Haiti Peanut Research Report

Haiti Peanut Research Report Efforts to improve peanut production in Haiti by investigating management options for foliar diseases, low soil fertility, and other yield-limiting agronomic issues Feed the Future Innovation Lab for Collaborative Research on Peanut Productivity and Mycotoxin Control (Peanut & Mycotoxin Innovation Lab) Haiti Peanut Research Report Summary Report of 2015 to 2017 Data Edited by Abraham Fulmer Developed by Haiti Peanut Value Chain Intervention Project C1. Production to Consumption: Technologies to Improve Peanut Production, Processing and Utilization in Haiti http://ftfpeanutlab.caes.uga.edu/ Authors Name Institution Country Title Affiliation Greg MacDonald University of Florida USA Principal Investigator Agronomy Timothy Brenneman University of Georgia USA Co-Principal Investigator Plant Pathology Robert Kemerait University of Georgia USA Co-Principal Investigator Plant Pathology Jamie Rhoads University of Georgia USA Assistant Director of PMIL Crop and Soil Sciences Abraham Fulmer University of Georgia USA Graduate Research Assistant Plant Pathology Alex Carroll Meds and Food for Kids Haiti Agronomist Partner Rick Macajoux Meds and Food for Kids Haiti Agronomist Partner Georgy Faroutine Meds and Food for Kids Haiti Agronomist Partner Will Sheard Meds and Food for Kids Haiti Agronomist Partner Dominique Laforest Meds and Food for Kids Haiti Agronomist Partner Patrick Dorzin Acceso Peanut Enterprise Haiti Agronomist Partner Table of Contents ACKNOWLEDGEMENTS.................................................................................................4 CHAPTER 1. Introduction 1.1 Peanut Production in Haiti................................................................................5 1.2 PMIL Project.................................................................................................11 1.3 Recommended Best Practices for Peanut Production in Haiti. .....................12 1.4 Lessons Learned............................................................................................14 1.5 Research Field Sites.......................................................................................18 2. Foliar Disease Management 2.1 Runner and Valencia Fungicide Timing Trials..............................................19 2.2 Top Six Valencia Evaluation with/without Fungicide. ..................................23 2.3 2016 ACI Seed Variety Trials........................................................................25 2.4 ICRISAT Variety Trial..................................................................................28 2.5 Tillman Breeding Line Screening.................................................................30 3. Soil Fertility 3.1 Inoculant × Fertilizer Interaction Studies......................................................34 3.2 Foliar × Granular Fertilizer Interaction Trials. .............................................36 3.3 2016 Fulvic Acid Trials..................................................................................38 4. Planting Method Trials 4.1 Seed/Row Spacing Trials. .............................................................................40 4.2 Planting Methods: Rows vs. Traditional Scatter Planting Trials...................45 REFERENCES Appendices I. Monthly Rainfall at MFK. ...................................................................................49 II. Results from Soil Samples at MFK and Acceso. .................................................50 III. Leaf Spot Florida 1-10 and Rust 1-9 rating scales................................................52 IV. 2010-2011 ICRISAT variety trials........................................................................53 V. 2015 Seed/Row Spacing Trials.............................................................................56 VI. Virus Symptoms on Peanut..................................................................................57 Acknowledgements We would like to thank the USAID Peanut & Mycotoxin Innovation Lab for funding this research. We would like to express appreciation to our in-country partners for their contribution to this research: Meds & Foods for Kids Acceso Peanut Enterprise Corp Premier Steppe Ferme We would also like to acknowledge the valuable assistance of student interns from the Université Roi Henri Christophe, Université Solidarité d'Haïti, Université d'État d'Haïti, Campus Roi Henri Christophe, and the Université Chrétienne du Nord d'Haïti for their contributions to this research: Name of Student Gender University Abbreviation Study Fredo Joseph Male URHC Seed Spacing Junie Pachoute Female URHC Seed Spacing Rodlin Jean Male URHC Seed Spacing Daphenie Jean Female URHC Runner Fungicide Timing Kinson Pierre Male URHC Runner Fungicide Timing Fedeline Charles Female URHC Runner Fungicide Timing Rico Mondestin Male URHC Seed Spacing Rostiny Frédérick Male URHC Seed Spacing Emile Blaise Female URHC Seed Spacing Junior Abraham Male URHC Seed Spacing Exan Desamours Male URHC Runner Fungicide Timing Gary Benoit Male URHC Runner Fungicide Timing Rodnie Valmy Female URHC Top 6 Valencia Rodemane Saint Louis Female URHC Top 6 Valencia Elizé Léandre Male UCNH Valencia Fungicide Timing Marilène Saint-Juste Female URHC Valencia Fungicide Timing Galeine Queranor Female URHC Valencia Fungicide Timing Lainé Dorinvil Male URHC Top 6 Valencia Frisnel Pierre Male URHC Top 6 Valencia Judeline Joseph Female UCNH Seed Spacing Jean-Baptiste Fontilus Male UCNH Top 6 Valencia Jean Jones Joseph Male UCNH Tillman Variety Rolcky Butois Male UCNH Runner Fungicide Timing Norvilmar St Firmin Male UCNH Planting Method Dapheney Dolcé Female UCNH Soil Fertility Telson Richard Male USH Valencia Fungicide Timing Rodson Charles Male USH Runner Fungicide Timing Wendy Antoine Male URHC Planting Method Joseph Job Male URHC Planting Method Ruth Eustache Female URHC Seed Spacing Myrvelise Jules Female URHC Seed Spacing Pierre Richard Charles Male UNEPH Planting Method Yonel Louis Male UCNH ICRISAT Variety Ronald Jean Male CRHC-UEH Soil Fertility UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 4 Chapter 1.1 Peanut Production in Haiti Introduction: Peanut has been cultivated in Haiti for at least 500 years and most likely dates back to prehistoric times. According to Bartolomé de las Casas, a priest who accompanied Christopher Columbus on his expedition to the New World and who is accredited with the first written description of the peanut, the native Amerindians cultivated peanut as a food crop on the island of Hispaniola prior to the arrival of the Europeans (Hammons, 1982). Peanut production in Haiti has continued up to the present day, and it is a popular crop because it brings a high market price and is an important and enjoyable food source for many Haitians. Dried peanuts can be found yearround in most open-air markets, and locally made peanut products such as peanut butter (including sweet, spicy, and unflavored forms) are commonly sold in shops and supermarkets (Nelson et al., 2003). Although peanuts are grown throughout the country, there are regions with more concentrated production (Figure 1.1.a). The heaviest centers of production occur in the Northeast near Ouanaminthe, Haiti, and in the Central Plateau region (from Mirebalais eastward to the Dominican border). It is estimated that, on average from 2012 to 2014, peanuts were planted on 45,590 ha, representing roughly 3% of the land area devoted to agricultural production in Haiti (FAOSTAT, 2016). There are two primary peanut market-types grown in Haiti – the local Haitian runner and the local Haitian Valencia. Interestingly, farmers in the North traditionally sow only the runner variety, whereas the local Valencia is confined to the Central Plateau. To the best of our knowledge, there is no information available concerning the origin of these peanut varieties, and our assumption is that they are landraces that were introduced into Haiti at some point in the past and have continued to be cultivated to the present day. Farmers generally save their own seed or purchase their seed from the market. Similar to other varieties within these market types, the local Valencia and the local runner reach full maturity on average from 80 to 90 days and 120 to 130 days, respectively. Generally speaking, cropping practices are similar throughout Haiti, and are primarily characterized as low-input agro-ecosystems under rain-fed conditions. For instance, peanuts are often manually planted and harvested by groups of neighboring farmers in rural communities on gardens/farms 1 hectare or less in size. Corn, sorghum or sugarcane are often rotated between peanut crops. Such production systems require farmers to use hoes or ox-drawn plows to prepare the land for seeding, and do not involve fertilizer or pesticides. Additionally, peanuts are typically planted during late spring/early summer (i.e., to align with the rainy season), and depending on the zone, a late summer/fall/winter crop may be planted, such as in the northern part of the country where fall/winter rainfall is generally more abundant. