No ambiente agrícola, a possibilidade de prever eventos futuros para poder estabelecer prioridades e planejar atividades são indispensáveis para um manejo adequado. Modelos matemáticos têm se tornado ferramentas valiosas para o entendimento de fenômenos e simulação de soluções de um dado sistema de interesse para diferentes condições iniciais e valores de parâmetros. O crescimento das plantas obedece a certos princípios fisiológicos que podem ser descritos, em termos quantitativos em resposta ao meio ambiente, através de equações matemáticas. Nos agrosistemas, a dinâmica da população de plantas daninhas pode ser descrita por modelos matemáticos que relacionam as densidades de sementes produzidas e de plântulas em áreas de cultivo. Os valores dos parâmetros dos modelos podem ser inferidos diretamente de experimentação e análise estatística, ou extraídos da literatura. O presente trabalho tem por objetivo investigar as particularidades dos parâmetros de modelos dinâmicos para populações de plantas daninhas, a partir de um experimento conduzido em campo, usando inferência Bayesiana via método de Monte Carlo com cadeias de Markov e analisar situações que podem alterar a dinâmica do comportamento populacional por meio de simulações. / In the agricultural environment, the possibility to predict future events to establish priorities and to plan activities is indispensable for an appropriate management. Mathematical models have become precious tools for the understanding of phenomena and simulation of solutions of a given system for different initial conditions and values of parameters. The growth of plants obeys the certain physiological principles that can be described, in quantitative terms in reply to the environment, through mathematical equations. In agrosystems, the dynamics of weed populations can be described by mathematical models that relates the produced seeds density and seedlings density in areas of a crop. The parameter models can be either directly inferred from experimentation and statistics analysis, or can be extracted from literature. The goals of this work is to investigate the particularitities of the dynamic models parameters for weed populations, from field experiment, using Bayesian inference by Monte Carlo method with Markov chains and to analyze situations that can modify the population behavior by simulations.

O presente trabalho teve o objetivo de avaliar os efeitos da vegetação intercalar composta por gramíneas e leguminosas perenes e da cobertura morta, obtida pelo manejo mecânico da vegetação intercalar, e herbicidas aplicados em pós-emergência, na linha da cultura, sobre a população de plantas daninhas, características químicas do solo, estado nutricional, desenvolvimento vegetativo e produção de citros. O experimento foi conduzido, durante as safras 2004/2005 e 2005/2006, na Fazenda São Judas, Lucianópolis - SP, em um pomar de laranja Pêra (Citrus sinensis (L.) Osbeck.), plantado em outubro de 2002. O delineamento utilizado foi o de blocos ao acaso, com parcelas subdivididas sendo 7 tratamentos e 2 subtratamentos, com 4 repetições. Os tratamentos foram: 1) vegetação intercalar de Brachiaria ruziziensis R. Germ. & Evrard, sem cobertura morta na linha da cultura; 2) vegetação intercalar de B. ruziziensis, com cobertura morta1 na linha da cultura; 3) vegetação intercalar de B. ruziziensis e Arachis pintoi Krap. & Greg com cobertura morta1 na linha da cultura; 4) vegetação intercalar de B. ruziziensis e Stylosantes spp. com cobertura morta1 na linha da cultura; 5) vegetação intercalar composta por Brachiaria plantagineacom cobertura morta1 na linha da cultura 6) vegetação intercalar de Arachis pintoi com cobertura morta1 na linha da cultura; 7) vegetação intercalar de estilosantes \'mineirão\' (Stylosathes macrocephala (25%), S. capitata (60%) e S. guianensis (15%)) com cobertura morta1 na linha da cultura. Os herbicidas aplicados às subparcelas foram: 1) glyphosate a 1260 g e. a. ha-1 e 2) paraquat a 400 g i. a. ha-1. Pelos dados obtidos