the organism in Canada shifted during the mid-1990’s as the US-1 strain (A1, metalaxyl-sensitive) was displaced by the highly aggressive, US-8 strain (A2, metalaxyl-insensitive). An increase in the incidence and severity of late blight has followed. Late blight is controlled by cultural practices aimed at eliminating disease sources and by the application of foliar fungicides. Tubers can become infected at harvest from contact with blighted vines leading to severe losses in storage. In many production areas, growers desiccate vines two to three weeks prior to harvest to reduce late blight tuber rot. However, in Manitoba, because of the loss of potential yield that results from vine killing prior to harvest in a late maturing cultivar such as Russet Burbank, growers are reluctant to adopt this practice. The objective of this study was to develop recommendations for preharvest management practices that reduce storage losses due to late blight. Field trials were conducted in 1997 and 1998 to investigate the effect of vine desiccation with diquat and/or a late season application of chlorothalonil and copper hydroxide on yield, processing quality, and disease development in storage. Desiccating vines with diquat two weeks prior to harvest reduced yield and tuber size. Compared to the untreated control, the largest reductions in marketable yield were observed for the early September harvest. By the late September harvest, however, the effect of vine killing in reducing marketable yield was less apparent. Specific gravity was lower in the vine killed treatment for all harvest dates in 1997 and in the early and mid September harvests in 1998. Vine killing did not contribute to elevated levels of reducing sugars or consistently darker fry colour at harvest or during storage. Skin-set was improved when vines were desiccated for all harvest dates in 1997 and at the early September harvest date in 1998. Vine killing reduced tuber rot in storage caused by Fusarium dry rot and Pythium leak for the early and mid September harvest dates in 1997. The incidence of late blight tuber rot was reduced in storage for the early September harvest in 1998 when vines were desiccated. The late-season application of chlorothalonil and copper hydroxide did not reduce tuber rot in storage in storage either year. Results from this study indicate that vine killing two weeks before an early to mid- September harvest is not recommended in Manitoba because of reductions in yield and specific gravity. Alternative management practices to reduce late blight tuber rot in storage should be investigated.

Phytophthora infestans (Mont) de Bary is a devastating pathogen in potato producing regions around the world. Populations of the organism in Canada shifted during the mid-1990’s as the US-1 strain (A1, metalaxyl-sensitive) was displaced by the highly aggressive, US-8 strain (A2, metalaxyl-insensitive). An increase in the incidence and severity of late blight has followed. Late blight is controlled by cultural practices aimed at eliminating disease sources and by the application of foliar fungicides. Tubers can become infected at harvest from contact with blighted vines leading to severe losses in storage. In many production areas, growers desiccate vines two to three weeks prior to harvest to reduce late blight tuber rot. However, in Manitoba, because of the loss of potential yield that results from vine killing prior to harvest in a late maturing cultivar such as Russet Burbank, growers are reluctant to adopt this practice. The objective of this study was to develop recommendations for preharvest management practices that reduce storage losses due to late blight. Field trials were conducted in 1997 and 1998 to investigate the effect of vine desiccation with diquat and/or a late season application of chlorothalonil and copper hydroxide on yield, processing quality, and disease development in storage. Desiccating vines with diquat two weeks prior to harvest reduced yield and tuber size. Compared to the untreated control, the largest reductions in marketable yield were observed for the early September harvest. By the late September harvest, however, the effect of vine killing in reducing marketable yield was less apparent. Specific gravity was lower in the vine killed treatment for all harvest dates in 1997 and in the early and mid September harvests in 1998. Vine killing did not contribute to elevated levels of reducing sugars or consistently darker fry colour at harvest or during storage. Skin-set was improved when vines were desiccated for all harvest dates in 1997 and at the early September harvest date in 1998. Vine killing reduced tuber rot in storage caused by Fusarium dry rot and Pythium leak for the early and mid September harvest dates in 1997. The incidence of late blight tuber rot was reduced in storage for the early September harvest in 1998 when vines were desiccated. The late-season application of chlorothalonil and copper hydroxide did not reduce tuber rot in storage in storage either year. Results from this study indicate that vine killing two weeks before an early to mid- September harvest is not recommended in Manitoba because of reductions in yield and specific gravity. Alternative management practices to reduce late blight tuber rot in storage should be investigated.

One of the most frequently encountered types of data is where the response variables are measured on an ordinal scale. Although there have been substantial developments in the statistical techniques for the analysis of ordinal data, methods appropriate for repeatedly assessed ordinal data collected from field experiments are limited. A series of biennial field screening trials for evaluating cultivar resistance of potato to the disease, late blight, caused by the fungus Phytophthora infestans (Mont.) de Bary has been conducted by the New Zealand Institute of Crop and Food Research since 1983. In each trial, the progression of late blight was visually assessed several times during the planting season using a nine-point ordinal scale based on the percentage of necrotic tissues. As for many other agricultural field experiments, spatial differences between the experimental units is one of the major concerns in the analysis of data from the potato late blight trial. The aim of this thesis is to construct a statistical model which can be used to analyse the data collected from the series of potato late blight trials. We review existing methodologies for analysing ordinal data with mixed effects particularly those methods in the Bayesian framework. Using data collected from the potato late blight trials we develop a Bayesian hierarchical model for the analyses of repeatedly assessed ordinal scores with spatial effects, in particular the time dependence of the scores assessed on the same experimental units was modelled by a sigmoid logistic curve. Data collected from the potato late blight trials demonstrated the importance of spatial effects in agricultural field trials. These effects cannot be neglected when analysing such data. Although statistical methods can be refined to account for the complexity of the data, appropriate trial design still plays a central role in field experiments. / Accompanying dataset is at http://hdl.handle.net/2292/5240

One of the most frequently encountered types of data is where the response variables are measured on an ordinal scale. Although there have been substantial developments in the statistical techniques for the analysis of ordinal data, methods appropriate for repeatedly assessed ordinal data collected from field experiments are limited. A series of biennial field screening trials for evaluating cultivar resistance of potato to the disease, late blight, caused by the fungus Phytophthora infestans (Mont.) de Bary has been conducted by the New Zealand Institute of Crop and Food Research since 1983. In each trial, the progression of late blight was visually assessed several times during the planting season using a nine-point ordinal scale based on the percentage of necrotic tissues. As for many other agricultural field experiments, spatial differences between the experimental units is one of the major concerns in the analysis of data from the potato late blight trial. The aim of this thesis is to construct a statistical model which can be used to analyse the data collected from the series of potato late blight trials. We review existing methodologies for analysing ordinal data with mixed effects particularly those methods in the Bayesian framework. Using data collected from the potato late blight trials we develop a Bayesian hierarchical model for the analyses of repeatedly assessed ordinal scores with spatial effects, in particular the time dependence of the scores assessed on the same experimental units was modelled by a sigmoid logistic curve. Data collected from the potato late blight trials demonstrated the importance of spatial effects in agricultural field trials. These effects cannot be neglected when analysing such data. Although statistical methods can be refined to account for the complexity of the data, appropriate trial design still plays a central role in field experiments. / Accompanying dataset is at http://hdl.handle.net/2292/5240