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 5 Yield-Limiting Factors: In comparison to high-input systems in the United States, peanut yield in Haiti is very low. For example, in the U.S. state of Georgia, the average peanut yield in 2016 was 4,416 kg/ha (USDANASS, 2016), but in Haiti, average yields were estimated to range from 448 to 897 kg/ha (FAOSTAT, 2016; Nelson et al., 2003), and field experience has shown farmers consistently see even lower yields. As will be seen in the results of this research, both of the local Haitian varieties are capable of achieving yields greater than 4,000 kg/ha in Haiti. This highlights the major gap between actual and potential peanut yield in Haiti. A number of obvious factors for low-yielding peanuts in Haiti can be explained with Figure 1.1.b. This picture is representative of many fields in Haiti and illustrates the following: i. .Brown leaves and defoliated plants. Foliar diseases (see section below for details) are major yield limiting factors in Haiti. In this case, this field was planted with a highly susceptible local variety and not treated with fungicide, which resulted in premature defoliation, and, ultimately, fewer mature pods. ii. .Decreased plant growth. Stunted plants indicate poor soil fertility and moisture deficiencies throughout the growing season. iii. .Low plant density. Large gaps between plants decreases the potential yield per unit of land area and increases the likelihood of weed pressure. In this case, low plant density was mostly due to improper seeding rate, but was most likely also affected by poor seed quality, lack of seed treatment and/or lack of moisture after planting. Many growers are not aware that their peanut yields are actually considered extremely low. Rather, the fields look natural, the same as they always have looked. Therefore, it is often the case that growers express great surprise (and delight) at the discovery of the actual yield potential of peanut in Haiti. Quality Reducing Factors: In addition to low yields, peanuts grown and sold in Haiti are often contaminated with dangerous, health-damaging levels of aflatoxins. Aflatoxins are carcinogenic mycotoxins caused by the fungus Aspergillus flavus, and chronic exposure even to low doses can lead to severe health problems, including increased incidence of liver cancer and childhood stunting, and in acute high doses, aflatoxicosis and death. In a recent study, samples taken from locally produced Haitian peanut butter, aflatoxin levels ranged from 7.9 to 799.8 micrograms/ kg aflatoxin, and 16 out of 18 samples had more than 20 ng/kg, the U.S. Food and Drug Administration (FDA) regulatory limit (Schwartzbord and Brown, 2015). Foliar Diseases of Peanut in Haiti: Figure 1.1.b. A typical example of the condition of the local Haitian runner in a grower field nearing the time of harvest close to Ouanaminthe, Haiti. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report As previously mentioned, foliar diseases of peanut are major factors responsible for the gap between actual and potential pod yield in Haiti, and have therefore been a major focus of this project. These diseases include 6 early leaf spot caused by Cercospora arachidicola, late leaf spot caused by Cercosporidium personatum, and peanut rust caused by Puccinia arachidis (see Fig 1.1.c and d). While all three of these fungal pathogens are capable of infecting peanuts in Haiti, our studies demonstrate that those causing peanut rust, followed by late leaf spot, are the most important. These diseases occur on leaves, petioles, and stems, and thrive in conditions of prolonged moisture (>12 h), such as prolonged rainfall, extended dew interval or extreme relative humidity (>90%) (Shokes and Culbreath, 1997; Subrahmanyam, 1997). In Haiti, these diseases often lead to 100% defoliation prior to the plants reaching full maturity (e.g., as depicted in Figure 1.1.e). Yield loss results because fewer leaves are capable of photosynthesis and harvest must come before many of the pods reach full maturity. Figure 1.1.c. Leaves of the local Haitian Valencia peanut market type with late leaf spot and peanut rust. Options for managing foliar peanut diseases in the tropics include planting resistant varieties, rotating crops, destroying volunteer peanuts and infested plant residue, and using Figure 1.1.d. Leaves of the local Haitian runner with early leaf spot (light brown lesions), late leaf spot (dark brown lesions) and peanut rust (light/dark orange pustules). fungicide judiciously (MacDonald et al., 1985; Subrahmanyam et al., 1985). Currently, there are no high-yielding disease-resistant varieties identified for use in Haiti, and the local Haitian runner and Valencia are extremely susceptible to both diseases. However, previous studies conducted in Haiti confirmed that resistant varieties can significantly reduce disease intensity and increase yield (Fulmer et al., 2012). The data from these studies can be referenced in Appendix IV of this report. Similar to other tropical areas where peanut is cultivated (MacDonald et al., 1985), crop rotation and destruction of crop residue likely has only a limited impact on reducing the overall inoculum level since there are almost always abandoned/ fallow fields with volunteer peanuts. Furthermore, since there are almost always peanuts grown in every month in Haiti at some locale, and because the spores of the causal fungi are aerially dispersed by wind and insects, it is probable that there is a constant source of spores, making it difficult for the peanut plants to evade contact with the pathogens. However, as there is little research on this topic, this tactic should still be encouraged as part of an integrated disease management program. Fungicides are relied upon in many developed and developing countries for managing the foliar diseases in question. In other tropical countries, three to four applications have been shown to significantly increase yield (Naab et al., 2009; Waliyar et al., 2000), and preliminary results in Haiti indicates that two applications significantly reduce disease severity and increase yield (J. Rhoads, unpublished data). However, the best timings and number of these applications has not been well understood. Recently, our research in Haiti has confirmed that low-input fungicide regimes are extremely effective for reducing disease and increasing yield; results have been used to develop more specific use recommendations (Chapter 1.3, and Chapter 2.1). UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 7 In addition to the fungal diseases of peanut just mentioned, our studies have identified the presence of a tospovirus that can occur on peanut in Haiti (Adegbola et al., 2016). Symptoms of this disease are described in Appendix VI of this report. However, after three years of monitoring peanuts for this disease in Haiti, we have found that incidence is rather low (<5%) and sporadic and that the disease is mainly confined to the northern regions. Overall, our results indicate that while it could be a yield-limiting factor in some instances, such as in plots planted with low plant density (See Chapter 4.1), we believe that there is generally very little (if any) yield loss that results from this disease in Haiti. Figure 1.1.e. Local Haitian runner with >95% defoliation caused by late leaf spot and peanut rust. Plots to the left were untreated; plots to the right were treated with four applications of fungicide. Soil Fertility: Soils in Haiti are generally considered infertile due to years of intense erosion caused by unmitigated deforestation and near continuous cropping, as well as a natural fragility resulting from the underlying bedrock formations and soil types (Bargout and Raizada, 2013). Overall, most of the soil in Haiti is highly alkaline and high in calcium (Bargout and Raizada, 2013). In a previous study conducted at the University of Florida that evaluated 1,500 soil samples from locations throughout Haiti, most were low in nitrogen and 62% were reported to be deficient in phosphorus; in 96% of the cases, potassium was not a limiting factor (Hylkema, 2011). From soil samples taken at our research sites in Haiti, our results corroborate Hylkema (2011) in that they consistently had a high pH and high levels of calcium. However, phosphorus was mainly only below average in fields located in the Central Plateau (see Appendix II of this report). It should be noted, however, that the fields sampled in the Central Plateau did not have a history of fertilizer inputs, whereas, the other fields sampled at MFK had a prior history of fertilizer inputs. In addition to our observations, results from our soil samples also indicate that many of the soils in Haiti have a heavy clay content (see Appendix II of this report), although there are some areas with extremely sandy soils (J. Rhoads, personal communication). Peanuts generally require a well-drained, sandy soil (e.g., loamy sand, sandy loam, or sandy clay loam) (Henning et al., 1982). This soil type not only promotes growth of the plant, it also facilitates the harvesting process by making it easier to dig the pods from the ground and leaving less soil clinging to the pods (Stalker, 1997). High-pH soils (such as in Haiti) are challenging environments for peanut production as this can lead to other nutrient deficiencies (e.g., by binding up nutrient availability) and can also result in iron chlorosis and zinc toxicity (Stalker, 1997). Peanuts also require soils with high calcium content, which is a positive aspect of the soils in Haiti (Stalker, 1997). Boron is an important micronutrient for proper seed development (Cox et al., 1982) and samples have shown the soils to be deficient. As with other legumes, peanut growth is generally not thought to be limited by low nitrogen soils, since they can receive nitrogen from the symbiotic relationship with Rhizobium root nodulating, nitrogenfixing bacteria that convert atmospheric nitrogen to available nitrogen used in plant growth. There are a number of instances where peanut yields have been increased due to applications of nitrogen, but it was presumed that these results were due to lack of Rhizobium strains in the soil (Cox et al., 1982). UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 8 In short, growing peanuts in Haitian soils is challenging primarily due to heavy clay content, complications from high soil pH, and uncertainty in regards to the capacity of peanut plants to utilize native Rhizobium strains. In order to help close the actual yield vs. potential yield gap, more research is needed to better understand how to manage the fertility of the soils in Haiti. Planting Method: Proper seed spacing between rows and within rows is a cultural practice that directly relates to yield per unit of area (Henning et al., 1982). As a result, much research has been conducted on determining the most appropriate planting density, which ultimately depends on seed quality, seed size, row spacing and variety (Henning et al., 1982). In previous studies where cultivation methods would have been most relative to those in Haiti, “the highest yields of Spanish varieties were realized from plantings 45-60 cm (1824 in) between rows with plants 15-20 cm (6-8 in) within the row. Cultivars of the runner and Virginia types yielded highest when planted with row spacings of 75-90 cm (30-36 in) and with plant spacings of 15-20 cm (6-8 in) in the row” (Henning et al., 1982). In short, bunch type varieties (Spanish and Valencia market types) generally benefit from higher plant density per unit of land, whereas prostrate-growth types (Virginia and runner market types) do not benefit from as high a plant density per unit of land. In Haiti, planting method is one of the few things