pode-se observar que: a vegetação intercalar dos tratamentos 2, 3 e 4 obtiveram as maiores produções de fitomassa e cobertura morta, seguido pelos tratamentos 5 e 7 e a menor produção de fitomassa foi obtida pelo tratamento 6. A cobertura morta de B. ruziziensis proporcionou menor densidade de plantas daninhas por até 90 dias. O teor de matéria orgânica, nas entrelinhas, na camada de 0-0,10 m, nos tratamentos 2 e 4 foi superior em 3,79 e 3,48 g cm-3 em relação ao tratamento 6, enquanto que na camada de 0,20-0,40 m, o teor de matéria orgânica, do tratamento 6, foi superior em 4,6, 5,1 e 3,9 g cm-3 em relação aos tratamentos 1, 3 e 4, respectivamente. Nas linhas da cultura, o tratamento 4 e 3 apresentaram teor de K superior em 1,28 e 0,95 mmolc dm-3, respectivamente, em relação ao tratamento 6, nas camadas de 0-0,10 e 0,10-0,20 m de solo. O estilosantes proporcionou incrementos no teor foliar de N de 2,58 e 2,55 g kg-1, em relação aos tratamentos 1 e 2. O teor foliar de P do tratamento 4 foi superior ao tratamento 1 em 0,31 g kg-1, sendo observada uma correlação entre a densidade do sistema radicular na camada de 0-0,10 m de solo e o teor foliar de P. Não foi verificado efeitos dos tratamentos e herbicidas no desenvolvimento vegetativo, bem como na produção total no período avaliado. / The objective of this research was to evaluate the effects of cover crops from grass and perennial leguminous in the orchard inter-rows, mulch, produced through mechanical management with a lateral rotary mower from inter-row cover crops associated with herbicides applied in pos-emergence, on the rows, on population dynamics of weeds, soil chemical characteristics and development, nutrition and citros production. The experiment was carried out at the São Judas farm, Lucianópolis - SP - Brazil during the harvest of 2004/2005 and 2005/2006 in a citrus orchard of Citrus sinensis L. Osbeck cv. Pera, planted in October, 2002. The experimental design used was of randomized blocks with subdivided plots and 4 replications. The treatments were; 1) Cover crop of Brachiaria ruziziensis R. Germ. & Evrard, without mulch on the row; 2) Cover crop of B. ruziziensis and mulch on the row, produced through mechanical management with a lateral rotary mower (REL) from this cover crop; 3) Cover crop of B. ruziziensis and Arachis pintoi Krap. & Greg and mulch on the row, produced through REL from this cover crop; 4) Cover crop of B. ruziziensis and estilosantes \"mineirão\" (Stylosathes macrocephala (25%), S. capitata (60%) e S. guianensis (15%)) and mulch on the row, produced through REL from this cover crop; 5) Cover crop of Brachiaria plantaginea and mulch on the row, produced through REL from this cover crop; 6) Cover crop of Arachis pintoi and mulch on the row, produced through REL from this cover crop; 7) Cover crop of estilosantes \"mineirão\" and mulch on the row, produced through REL from this cover crop.The herbicides applied into the subplots were: 1) glyphosate at the rate 1260 g a.e. ha-1 and 2) paraquat at the rate 400 a.i. ha-1. In these experiments, it could be observed that: the cover crops from the treatments 2, 3 and 4 got higher wet biomass and mulch than others treatments; B. ruziziensis mulch decrease the weeds density until 90 days after the distribution; in the inter-row, into 0-0,10 m soil layer, the treatments 2 and 4 organic matter, was higher than treatment 6, however, into 0,20-0,40 m soil layer, treatment 6 organic matter was higher than treatments 1, 3 and 4. Into the rows, treatments 4 and 3 achieved K tenor higher than treatment 6 in 1,28 and 0,95 mmolc dm-3, respectively, into 0-0,10 and 0,10-0,20 m soil layers. Estilosantes promoted N foliar increases of 2,58 and 2,55 g kg-1 regarding treatments 1 and 2. Treatment 4 P foliar tenor was higher than treatment 1 in 0,31 g kg-1. It was observed a correlation between root density into superficial layer soil and P foliar tenor. It was not verified any effect of treatment or herbicides on development and total citros production.