growers have a very strong opinion about and a sense of control over. While nearly all peanuts are sown by hand, the exact planting method often differs by region or farmer, but generally consists of scatter planting or single-furrow planting. The first method involves using a hoe to make a divot in the ground, dropping seed into the hole/divot and covering it with the feet. Depending on the farmer, these divots may be spaced approximately 30 cm to 45 cm (12 in to 18 in) and may include one or two seed per divot/hole. In the second method, a farmer uses an ox-drawn plow to make a single furrow in the ground, while walking in a circular pattern around the field; another person comes behind dropping one to two seed at 30 cm to 60 cm (12 in to 24 in) spacing within the row. Furrows are generally spaced 45 to 60 cm apart and as the new furrow is made, the soil is pushed in the direction of the previously seeded furrow and is generally enough to bury the seed in the adjacent row. It is generally understood that planting in rows is more beneficial than scatter planting. This makes sense where tractor operated mechanical planters can create a much more efficient, uniform and precise placing of seed. However, in Haiti, where there is little to no mechanical implementation beyond soil preparation, it is uncertain whether the additional labor required for planting in rows would benefit growers. From a practical standpoint and without considering the economic implications, planting in rows appears to be more beneficial for a number of reasons. First, it is a more precise way of utilizing and estimating seed for planting. Secondly, it facilitates practices aimed at the maintenance of the crop, such as easier weeding and harvesting with a hoe, and makes the uniform application of fungicide by backpack sprayer much easier. Not only would this likely lead to better spray coverage, it would also decrease the probability of damaging the main stem (and pegs around the main stem) by stepping on the plant. Increased Demand of Peanut in Haiti: Since 2007, Meds & Foods for Kids (MFK) has been making a peanut-based Ready-to-Use Therapeutic Food (RUTF), locally known as Medika Mamba and commonly known as Plumpy’nut (MFK, 2017). This product is the gold standard for the treatment of severe acute malnutrition in children both globally (WHO, 2017) and in Haiti (Iannotti et al., 2015). Childhood malnutrition continues to be a major issue in Haiti due to the high levels of poverty and food insecurity (PMIL, 2017). MFK produces and distributes RUTF and other similar peanut-based supplementary products through partnerships with UNICEF, the World Food Programme and many other local and international humanitarian organizations. As a Non-Governmental Organization (NGO), MFK’s approach is to facilitate treatment of malnutrition, but also to address the root of the problem by developing the economic sector through employment at their factory and driving demand for locally sourced agricultural products. As such, MFK’s desire has been to purchase 100% of the peanuts for the factory from local Haitian farmers. However, this remains difficult due to a number of reasons, including low quality, aflatoxin contamination, and inconsistent and uncompetitive high market UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 9 prices due to low productivity. By increasing the production of high-quality peanuts, MFK would be able to accomplish this goal of 100% local sourcing. The strategy to achieve this has been to lower the overall cost by increasing yields. The end-result theoretically would be a win-win situation for growers and consumers alike. Growers would benefit from yield increases, despite lower market prices, as long as net profitability remained high through cost controls. A lower market price and supply of aflatoxin-free peanuts would benefit the average consumer both from a financial standpoint and from a health standpoint and facilitate the in-country purchase of value-added products such as Medika Mamba. Additionally, Acceso Peanut Enterprise, a for-profit peanut value-chain business, has been operating in Haiti since 2015 with the aim of increasing farmer productivity and easing difficulties of aggregation to meet this demand for high-quality peanuts. The Acceso model directs technology exchange by establishing a system of depots in small, rural communities. Depot managers trained on best production practices are able to extend this information to local growers. From these depots, local farmers in the Acceso program are able to obtain yield increasing inputs such as high quality seed, fertilizers and fungicide on credit. Following harvest, these same farmers are able to sell their peanuts back to Acceso at a fixed, competitive price. Once Acceso sorts and tests the peanuts for aflatoxin and kernel moisture content, they are then able to deliver a high quality product to local businesses with whom they have previously established a purchasing contract. References: Adegbola, R., Fulmer, A., Williams, B., Brenneman, T., Kemerait, R., Sheard, W., Woodward, J., Adkins, S., & Naidu, R. (2016). First report of the natural occurrence of tomato chlorotic spot virus in peanuts in Haiti. Plant Dis. 100:8, 1797. Bargout, R. N., & Raizada, M. N. (2013). Soil nutrient management in Haiti, pre-Columbus to the present day: lessons for future agricultural interventions. Agric. Food Secur. 2:11. Cox, C. R., Adams, F., & Tucker, B. B. (1982). Liming, Fertilization and Mineral Nutrition. Pages 139-163 in: Peanut Science and Technology H. Pattee and C. Young, eds. American Peanut Research and Education Society, Yoakum, TX. FAOSTAT (2016). Food and Agriculture Organization of the United Nations. Statistics Division. Retrieved from http://www.fao.org/faostat Fulmer, A. M., Kemerait, R. C., Sherwood, J.L., Jordan, D.L., Rhoads, J., & Brenneman, T. B. (2012). Evaluation of ICRISAT varieties for resistance to foliar peanut diseases in Haiti. Phytopathology 102 (Suppl.) S2:4. Hammons, R. (1982). Origin and early history of the peanut. Pages 1-20 in: Peanut Science and Technology H. Pattee and C. Young, eds. American Peanut Research and Education Society, Yoakum, TX. Henning, R. J., Allison, A. H., & Tripp, L. D. (1982). Cultural Practices. Pages 123-138 in: Peanut Science and Technology H. Pattee and C. Young, eds. American Peanut Research and Education Society, Yoakum, TX. Hylkema, A.L. (2011). Haiti soil fertility analysis and crop interpretations for principal crops in the five winner watershed zones of intervention. M.S. Thesis. University of Florida, Gainesville. Iannotti, L. L., Henretty, N. M., Delnatus, J. R., Previl, W., Stehl, T., Vorkoper, S., Bodden, J., Maust, A., Smidt, R., & Nash, M. L. (2015). Ready-to-use supplementary food increases fat mass and BMI in haitian school-aged children. J. Nutr. 145:813-822. MacDonald, D., Subrahmanyam, P, Gibbons, R., & Smith, D. (1985). Early and late leaf spots of groundnut. Information Bull. No. 21. ICRISAT, Patancheru. MFK (2017). Meds and Food for Kids. Retrieved from https://mfkhaiti.org Naab, J., Prasad, P., Boote, K., & Jones, J. (2009). Response of peanut to fungicide and phosphorus in on-station and on-farm tests in Ghana. Peanut Sci. 36:157-164. Nelson, R., Jolly, C., Hinds, M., Donis, Y., & Prophete, E. (2003). Consumer preferences for peanut butter (mamba) products in Haiti: A conjoint analysis. Peanut Sci. 30:99-103. PMIL (2017). University of Georgia Peanut Mycotoxin Innovation Lab. Retrieved from http://www.caes.uga.edu/global/feed-the-future-innovation-labs/peanutmycotoxin-innovation-lab.html Schwartzbord, J. R., & Brown, D. L. (2015). Aflatoxin contamination in Haitian peanut products and maize and the safety of oil processed from contaminated peanuts. Food Control 56:114-118. Shokes, F., & Culbreath, A. K. (1997). Early and late leaf spots. Pages 17-20 in: Compendium of Peanut Diseases, 2nd Ed. N. Kokalis-Burelle, D. Porter, R. Rodriguez-Kabana, D. Smith and P. Subrahmanyam, eds. American Phytopathological Society Press, St. Paul, MN. Stalker, H.T. (1997). Peanut (Arachis hypogaea L.). Field Crops Res. 53:205-217. Subrahmanyam, P. (1997). Rust. Pages 31-33 in: Compendium of Peanut Diseases, 2nd Ed. N. Kokalis-Burelle, D. Porter, R. Rodriguez-Kabana, D. Smith and P. Subrahmanyam, eds. American Phytopathological Society Press, St. Paul, MN. Subrahmanyam, P., Reddy, L., Gibbons, R., & MacDonald, D. (1985). Peanut rust: a major threat to peanut production in the semiarid tropics. Plant Dis. 69:813-819. USDA-NASS (2016). Agricultural Statistics Database. Retrieved from https://www.nass.usda.gov Waliyar, F., Adamou, M., & Traoré, A. (2000). Rational use of fungicide applications to maximize peanut yield under foliar disease pressure in West Africa. Plant Dis. 84:1203-1211. WHO (2017). Malnutrition. World Health Organization. Retrieved from http://www.who.int/maternal_child_adolescent/topics/child/malnutrition/en/ UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 10 Chapter 1.2 PMIL Project Objectives and Role of the PMIL Project: The Peanut and Mycotoxin Innovation Lab (PMIL) is part of the U.S. government’s Feed the Future initiative administered by the U.S. Agency for International Development (USAID) and is intended to improve livelihoods and health in developing countries through advances in peanut research in production, processing, and markets. The Haiti Value Chain project is multidisciplinary collaboration and involves specialists from the University of Florida, the University of Georgia and Cornell University. Key to this project is the collaboration of in-country partners–MFK, Acceso, and Premiere Steppe Ferme. In addition, PMIL also has a global peanut breeding program, including a relatively new initiative in Haiti led by Barry Tillman at the University of Florida in collaboration with Raphael Colbert at Quisqueya University in Haiti. From 2007 to 2012, PMIL’s predecessor program, the Peanut Collaborative Research Support Program (PCRSP) worked directly with MFK to improve local production. During this phase of the project, efforts were primarily focused on working directly with local farmers to provide support from a local agronomist employed with MFK. Local production was linked to providing peanuts for the MFK factory but also to local markets. In 2013, the project linked with TechnoServe, a global NGO known for agribusiness development that was providing technical support