A identificação de plantas invasoras é de extrema importância em diversos procedimentos utilizados na agricultura. Apesar de ser uma tarefa computacionalmente difícil, esta identificação tem se tornado muito importante no contexto da agricultura de precisão. A agricultura de precisão substitui os tratos culturais de grandes áreas da cultura, feitos pela média do nível dos problemas encontrados nessas áreas, por tratamento específicos e pontuais. As pricipais vantagens são o aumento de produtividade, relacionado com a diminuição da variabilidade na produção, a economia de insumos e a preservação do meio ambiente. Este trabalho enfoca o reconhecimento de plantas invasoras em tempo real. Para manter o requisito de tempo real, são utilizadas redes neurais artificiais como meio para o reconhecimento de padrões. Entre as diversas plantas invasoras de ocorrência freqüente no cerrado brasileiro, foi selecionado o picão preto para a avaliação das técnicas adotadas. Uma arquitetura modular de reconhecimento é proposta, com o uso de processamento paralelo, facilitando a inclusão de módulos de reconhecimento de outras plantas invasoras sem a deterioração do desempenho do sistema. Os resultados obtidos são amplamente satisfatórios, demonstrando a possibilidade do desenvolvimento de um sistema embarcado completo de identificação de plantas invasoras em tempo real. Este sistema, apoiado pelo sistema de posicionamento global GPS, pode servir de base para uma série de máquinas agrícolas inteligentes, como pulverizadores de herbicidas e outros defensivos utilizados na agricultura. / Weed identification is an important task in many agricultural procedures. In spite of being a computation intensive task, this identification is very important in the role of precision agriculture. Conventional procedures in agriculture are based on the average level of the problems found in large areas. Precision agriculture introduces new punctual management procedures, dealing with very small areas. The main advantages are: productivity increase, related with the decrease in production unevenness, economy and environment preservation. This work focuses on the real time recognition of weeds. To maintain the real time requirement, neural networks are used to carry out the recognition of image patterns. Among the several weeds frequently found in the Brazilian savannah, the "picão preto" was selected for the evaluation of the adopted techniques. A modular architecture is proposed, using parallel processing, making easier the use of new recognition modules (for other weeds), still preserving the real time capabilities of the system. Results obtained are thoroughly adequate, demonstrating the possibility of the development of embedded systems for the identification of several weeds in real time. These systems, jointly with the global positioning system (GPS), can be used in a family of intelligent equipment, such as spraying machines for herbicides and other agricultural products.

Les écosystèmes littoraux lacustres, ou ceintures des lacs, ont une valeur écologique très importante. Zones de transition et d'échanges entre les écosystèmes terrestre et aquatique, les rives des lacs et leurs zones littorales sont reconnues pour abriter une forte biodiversité, dont des plantes aquatiques visibles à l'œil nu : les macrophytes. La structure spatiale, l'abondance et la composition spécifique de ces peuplements végétaux sont conditionnées par le caractère et le fonctionnement physiques des biotopes littoraux lacustres. Ce travail de thèse est dédié à l'acquisition de connaissances sur les relations entre les peuplements macrophytiques et les déterminants physiques de biotopes littoraux lacustres dans les lacs peu profonds. Il s'applique à définir, à différentes échelles, les préférences hydroécologiques et la structure fonctionnelle et spatiale de communautés et populations de plantes aquatiques. Cette recherche s'appuie sur la modélisation et sur des observations, mesures et prélèvements menés in situ sur les macrophytes et les variables environnementales dans les grands lacs naturels du littoral aquitain. Les principaux résultats indiquent que l'action du vent et des vagues, la pente des rives, la profondeur, la granulométrie et la matière organique des sédiments influencent l'organisation spatiale et les assemblages d'espèces de macrophytes. Ces variables peuvent être utilisées pour prédire la distribution potentielle de certaines communautés végétales. Les altérations anthropiques de l'hydromorphologie favorisent le développement des espèces exotiques tout en compromettant le maintien des espèces indigènes patrimoniales. Les enjeux liés à la gestion des espèces exotiques et des espèces indigènes ont également été identifiés à partir d'une analyse diachronique des dynamiques écologiques de la végétation depuis quarante ans dans les lacs aquitains. Il est désormais urgent d'établir une stratégie de conservation des espèces indigènes, voire de restauration pour les patrimoniales, et de limiter la dispersion des espèces exotiques dans les lacs. Présentant de nombreuses analogies avec les lacs d'Europe du Nord tels que les softwater lakes ou les Lobelia lakes, les processus écologiques en cours dans les lacs du sud-ouest de la France pourraient préfigurer la situation à venir pour l'ensemble des lacs du même