to the Partners in Health/Zanmi Lasante RUTF factory in the Central Plateau. During this time, Acceso was formed and took over that role and adopted a small business model approach. As previously noted, their model provides technical inputs to improve production, such as tillage services, improved varieties, fertilizer and fungicide. From 2013 to 2017, PMIL, in collaboration with MFK and Acceso, focused on applied research to support outreach programs and technical inputs for small holder farmers. This involved working with Acceso agronomists and depot managers. Applied research was conducted through MFK in collaboration with Acceso and allowed for more formal training of students and local agronomists. Throughout this time, information generated from the research program and the technical guidance of the PMIL specialists continued to provide input for extension related materials. Toward the latter phase of the project, the focus of PMIL shifted from extension-based programs (trainings and materials directed to growers) to applied production research and the training of agronomists, Acceso depot managers (train the trainer), and agronomy students from local universities. The latter resulted in training 34 undergraduate students in applied research through a rigorous and competitive internship/undergraduate research project program offered at MFK. From the applied research relating to enhancing yield, specific objectives were to: 1. Evaluate and screen multiple varieties in order to identify a high-yielding, disease-resistant variety suitable to the environment in Haiti; 2. Determine the optimum number and timing of fungicide applications for runner and Valencia market types grown in Haiti; 3. Test the effect of different treatments aimed at boosting soil fertility; and 4. Determine the most appropriate method for planting runner and Valencia peanut market types. References: MFK (2017). Meds and Food for Kids. Retrieved from https://mfkhaiti.org PMIL (2017). Peanut Mycotoxin Innovation Lab. Retrieved from http://www.caes.uga.edu/global/feed-the-future-innovation-labs/peanut-mycotoxin-innovation-lab.html UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 11 Chapter 1.3 Recommended Best Practices for Peanut Production in Haiti Purpose: To provide agronomists and grower advisors in Haiti with the best possible recommendations based on our research regarding variety, planting, fertility, and fungicide in order to aid in the development of a more complete technology package for Haitian peanut farmers. These recommendations are based on the results of our research and experience in Haiti and in the Southeast U.S. Variety Selection: Valencia. Currently, our research efforts have failed to identify a high yielding, disease-resistant variety that consistently outperforms the local Valencia. The New Mexico Valencia A was extensively screened in the fungicide timing trials (Chapter 2.1), but did not offer any advantage over the local Valencia in terms of disease severity or yield (Figure 2.1.b). The Top 6 Valencia trials (Chapter 2.2) demonstrated that 309 Tan, a Valencia variety from the New Mexico State University (NMSU) breeding program, had excellent resistance to foliar diseases in Haiti, but did not yield as high as the local Valencia in plots with and without fungicide (Figure 2.2.b). Similarly, while other varieties from the NMSU breeding program did have a higher numerical yield compared to the local Valencia, the increase does not appear to be significant enough to justify the introduction of any of these varieties into Haiti at present. Current recommendation: Farmers who desire to plant a Valencia market type should continue to use the local Haitian Valencia. Runner. Georgia-06G consistently out-yielded the local Haitian runner in the fungicide timing trials (Figure 2.1.b). From a physiological standpoint, the Georgia-06G variety generally exhibits stunted growth in Haiti and, compared to its growth in the southeastern U.S., does not appear to thrive in most Haitian soils. However, upon digging, the pod load is still consistently higher than the local runner. Also, it should be noted that Georgia-06G is extremely susceptible to peanut rust while the converse is true for the local runner for late leaf spot (Figure 2.1.a). Current recommendation: Although the Georgia-06G variety is not the perfect ideal for Haiti, it could be a good option for growers until a better option is identified. It should be noted that the local Haitian varieties have several positive traits that appear to have been selected for over years of cultivation in Haiti. Both varieties have excellent seed vigor; they are almost always the first to germinate and both seem to grow well in Haitian soils. This may also include greater resistance to native soilborne diseases. This suggests that a local breeding program to improve upon these local adaptations may be a feasible strategy in the future. Fertility: Thus far, we have failed to see a significant yield increase from any of the fertility treatments that we have evaluated in Haiti (Chapters 3.1, 3.2, and 3.3). However, we must emphasize that many of the fertility studies were conducted in fields that likely had a good residual level from previous peanut crops that were fertilized. More studies need to be conducted in fields with known fertility deficiencies before a definitive recommendation can be made. It should be noted, however, that we did see a numerical trend in the most recent fertility studies (Chapter 3.2) that suggests that a significant yield response would be obtained in nutrientdeficient soils. As noted in the research on fulvic acid and foliar applications, the high-pH soil greatly contributes to the challenge of determining highly responsive fertilizer recommendations, in spite of the known deficiencies in most soils. It appears that the use of an inoculant (to increase root nodulation by Rhizobia species) is not necessary in Haiti. We did not see a positive yield increase in either of the inoculant trials we conducted (Chapter 3.1) in 2015, and we have consistently found good nodulation on the roots of plants in noninoculated fields. It should be also noted that inoculants require cool storage conditions (which make implementing this practice very difficult) and are not UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 12 readily available in Haiti. Furthermore, evidence from alkaline soils in Texas suggests that liquid inoculum greatly outperformed granular forms and that seed coated inoculant was not found to be effective at all. Current recommendation: Given that soil tests are not very practical in Haiti and the general infertility of most soils, we still suggest that growers apply 40-60 kg/ha of 20-20-10 or DAP at planting or as a side-dress application two to three weeks after planting. However, we do not suggest using an inoculant at this time. Planting: Seed/row spacing trials (Chapter 4.1) have consistently demonstrated that the ideal planting density is not the same for the local runner and the local Valencia. Valencia. Overall, regardless of between-row spacing or within-row spacing, the Valencia variety yield consistently increases with increasing planting density (Table 4.1.c and Figure 4.1.c). However, there is generally less of a yield gap between 12- and 18-inch row spacing than 24- and 18-inch row spacing, and three and six seed/ ft than between one and three seed/ft (Figure 4.1.d). In the absence of a formal cost-benefit analysis, this suggests that 18-inch rows planted at three seed/ft may be the best option for growers in Haiti. However, it should be noted that these studies were conducted with high-quality seed with excellent germination. This is often not the case in Haiti–therefore, if the seed germination is questionable, we would advise using 12-inch row spacing with three seed/ft spacing within the row, or 18-inch row spacing with six seed/ft spacing within the row. The positive correlation between planting density and yield for the local Valencia was corroborated with the row vs. scatter planting method trials (Chapter 4.2). In these studies, we found that the traditional scatter planting method did not result in a yield loss when compared to the same amount of seed sown in rows. As mentioned elsewhere, while we believe that there are number of advantages for planting in rows (better plant density control, ease of crop protection and harvest, etc.) and will continue to advocate that practice, these data suggest that the traditional method can provide equivalently high yields at higher densities. Runner. We did not find the same consistency in the response to seed/row spacing treatments for the runner variety (Table 4.1.c and Figure 4.1.c). However, yield in plots with three and six seed/ft within-row spacing were more often higher than plots with the 1 seed/ft spacing (Figure 4.1.d). Row spacing did not have an effect on yield when planted at three or six seed/ft (Table 4.1.c), suggesting that the within-row spacing is more important for the runner variety. Results from the row vs. scatter planting method trials (Chapter 4.2) corroborate that an increase in the planting density of the local runner variety does not necessarily result in a significant increase in yield (Figure 4.2.c). Also, similar to the results for the Valencia market type, the results from these trials suggest that planting in rows does not necessarily signify an increase in yield compared to the traditional scatter method. As long as the planting density is similar, similar yields can be obtained from both methods. Current recommendation: For the Valencia, we suggest planting 18-inch rows with three seed/ft or 24-inch rows with six seed/ft. For runner, we suggest planting 24-inch rows at six seed/ft. Both of these suggestions assume a relative low seed germination of 50-70%, which is often the case in Haiti. Fungicide Applications: Results from multiple studies conducted from 2015 to 2017 on runner and Valencia market types emphasize the importance of managing foliar diseases in Haiti (Chapter 2.1). However, yield loss for runner varieties is higher than Valencia varieties, likely due to the phenomenon of disease escape. Runner varieties require ~40 more days to reach maturity than Valencia varieties and are therefore exposed to the threat and impact of foliar diseases for a longer period of time. Overall, for both Valencia and runner varieties, we report an inverse relationship between disease severity and fungicide applications, namely, disease increases with decreasing fungicide applications (Figure 2.1.a). Similarly, pod yield tended to increase with increasing fungicide applications (Figure 2.1.a). As a result, a straightforward recommendation cannot be made simply based on the differences between treatments alone. Practical, biological and economic factors must also be considered in the decision process. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 13 For example, most growers in Haiti do not have access to fungicide spray equipment and the appropriate products are not widely available. Therefore, most farmers must purchase the application service from a contract service provider (such as Acceso). In this scenario, three to four applications would likely be the maximum number of applications that could be made during the season. From a biological standpoint, fungicide resistance is something that must be considered when attempting to make a fungicide recommendation. The current fungicide available for growers in the Acceso program is a mixture of tebuconazole and chlorothalonil (Muscle® ADV). Tebuconazole resistance has been detected in populations of both leaf spot pathogens in the United States, and the rust pathogen could also have a similar risk. A key principle of avoiding resistance is to make applications when the pathogen population is low. In Haiti, we have found that this is generally between 30 and 60 days after planting (DAP), depending on which pathogen is in question. A second factor that aids this key principle is to use more frequent applications, thereby ensuring that the pathogen population is still low when the second/third/fourth application of fungicide is made. Ideally, therefore, in Haiti, applications should start around 45 DAP and continue every 14 days in order to decrease the risk of fungicide resistance. A more formal cost/benefit analysis will be needed to help determine the most appropriate recommendation. Just because there are more peanuts in the plots treated with the greatest number of fungicide applications doesn’t mean that it will translate into the most profitable return on investment. For this reason, a more in-depth analysis is currently being conducted by the economist in order to help guide the decision-making process. Current recommendation: Without considering an economic analysis–strictly from a disease-management perspective–the data and the aforementioned practical and biological considerations suggest that three applications at either of the timings evaluated would be the best practice for Valencia varieties grown in Haiti. For runner varieties, we suggest that four applications made at either of the timings evaluated would be advisable for growers in Haiti, particularly during the rainy season. Chapter 1.4 Lessons Learned Purpose: The objective of this chapter is to reconsider critical steps that have led to success and failure in this research project undertaking to evaluate potential technologies aimed at improving peanut productivity in Haiti. The other chapters review completed projects with relatively comprehensive data collection; however, many other trials were attempted during this period that were deemed inadequately definitive for inclusion or were not completed for various reasons. Undertaking field research in Haiti has proven challenging in often unanticipated ways, and there are numerous lessons to be learned for future research endeavors in Haiti or in similar environments with limited research infrastructure or technical experience. Site Selection: Without a proper research station with historic field data, weather data collection, equipment for field preparation, planting, irrigating, harvesting, or sample processing, trained experienced staff or security from livestock or inclement weather, conducting rigorous repeatable research is a real challenge. Initial efforts to work on collaborating farmers’ fields, leased land or university land, led to mixed results. Examples: • Two unfenced trials were lost to livestock invasion. • One trial was lost at a university for failure to anticipate student vacation. • At least two trials were lost to inclement weather, including flooding and drought, before irrigation and extensive drainage systems were established. • Soil variability (texture and fertility) is quite high and resulted in data exclusion in some trials. Soil data was a challenge, since the quality, cost, and timeliness of local analysis was limited. Samples were sent to the U.S., requiring permits and shipping costs. Also, the high-pH soil required different analytical techniques than commonly used in Georgia. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 14 • Access to equipment for field preparation was a limiting factor for early trial efforts due to the limited number of tractors in the country. Manual preparation was possible, but it was often exceedingly expensive and inconsistent in quality. Collaborating Partners: Haitian farmers, agricultural technicians, and agronomists have varying, but generally limited, experience designing and conducting controlled research experiments. With a lack of mechanization for all processes of preparation, planting, management, harvest, and sample processing, the extensive use of manual labor (often untrained labor that may have extensive production experience, but no research experience) opens the door for mistakes to be made. Communication across languages and cultures proved to be challenging. Examples: • After plots testing a biocontrol product were harvested, field laborers commingled all the replications of treatment and control plots to make drying easier. • Hand-seeded variety trials were abandoned after the crops started to mature and obvious, consistent physiological differences appeared between rows, likely due to inattentive labor during seeding. This resulted in the loss of valuable imported seed and a failure of the trial. • Eager field assistants reasonably not wanting to work in the hot afternoon sun completed the harvest of several trials without proper labeling resulting in a complete loss of data. • The in-country partners often relied on short-term foreign interns, which resulted in a loss of continuity and lack of institutional learning. The experience level of interns with field agriculture varied greatly and led to handover issues and a lack of strategic direction. • Relying on U.S. experts for research design and oversight led to an occasional lack of prioritization of research with farmer conditions or the inability to adequately supervise the implementation of trials and data collection. Equipment and Facilities: As previously noted, the mechanization of field processes was an initial limiting factor but was resolved over time by concentrating efforts in targeted locations. Some equipment that was integral to quality data collection, but was often overlooked, included: • Adequate and appropriate storage for seed (ideally conditioned), inputs, tools, etc. • Weather stations with data loggers and backup rain gauges. Simple tools for plot marking, labeling tags, mesh bags for plot samples, battery-powered scales, robust moisture meters (Dicky John miniGAC®), and simplified quantitative aflatoxin-testing equipment (Mobile Assay mReader). • Backpack sprayers that are robust and easily calibrated for use by local technicians, including booms for multiple-row spraying to reduce variation. • Access to quality internet has improved and allowed for better communication, including the sharing of photos and real-time data. • Field preparation equipment, such as the use of Chinese-made two-wheel tractors were an initial improvement. The later use of larger, but still relatively small, tractors for disc plowing, harrowing, rototilling, seeding, and threshing were key to scaling beyond small plots. Successful implementation of this equipment for farmers is still elusive due to the high costs of operation, disparate small plots, slope and soil variability, and the large number of trees and rocks in many plots. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 15 Notable Successes: • During the final three years of the project, the intern program at MFK managed to incorporate 34 student projects from four different local universities. Agronomy students must complete a final project and the infrastructure provided by MFK, including the research agenda, field support, and supervision, was key to establishing these connections. This is a true win, but it required learning from previous efforts and establishing a functioning research team. • Barry Tillman was able to establish a relationship with a recent doctoral graduate, Raphael Colbert, to work on bean breeding and expand his efforts into peanut evaluation. After some initial learning, the research team at CHIBAS/Quisqueya University has established a quality system for evaluating germplasm at their research farm in Cabaret. This means that three reliable sites are now available for evaluation and that there is a potential for a real long-term strategy of incorporating improved traits into the existing adapted varieties through plant breeding. • A local commercial-scale farm, Premier Steppe Ferme, has collaborated to scale these and other inputs on their farm, planting blocks of up to 10 ha and averaging yields two to three times the local norm. There has also been learning on the adaptation of advanced mechanized technology for medium-scale ventures, such as the use of used two-row equipment from the U.S. and the Brazilian-made Colombo multi-crop thresher. Notable Teachable Moments: • Initial trials with breeding lines from ICRISAT India showed great promise in disease resistance and drought tolerance (Appendix IV). These varieties were very attractive to local farmers due to their agronomic traits, consistently yielding three times the local varieties even when there was crop failure due to drought. However, when a leading variety (ICGV 99030) that had been multiplied was finally tried for consumption, it was found that the flavor and oil content/quality was completely unacceptable. Assuming that this was perhaps due to improper postharvest handling, a second crop was produced with similar results. It was later learned that these varieties were not intended for release, but for sources of genetic material for making local crosses and had never been screened for consumer traits. • While these data suggest that improved varieties have potential, the local varieties continue to be dominant. However, they are only dominant locally. Trials of runner varieties in the Central Plateau proved to be a failure, not due to yield limitations, but due to the local dislike of that variety for lack of market demand (reportedly too high in oil), farmer preference for the ease of harvesting bunch types, and the predominant strategy of squeezing two crops into the rainy season, which is only possible with the short-duration Valencia. The strategy of focusing on a single, high-yielding plantation was unacceptable to most farmers, likely for its risk. Conversely, in the North and Northeast, the Valencia variety was deemed unacceptable because it was too low yielding and supposedly difficult to harvest (the opposite of the Central Plateau) and lack of market demand for that variety (reportedly too low in oil). Even over a relatively small geographic area, people have strong preferences and research should not try to overcome these preferences, but work within their existing strategies. Likely Lynchpins of Future Success: • Seed affordability and quality remain critical limitations. Continued expansion of seed production, aimed at improving seed quality (maintaining germination and vigor) and reducing cost for farmers during the planting season, will be critical for improving productivity and reducing costs. It will also be critical for introducing improved varieties. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 16 • Finding varieties that are higher yielding, more disease resistant and drought tolerant and meet local demands for quality will be critical to move past the current low yields. Progress can be made in the interim through improved agronomic practices, but the potential for genetic gains is clear in the data presented in this report. • Efforts at mechanization at the farmer level (small-scale tillage, planters) have not proven to be cost effective to date. However, scaling production without some mechanization will not be cost effective in the future, especially as availability of rural labor continues to decline and costs increase. Both proven (animal traction) and new (mobile threshers) technologies should be evaluated. Additional Areas for Future Research: • Seed germination was a continuous problem during trials and resulted in higher recommended seeding rates and associated costs than if germination rates were consistently high. Seed vigor, as measured be time to emergence, revealed a surprising difference between the local varieties, which were generally quick to emerge, and imported varieties, which often took several days longer. A future project should evaluate the environmental (soil fertility and maturity impact of disease) and postharvest handling (high drying temperatures due to solar drying, storage conditions) variables that impact seed vigor and germination. • The recommendations are based on rigorous data collected over multiple seasons, in multiple locations. However, the implementation of these recommendations needs to be evaluated at scale and on farmer fields. Gathering quality empirical data from farmer fields remains a challenge, but should also be addressed with concerted efforts with targeted farmers. Concluding Remarks: Finally, there is great wisdom in several commonly used Haitian proverbs that are worth recounting related to our research experience: 1. “Kabrit ak twòp met mouri anba soley” (A goat with too many owners dies in the sun). There is a requirement of clear communication and delineation of responsibility, which becomes especially important across multiple languages and cultures. With many personalities, both from abroad and locally, and high turnover of key personnel, several key lessons had to be learned more than once. 2. “Pise gaye pa kimen” (Urinating all over the place never forms foam). Though a bit crass, the idea of not spreading efforts too thin or trying to achieve too many things was key to later program success. Trials were conducted on two well-monitored and managed locations where adequate controls were taken for unexpected variables and lessons learned over time. 3. “Wòch nan dlo, pa konn doulè wòch nan soley” (The rock in the water doesn’t understand the suffering of the rock in the sun). There are two key lessons from this proverb: 1) U.S.-based collaborators need to take time to thoroughly communicate and understand the local limitations, including labor and personal constraints of staff, and invest in long-term solutions; and 2) research priorities should be soundly based in the reality of the local farming system to ensure relevance of the research outputs. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 17 Section 1.5 Research Field Sites Locations: Field trials were conducted at research sites belonging to our in-country partners, Meds and Food for Kids (MFK) and Acceso Peanut Enterprise Corporation. The Acceso research farm is located in the Central Plateau, in the community of Coupe Gorge (located just outside of Mirebalais) (18°50’21.05”N latitude, 72° 3’29.33”W longitude), Haiti. The research plots at the MFK factory are located in the community of Quartier Morin (located east of Cap-Haïtien), Haiti (19°41’32.17”N latitude, 72° 9’16.91”W longitude). Field rotations. At MFK, all fields had a previous history of peanut. In most cases, peanuts were planted behind peanut, with a 3- to 6-month fallow period between crops. In some cases, peanut followed a rotation of sorghum. At the Acceso research farm, peanut often followed several years of bean and/or sorghum production. As peanut studies were conducted year-round, we were ensured that a consistent inoculum source was present for fungicide and variety trials conducted for leaf spot and rust. Soil types. Based on soil samples evaluated at the University of Georgia, the soil type in fields used at MFK was a sandy clay loam with an average of 42% sand, 27% silt, and 30% clay with a pH of 7.4, and calcium levels were also over 3,000 kg/ha. Fields used at the Acceso research farm were a clay-based comprised of 25% sand, 25% silt, and 50% clay with a pH of 7.1 (CaCl22), and calcium levels were over 3,000 kg/ha (see Appendix II for soil test results). Part of the land devoted to research plots at Meds and Food for Kids (MFK) factory located outside of CapHaïtien (Quartier Morin), Haiti. Back field being prepared for planting at the Meds and Food for Kids (MFK) factory located outside of Cap-Haïtien (Quartier Morin), Haiti. Research plots used at the Acceso research farm located in Coupe Gorge (part of the Mirebalais commune) in the Central Plateau of Haiti. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report Additional fields used for research at the Acceso research farm located in Coupe Gorge (part of the Mirebalais commune) in the Central Plateau of Haiti. 18 Chapter 2.1 Runner and Valencia Fungicide Timing Trials Purpose: Determine the appropriate number and best timings of fungicide applications for reducing disease severity of foliar diseases and increasing yield for runner and Valencia market types grown in Haiti. Experimental Design: Five trials with runner market type peanuts were conducted at the MFK research site from 2015 to 2017. All trials were laid out in a split plot design with four replications. Variety was the main plot treatment and was planted to the local Haitian runner or Georgia-06G, the predominant cultivar planted in the Southeast U.S. Fungicide treatment was the subplot and consisted of six different application regimes plus an untreated check (see Table 2.1.a for details). Dates from planting to harvest for each trial were as follows: 23 March to 13 August 2015; 9 November 2015 to 8 March, 2016; 19 February to 23 June 2016; 16 October 2016 to 19 February 2017; 31 March to 12 August 2017. Two trials with Valencia market-type peanuts were conducted in the Central Plateau at the Acceso research site in the spring of 2015 and 2016, and three trials were conducted at the MFK research site in the spring and fall of 2016 and spring of 2017. All trials were laid out in a split plot design with four replications. Variety was the main plot treatment and was planted to the local Haitian Valencia or New Mexico Valencia A, an improved cultivar planted in the southwestern United States. Fungicide treatment was the subplot treatment and consisted of six different application regimes plus an untreated check (see Table 2.1.a for details). Dates from planting to harvest in the Central Plateau for each trial were as follows: 31 March to 30 June 2015; 8 April to 16 July 2016. Dates from planting to harvest at MFK for each trial were as follows: 14 March to 16 June 2016; 29 August to 28 November, 2016; 15 May to 29 August, 2017. Table 2.1.a: Treatment details used for runner and Valencia fungicide timing trials in Haiti. y z Market type Total applications Initiationy Spray intervalz Application timings Runner 6 4 4 3 3 2 0 30 37 45 37 45 60 - 14 21 21 28 28 28 - 30, 44, 58, 72, 86, 100 37, 58, 79, 100 45, 66, 87, 108 37, 65, 93 45, 73, 101 60, 88 - Valencia 4 3 3 2 2 1 0 30 30 45 45 45 45 - 14 21 14 21 28 - 30, 44, 58, 72 30, 51, 72 45, 59, 73 45, 66, 45, 73 45 - Days after planting when the first application was made. Days between applications. For each fungicide treatment, applications of tebuconazole (0.23 kg/ha) + chlorothalonil (0.84 a.i. kg/ha) (Muscle ADV, Sipcam Agro USA, Durham, NC) were made at 188 liters per hectare with a hand-pumped backpack sprayer. For all trials, plots were 1.2 m wide and 4.6 m in length, consisting of two rows of peanuts bordered by a single untreated row of peanuts. Blocks were separated by a 1.5-m alley. Runner trials were planted at three seed per 30.5 cm in the first two trials and six seed per 30.5 cm in the last two trials. Valencia trials were planted at a rate of three seed per 30.5 cm in the Central Plateau and six seed per 30.5 cm in the North. Prior to planting, seeds were treated with azoxystrobin, fludioxonil and mefenoxam (Dynasty PD®, Syngenta Crop UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 19 Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Fields were disked two to three times prior to planting, and rototilled within two days prior to planting. A few days prior to planting, fields in the Central Plateau were fertilized with diammonium phosphate at a rate of 45 kg/ha; at