type dans le contexte du changement global. Les écosystèmes littoraux lacustres concentrent des enjeux écologiques, économiques et sociétaux, soulevant des questions et problématiques scientifiques sur leur fonctionnement et sur la bioindication de leur état écologique qu'il est nécessaire de continuer à explorer. / Lakeshores zones are of very high ecological value. As they are ecotones between terrestrial and aquatic ecosystems, lake shoreline and littoral zones are characterized by a high level of biodiversity, including macroscopic aquatic plants known as macrophytes. Plant distribution, abundance and species composition are determined by physical features of lakeshores. This work aims to assess relationships between macrophyte populations and communities and hydromorphological components in shallow lakes at different scales. Plant and environmental variables were modeled or collected on the field in large lakes located along the western French Atlantic coast. Wind and wave action, slope, depth, sediment grain size and organic matter content are well known for their influence on macrophyte species distribution and assemblages. Physical variables can be used to predict the potential distribution of plant communities. Anthropogenic physical disturbances enhances the development of invasive species while they threaten rare and endangered native species. Plant management issues were also highlighted from a long-term field data (fourty years) analysis of macrophyte dynamics in these shallow lakes. Management strategies are needed for the conservation of native species, or even their restoration, and the control of alien species spread in these lakes. Similar in some ways to northern European lakes such as softwater lakes or Lobelia lakes, the ecological processes identified in these Atlantic shallow lakes could foreshadow the future situation for all lakes of the same type in the context of global change. Lakeshore ecosystems gather ecological, economic and societal issues, raising scientific questions and issues about their functioning and the bioindication of their ecological status that needs to be further explored.

Plants regulate responses to their environment through complex hormone signaling; these hormones can be categorized broadly into two categories: growth and defense, though many have roles in both. Much remains to be understood about the complexity of hormone signaling in relation to environmental responses, especially species- and genotype-specific differences. Unraveling this complexity of hormone signaling will lead to the development of resilient crops that are able to respond appropriately to their environment. In this dissertation, I hypothesize novel roles for growth and defense hormones in <i>Solanum </i>spp. responses to 1) biochar, a black carbon soil amendment (Chapter 2), 2) infection with<i> Ralstonia solanacearum</i>, an economically important soilborne pathogen causing bacterial wilt (Chapter 3), and 3) endophytic colonization by the soil bacterial community (Chapter 4). In Chapter 2, I showed that biochar upregulates GA signaling and affects GA-related traits in a species- and cultivar-specific manner. Biochar amendment also downregulates defense signaling. In Chapter 3, I demonstrated a novel role for auxin in resistance against <i>R. solanacearum, </i>including differential expression of auxin signaling genes in resistant genotype H7996 compared to susceptible WV in response to <i>R. solanacearum</i> infection. In addition, I observed stronger and faster upregulation of defense hormone marker genes for SA and ET in H7996 compared to WV. In Chapter 4, I showed that SA and ET are required for normal tomato root microbial community assembly, affecting the colonization of a few key taxa in order to promote alpha diversity. H7996 and WV root communities differ in alpha diversity, and a panel of H7996 x WV RILs showed quantitative variation in alpha diversity that correlated negatively with the abundance of these key taxa. In conclusion, I elucidated novel roles for hormones in responses to the soil environment, pathogen infection, and root community colonization. These findings are important for developing resilient, sustainable crops.

Onion thrips (Thrips tabaci) and Iris Yellow Spot Virus (IYSV) constitute a pest complex of global concern for allium growers. Yield losses due to onion thrips alone can be heavy, and in tandem with IYSV, these losses can be complete. This study was one phase of a multi-phase research project intent on exploring the potential drivers of thrips and virus infestation in onion. Preliminary observations in participating growers’ fields in northern Utah pointed towards a potential link between weedy field borders and thrips/IYSV infestation in onion. Field work was carried out at the Kaysville experiment station examining the possible relationships between thrips and IYSV incidence in onion and the presence of adjacent weed populations. This research was conducted with the intent of guiding growers’ decisions regarding field border management, and to provide additional insights into onion thrips behavior. While implications for growers remain unclear, thrips preferences regarding host plant utilization were identified, as well as preliminary evidence of additional plant species capable of hosting IYSV.

Un projet visant à identifier des mycoherbicides pour lutter contre les adventices a été initié entre l’UMR Agroécologie de Dijon et la société DE SANGOSSE® (Agen). Trois volets ont structuré ce projet à l’issue d’une collecte de prélèvement de 475 plantes représentatives de 23 espèces d’adventices symptomatiques et asymptomatiques en Bourgogne et en Beauce. Le 1er volet reposait sur une approche de type metabarcoding (technologie Illumina), pour évaluer et comparer la diversité des communautés fongiques endophytes des plantes symptomatiques et asymptomatiques. 542 genres fongiques ont ainsi été identifiés. Des taxons associés aux plantes symptomatiques ont été identifiés. Parmi ceux-ci, certains sont des pathogènes connus, d’autres non et ils constituent des pistes à exploiter pour la recherche de candidats mycoherbicides. Le deuxième volet repose sur une approche conventionnelle de microbiologie et pathologie. Une collection de 194 champignons associés aux symptômes des adventices a été constituée. La pathogénicité de ces isolats a été testée grâce à une série de screenings de plus en plus sélectifs qui ont abouti à la sélection de cinq souches, identifiées par séquençage de l’ITS ou d’autres marqueurs taxonomiques. Une souche appartient à l’espèce Boeremia exigua var exigua, une autre à l’espèce Alternaria alternata, deux appartiennent à l’espèce A. penicillata et la dernière au genre Alternaria. Le troisième volet visait à identifier le mode d’action d’une souche par une double approche, métabolomique et microscopique. La souche de B. exigua var exigua secrète des métabolites phytotoxiques mais également infeste et semble détruire les tissus végétaux sous-épidermique de la plante hôte.Ce projet exploratoire a fourni des pistes de taxons fongiques associés à des symptômes observés sur adventices en analysant la diversité par une approche moléculaire et a fourni des souches fongiques, mycoherbicides potentiels, par une approche microbiologique dont on voit bien qu’elle reste une méthode incontournable, malgré ses limites, pour obtenir des candidats fongiques à action herbicide. / A project aiming at identifying mycoherbicides to control weeds has been initiated between the UMR Agroécologie (Dijon) and the company DE SANGOSSE® (Agen, France). Three axes structured this project after a sampling collection of 475 plants representative of 23 species of symptomatic and asymptomatic weeds was carried out in Burgundy and Beauce. The first part was based on a metabarcoding approach (Illumina technology), to evaluate end compare the diversity of endophytic fungi communities of symptomatic and asymptomatic weeds. 542 fungal genera have been identified. Taxa associated with symptomatic plants have been identified. Of these, some are known pathogens, others are not, and both constitute avenues to exploit for the research of mycoherbicide candidates. The second axe is based on a conventional approach to microbiology and pathology. A collection of 194 fungi associated with weed symptoms was established. The pathogenicity of these isolates was tested through a series of increasingly selective screenings that resulted in the selection of five strains that were identified by sequencing of ITS or other taxonomic markers. One strain belongs to the species Boeremia exigua var exigua, another species Alternaria alternata, two belong to the species A. penicillata and the last to the genus Alternaria. The third axe aimed at identifying the mode of action of a strain by a dual metabolomics and microscopic approach. The strain of B. exigua var exigua produced phytotoxic secondary metabolites but also infested and apparently destroyed the sub-epidermal plant tissues of the host plant.This exploratory project provided tracks to exploit fungal taxa associated with observed weeds symptoms, by analyzing the diversity, by a molecular approach and provided fungal strains, potential mycoherbicides by a conventional microbiological approach that we can see it remains an unavoidable method, despite its limitations, to obtain fungal candidates with herbicidal action.

The naturalisation and subsequent spread of non-indigenous plant species (NIPS) is a major problem for most regions of the world. Managing plant invasions requires greater understanding of factors that determine initial naturalisation and distribution of wild NIPS. By the year 2000, 2252 NIPS were recorded as wild (1773 fully naturalised and 479 casual) in New Zealand. From published literature and electronic herbaria records, I recorded year of discovery of wild populations, and regional distribution of these wild NIPS. I also recorded species related attributes hypothesised to affect naturalisation and/or distribution, including global trade, human activities, native range and biological data; and regional attributes hypothesised to affect distribution, including human population densities, land use/cover, and environmental data. I used interval-censored time-to-event analyses to estimate year of naturalisation from discovery records, then analysed the importance of historical, human activity, biogeographical and biological attributes in determining patterns of naturalisation. Typically, NIPS that naturalised earlier were herbaceous, utilitarian species that were also accidentally introduced and/or distributed, with a wide native range that included Eurasia, naturalised elsewhere, with a native congener in New Zealand. In the year 2000, 28% of wild NIPS occupied only one region, 18% occupied two regions, decreasing incrementally to 2.5 % for nine regions, but with 13.5% occupying all ten regions. I used generalised linear models (GLMs) with binomial distribution to determine predictors of whether a wild NIPS occupied ten regions or not, and GLMs with Poisson distribution for wild NIPS occupying 0 – 9 regions. As expected, the dominant effect was that species discovered earlier occupied more regions. Utilitarian wild NIPS that were also accidentally introduced and/or distributed, and wild NIPS with a native congener tended to be more widely distributed, but results for other attributes varied between datasets. Although numbers of wild NIPS recorded in regions of New Zealand were sometimes similar, composition of wild NIPS was often very different. I used nonmetric multidimensional scaling (NMDS) to determine dissimilarity in composition between regions. Then, after reducing correlation between predictor variables using principal components analyses (PCAs), I tested the importance of regional variables in determining the regional composition of wild NIPS using metaMDS. The density of human populations best explained the dissimilarity in composition, but temperature gradients and water availability gradients were also important. In the year 2000 more than 1100 (60%) of the 1773 fully naturalised NIPS in mainland New Zealand had each been recorded in Northland/Auckland and Canterbury, and at the other end of the scale, Southland and Westland each had fewer than 500 (30%). I used GLMs to analyse the importance of people and environment in determining the numbers of wild NIPS in each region. Because I conducted multiple tests on the same dataset I used sequential Bonferroni procedures to adjust the critical P-value. Only human population density was important in explaining the numbers of NIPS in the regions. Overall, humans were the dominant drivers in determining the patterns of naturalisation and spread, although environment helps determine the composition of NIPS in regions. Incorporating human associated factors into studies of wild NIPS helps improve the understanding of the stages in the naturalisation and spread process.

El objetivo general de esta tesis fue producir información científico-tecnológica básica para la implementación de un sistema integrado de control de malezas, que permita reemplazar al sistema mecánico en las plantaciones de álamo (Populus spp.) del centro-norte de la Provincia de Buenos Aires. Los trabajos se agruparon en tres líneas de investigación complementarias, en donde se abordaron grupos específicos de problemas. En la primera línea de trabajo se determinó la composición específica de la comunidad de malezas y la magnitud de su interferencia, se establecieron los niveles de control y fitotoxicidad que producen diferentes herbicidas, y se evaluaron en forma comparada los efectos sobre la supervivencia y el crecimiento del cultivo de distintos tratamientos integrados de control respecto del sistema actual. La segunda línea estudió la variación temporal de la interferencia en busca de la definición del período crítico para el control de malezas, y en la tercera línea de trabajo se evaluó el resultado económico potencial de la implementación de los cambios estudiados en el sistema de control, considerando diferentes escenarios económico-financieros, grados de incorporación de tecnología y de reemplazo de mano de obra. Las distintas líneas comprendieron experiencias en campo llevadas a cabo entre los años 1998 y 2004 en el Establecimiento Forestal "María Dolores", Partido de Alberti, Provincia de Buenos Aires (34° 50' Lat. Sur; 60° 30' Long. Oeste; 55 msnm). Los trabajos se efectuaron siguiendo una estrategia progresiva de experimentación: en una primera etapa se realizaron ensayos exploratorios y posteriormente los resultados iniciales obtenidos fueron evaluados en ensayos a escala de producción y parcelas demostrativas de 5 hectáreas de superficie.

Yellow toadflax, Linaria vulgaris (L.) Mill. (Plantaginaceae), is a non-native invasive plant. Rhinusa pilosa Germar (Coleoptera: Curculionidae) is a proposed biocontrol agent. Gall development by R. pilosa was described using histological methods and compared between plant populations from native and introduced ranges. Key stages of oviposition were isolated histologically to determine their importance in gall induction. Rhinusa pilosa galled and developed on four geographically distinct Canadian populations in a pre-release quarantine study. Low agent densities only negatively affected one population. High densities of R. pilosa reduced potential reproductive output and plant biomass. Conducting detailed investigations into the biology, impact, and development of R. pilosa on populations from invasive and native ranges may help predict the efficacy of R. pilosa in the field if approved for release and.goes beyond current pre-release testing requirements. / ix, 168 leaves : ill. (chiefly col.) ; 29 cm