MFK, the same rate of diammonium phosphate was used in all trials conducted in 2015. All other trials were fertilized with 20-20-10 at a rate of 67 kg/ha. Manual weeding occurred at 4, 6 and 8 weeks after planting. At the MFK research site, plots were irrigated with a rotary sprinkler system as needed (twice a week in the absence of rain) with approximately 1.3 cm of water per irrigation event. At the Acceso research site, plots were irrigated with flood irrigation similar to the local grower standard. Data Collection: Two to three weeks after planting, stand counts were made for each plot. Final leaf spot and rust severity ratings were taken immediately prior to digging. Leaf spot severity was assessed with the Florida 1 to 10 scale (Appendix III). Rust severity was assessed with a modified 1 to 9 scale (Appendix III). Peanuts were manually harvested by first pulling the entire plant from the ground and removing all the attached pods from the plant. Afterwards, the soil in each plot was filtered through by hand to recover the remaining pods left in the ground. Pods were placed in large, green, mesh cabbage bags (Cady Bag Company, LLC. Pearson, GA) and washed after harvest in order to remove any remaining soil, and then placed on a large concrete pad to dry in the sun. The bagged pods were allowed to dry for a minimum of three days, and were moved under a shelter each night. After drying bags were weighed, and immediately afterwards, a 100 pod sample was shelled to obtain the moisture content of the kernels. Final weights were adjusted to 10% pod moisture. Statistical Analysis: For this report, means and standard errors of final severity of leaf spot and rust and yield were calculated across all trials with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). A more in-depth analysis of variance will be reported in a forthcoming publication. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 20 Figure 2.1.a. Effect of fungicide program on total pod yield (kg/ha) for runner and Valencia market type peanuts averaged across all trials. Fungicide treatments are labeled as follows: number of applications _ day after planting of first application _ subsequent spray interval. Error bars represent the standard error of the mean. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 21 Figure 2.1.b. Effect of variety on total pod yield (kg/ha) for runner and Valencia market type peanuts averaged across all fungicide treatments and trials. Error bars represent the standard error of the mean. Figure 2.1.c. Trial 3: Runner fungicide timing trial 1 June 2016. Figure 2.1.d. Trial 3: Runner fungicide timing trial 23 June 2016. Figure 2.1.e. Trial 1 at MFK: Valencia fungicide timing trial 29 April 2016. Figure 2.1.f. Trial 1 at MFK: Valencia fungicide timing trial 1 June 2016. Figure 2.1.g. Trial 1 in the Central Plateau: Valencia fungicide timing trial 11 June 2015 (PC, PMIL). Figure 2.1.h. Trial 2 in the Central Plateau: Valencia fungicide timing trial 15 June 2016. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 22 Chapter 2.2 Top 6 Valencia With/Without Fungicide Trials Purpose: Evaluate the performance of advanced Valencia market type breeding lines developed at New Mexico State University by Dr. Naveen Puppala for possible introduction in Haiti. Primary interest is to determine varietal response to foliar diseases in relation to final severity of infection and pod yield. Experimental Design: Five trials were conducted in Haiti from 2015 to 2017, four of which were located at MFK and one at the Acceso research farm in the Central Plateau. Table 2.2.a. Planting dates, harvest dates, seeding rates and plot lengths for all trials in 2015 to 2017. Trial Location Planting date Harvest date Seeding rate Plot length 1 2 3 4 5 MFK 13 Nov, 2015 25 Feb, 2016 3 seed/30.5 cm 3.04 MFK 23 Mar, 2016 21 June, 2016 6 seed/30.5 cm 4.6 m MFK 24 Aug, 2016 29 Nov, 2016 6 seed/30.5 cm 4.6 m MFK 27 Jan, 2017 3 May, 2017 6 seed/30.5 cm 4.6 m Central Plateau 19 Jan, 2017 4 May, 2017 3 seed/30.5 cm 4.6 m All trials were laid out in a split plot design with four to six replications. Six Valencia varieties were the main plot treatments and fungicide treatments (with or without three applications) were the subplot treatments. The six varieties included in these studies included the local Haitian Valencia and the following five advanced breeding lines developed by the New Mexico Agricultural State University Experiment Station located at Clovis, NM: 309 Red, 309 Tan, M2, M3, and SCGV0801. Fungicide treatments consisted of a combination of tebuconazole (0.23 kg/ha) + chlorothalonil (0.84 a.i. kg/ha) (Muscle® ADV, Sipcam Agro USA, Inc.) sprayed at 45, 60 and 75 days after planting. A hand pumped backpack sprayer calibrated at 188 liters per hectare was used to make fungicide applications in all trials. Plots were 1.2 m wide and plot length ranged from 3.04 m to 4.6 m length. Each plot consisted of two rows of peanuts, and were planted by hand at a rate of three seed or six seed/30.5 cm. A 1.5-m alley separated blocks, and there was no border row between subplots. Other yield-reducing factors were managed in order to mitigate confounding results from the experimental factors. As such, seeds were treated with azoxystrobin, fludioxonil and mefenoxam (Dynasty PD®, Syngenta Crop Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Plots were weeded on a biweekly basis and irrigated biweekly in the absence of rain. Data Collection: Two to three weeks after planting, stand counts were made for each plot. Final leaf spot and rust severity ratings were taken immediately prior to digging. Leaf spot severity was assessed with the Florida 1 to 10 scale (Appendix III). Rust severity was assessed with a modified 1 to 9 scale (Appendix III). Peanuts were manually harvested by first pulling the entire plant from the ground and removing all the attached pods from the plant. Afterwards, the soil in each plot was filtered through by hand to recover the remaining pods left in the ground. Pods were placed in large, green, mesh cabbage bags (Cady Bag Company, LLC. Pearson, GA) and washed after harvest in order to remove any remaining soil, and then placed on a large concrete pad to dry in the sun. The bagged pods were allowed to dry for a minimum of three days, and were moved under a shelter each night. After drying bags were weighed, and immediately afterwards, a 100-pod sample was shelled to obtain the moisture content of the kernels. Final weights were adjusted to 10% pod moisture. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 23 Statistical Analysis: For this report, means and standard errors of final severity of leaf spot and rust and yield were calculated across all trials with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). A more in-depth analysis of variance will be reported in a forthcoming publication in “Peanut Science.” Conclusion: While the 309 Tan variety showed excellent resistance to foliar disease, it did not respond as anticipated with increased yield and while other varieties showed a potential for yield increase over the local Haitian Valencia, it was not significant enough to warrant investment in large scale seed introduction. Until a higher performing variety is found, farmers interested in producing Valencia should continue with the local Haitian Valencia, which performed reasonably well and is widely available. Figure 2.2.a. Example of plot layout at MFK for the NMSU Valencia trials. Figure 2.2.c. 309 Red: fungicide (left of blue line) vs untreated plots (right of blue line). Pictures were from Trial 2 and taken on 21 June 2016. Figure 2.2.d. 309 Tan: fungicide (left of blue line) vs untreated plots (right of blue line). Pictures were from Trial 2 and taken on 21 June 2016. Figure 2.2.b. Means of final leaf spot and rust severity and pod yield pooled across all trials. Error bars represent the standard error of the mean. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 24 Chapter 2.3 2016 ACI Seed Variety Trials Purpose: Evaluate the performance of short maturing, high oleic acid breeding lines, developed by Dr. Kim Moore in Tifton, GA, through ACI Seeds for possible use in Haiti by comparing them to known standards. Experimental Design: Field trials were conducted at two locations in Haiti: MFK and the Acceso research farm in the Central Plateau. All trials were originally laid out in a random complete block design with four to six replications. However, experiments were changed to split plot design to allow for a comparison of varietal performance with fungicide vs. no fungicide. As such, main plot treatment was fungicide (with or without two applications starting at 60 days after planting) and sub-plot treatment was peanut variety. This resulted in three replications at MFK and two replications in the Central Plateau. Table 2.3.a. ACI Seeds breeding lines and additional varieties evaluated in these studies. Variety Description Market type WT 11-1120 ACI Seeds Runner N 11-0087 ACI Seeds Runner N 11-0029 ACI Seeds Runner M 15-1085 ACI Seeds Runner M 15-0069 ACI Seeds Runner 309 Red New Mexico State University Valencia 308 Red New Mexico State University Valencia 308 Tan New Mexico State University Valencia Local Valencia Haitian landrace Valencia Local Runner Haitian landrace Runner Georgia-06G University of Georgia Runner At MFK, plots were planted on 25 May 2016 and harvested on 24 August (Valencia types) and 27 September 2017 (runner types). At the Acceso farm in the Central Plateau, plots were planted on 12 May 2016 and harvested on 11 August (Valencia types) and 6 September 2017 (runner types). For all trials, peanuts were planted at a rate of six seed/30.5 cm in single row plots that were 0.6 m wide and 3.0 m in length. Each Blocks were separated by a 1.5-m alley. Other yield reducing factors were managed in order to mitigate confounding results from the experimental factors. As such, fields were fertilized with 50 kg/ha of 20-20-10 and seeds were treated with azoxystrobin, fludioxonil and mefenoxam (Dynasty PD, Syngenta Crop Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Plots were weeded on a biweekly basis, sprayed every 15 days (starting 30 days after planting) with fungicide (Muscle® ADV, Sipcam Agro USA, Inc.) and irrigated biweekly in the absence of rain. Data Collection: Two to three weeks after planting, stand counts were made for each plot. Final leaf spot and rust severity ratings were taken immediately prior to digging. Leaf spot severity was assessed with the Florida 1 to 10 scale (Appendix III). Rust severity was assessed with a modified 1 to 9 scale (Appendix III). Peanuts were manually harvested by first pulling the entire plant from the ground and removing all the attached pods from the plant. Afterwards, the soil in each plot was filtered through by hand to recover the remaining pods left in the ground. Pods were placed in large, green, mesh cabbage bags (Cady Bag Company, LLC. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 25 Pearson, GA) and washed after harvest in order to remove any remaining soil, and then placed on a large concrete pad to dry in the sun. The bagged pods were allowed to dry for a minimum of three days, and were moved under a shelter each night. After drying bags were weighed, and immediately afterwards, a 100 pod sample was shelled to obtain the moisture content of the kernels. Final weights (kg/ha) were adjusted to 10% pod moisture. A post-harvest test was made with the same 100 pod sample to evaluate the percentage of sound mature kernels (%SMK). Percent SMK was calculated by dividing the weight of the sound mature kernels by the total weight of the weight of the unshelled sample. Statistical Analysis: Stand count, leaf spot severity, rust severity and yield were subjected to analysis of variance with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). Each trial was analyzed separately and the model for each trial was a split plot design with fungicide and variety considered as fixed effects, with replication and replication × fungicide as random effects. In all analyses the Kenward-Roger option was used to adjust the degrees of freedom, and differences in the least square means were tested by Tukey’s multiple comparisons test. When data violated the assumptions of normality, transformations were used. Results: Table 2.3.b. P-values for the fixed effects from the analysis of variance. z Leaf spot Rust Stand countz Yield P-value P-value P-value P-value 0.2522 0.4729 0.8162 0.2347 Variety 0.0089 0.2313 <.0001 <.0001 CP Fungicide × Variety 0.5694 0.4346 0.6585 0.3375 MFK Fungicide <.0001 0.0019 0.2231 0.0356 MFK Variety <.0001 0.0001 <.0001 0.0014 MFK Fungicide × Variety <.0001 <.0001 0.3651 0.1981 Location Fixed effect CP Fungicide CP Number of plants emerged at ~ 3 weeks after planting divided by the number of seeds planted. Table 2.3.c. Effect of variety on stand count and yield at two locations in Haiti. Variety Central Plateau y MFK y Yield (kg/ha) Stand count z Stand count Yield (kg/ha) 308 Red 73 ab 2029 ab 97 ab 3526 abc 309 Red 43 bc 1642 ab 85 abc 3678 abc 309 Tan 75 ab 1253 b 98 a 2258 abc Georgia-06G 64 ab 2789 a 74 c 4391 a Local runner 50 ab 1440 ab 38 e 3700 abc Local Valencia 73 ab 1799 ab 83 abc 3606 abc M150069 49 ab 1810 ab 79 bc 1866 c M151085 78 a 1505 ab 88 abc 4126 ab N110029 69 ab 2109 ab 85 abc 2735 abc N110087 68 ab 1609 ab 90 abc 2474 abc WT11_112 23 c 350 c 52 d 2021 bc y Number of plants emerged at ~ 3 weeks after planting divided by the number of seeds planted. z Means within the same column with the same letters are not significantly different based upon Tukey’s honestly significant difference test. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 26 Figure 2.3.a. Final severity of leaf spot and rust for each variety in both treated and untreated plots at two locations in Haiti. Conclusion: The data do not suggest that it would be worthwhile to pursue large scale introduction of these varieties. The data do confirm the previous finding that the local Haitian Valencia, local Haitian Runner and Georgia 06-G maintained respectable yields. The data also suggest that planting at a higher density (six seed per 30.5cm) with the local Valencia may be advisable in both the North and Central Plateau. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 27 Chapter 2.4 2017 ICRISAT Variety Trials Purpose: Evaluate the performance of advanced breeding lines developed by ICRISAT (International Crops Research Institute for the Semi-Arid Tropics). Experimental Design: A trial was conducted at MFK that was laid out in a random complete block design with two replications. Table 2.4.a. Varieties evaluated were as follows. Treatment Trait Botanical type Branching habit Seed color ICGV 00338 Short-duration Spanish Sequential Tan ICGV 02038 Short-duration Spanish Sequential Tan ICGV 06237 Short-duration Spanish Sequential Tan ICGV 07210 Short-duration Spanish Sequential Tan ICGV 07235 Drought tolerant Spanish Sequential Tan ICGV 07286 Drought tolerant Spanish Sequential Tan ICGV 07390 Drought tolerant Spanish Sequential Tan ICGV 07396 Drought tolerant Spanish Sequential Tan ICGV 06138 Diseases resistant Spanish Sequential Tan ICGV 06175 Diseases resistant Virginia Alternate Tan ICGV 06176 Diseases resistant Virginia Alternate Tan ICGV 07120 Diseases resistant Spanish Sequential Tan Local Valencia - Valencia - Tan Local Runner - Runner - Tan Georgia-06G - Runner - Tan At MFK, plots were planted on 6 April 2017 and harvested on 5 July (Valencia and Spanish types) and 4 August 2017 (Virginia and runner types). However, it should be noted that the maturity was not evaluated prior to harvest. Peanuts were planted at a rate of three seed/30.5 cm in single row plots that were 0.3 m wide and 4.5 m in length. Each block was separated by a 1.5 m-alley. Other yield-reducing factors were managed in order to mitigate confounding results from the experimental factors. As such, fields were fertilized with 50 kg/ha of 20-20-10 and seeds were treated with azoxystrobin, fludioxonil, and mefenoxam (Dynasty PD®, Syngenta Crop Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Plots were weeded on a biweekly basis, sprayed every 15 days (starting 30 days after planting) with fungicide (Muscle® ADV, Sipcam Agro USA, Inc.) and irrigated biweekly in the absence of rain. Data Collection: Two to three weeks after planting, stand counts were made for each plot. Final leaf spot and rust severity ratings were taken immediately prior to digging. Leaf spot severity was assessed with the Florida 1 to 10 scale (Appendix III). Rust severity was assessed with a modified 1 to 9 scale (Appendix III). Peanuts were manually harvested by first pulling the entire plant from the ground and removing all the attached pods from the plant. Afterwards, the soil in each plot was filtered through by hand to recover the remaining pods left in the ground. Pods were placed in large, green, mesh cabbage bags (Cady Bag Company, LLC. Pearson, GA) and washed after harvest in order to remove any remaining soil, and then placed on a large concrete pad to dry in the sun. The bagged pods were allowed to dry for a minimum of three days, and were moved under a shelter each night. After drying bags were weighed, and immediately afterwards, a 100-pod UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 28 sample was shelled to obtain the moisture content of the kernels. Final weights (kg/ha) were adjusted to 10% pod moisture. A post-harvest test was made with the same 100-pod sample to evaluate the percentage of sound mature kernels (%SMK). Percent SMK was calculated by dividing the weight of the sound mature kernels by the total weight of the weight of the unshelled sample. Statistical Analysis: Final severity of stand count, leaf spot severity, rust severity and yield were subjected to analysis of variance with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). In all analyses the Kenward-Roger option was used to adjust the degrees of freedom, and differences in the least square means were tested by Tukey’s multiple comparisons test. When data violated the assumptions of normality, transformations were used. Results: Table 2.4.b. Effect of variety on stand count, leaf spot, rust, and yield. Trt Treatment Trait Botanical type Stand count Leafspot Rust Kg/ha 1 ICGV 00338 Short-duration Spanish 100.0 a 1.0 a 1.25 a 2211.0 ab 2 ICGV 02038 Short-duration Spanish 100.0 a 1.0 a 1.00 a 2182.1 ab 3 ICGV 06237 Short-duration Spanish 100.0 a 1.0 a 1.25 a 1309.2 b 4 ICGV 07210 Short-duration Spanish 100.0 a 1.0 a 1.00 a 1789.6 ab 5 ICGV 07235 Drought tolerant Spanish 65.0 c 1.0 a 1.00 a 2442.7 ab 6 ICGV 07286 Drought tolerant Spanish 86.0 ab 1.0 a 1.00 a 2584.0 ab 7 ICGV 07390 Drought tolerant Spanish 79.0 abc 1.0 a 1.00 a 3042.8 ab 8 ICGV 07396 Drought tolerant Spanish 92.0 ab 1.0 a 1.00 a 2433.3 ab 9 ICGV 06138 Diseases resistant Spanish 96.0 a 1.0 a 1.00 a 2311.0 ab 10 ICGV 06175 Diseases resistant Virginia 88.0 ab 1.0 a 1.00 a 3532.2 a 11 ICGV 06176 Diseases resistant Virginia 95.0 a 1.0 a 1.00 a 2880.6 ab 12 ICGV 07120 Diseases resistant Spanish 92.0 ab 1.0 a 1.00 a 3008.2 ab 13 Local Valencia Valencia 99.0 a 1.0 a 1.00 a 1869.9 ab 14 Local Runner Runner 62.0 c 1.0 a 1.00 a 1200.1 b 15 Georgia-06G Runner 73.0 bc 1.0 a 1.00 a 2925.8 ab LSD P=.05 Standard Deviation CV 13.16 6.13 6.93 0.00 0.00 0.0 0.267 0.124 12.04 1179.14 549.72 23.08 Replicate F Replicate Prob (F) Treatment F Treatment Prob (F) 2.215 0.1588 8.938 0.0001 0.000 1.0000 0.000 1.0000 2.154 0.1643 1.000 0.5000 5.097 0.0405 2.826 0.0308 UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 29 Figure 2.4.a. Plots with ICRISAT varieties in a field at MFK taken on 5 June, 2017. Conclusion: This trial had two objectives: 1) to evaluate the potential of these advanced lines from the ICRISAT breeding program in India and 2) to multiply seed for future multi-location trials, which is why fungicide and irrigation were used to assure maximum yield. These initial data suggest that these lines do hold promise and should be evaluated under field stress situations in multiple locations. Chapter 2.5 2016-2017 Tillman Breeding Line Screenings Purpose: To evaluate the performance of advanced breeding lines developed by Barry Tillman for resistance to foliar diseases in Haiti. Experimental Design: In the spring of 2016, 45 breeding lines were compared to the local Haitian landraces and the predominant runner variety planted in the southeastern U.S., Georgia-06G. In the fall of 2016, the best performing lines were selected for further screening and compared to the same known standard varieties as the first trial. Both trials were laid out in a random complete block design. In the first trial, there were two replications and in the second trial there were three replications. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 30 The first trial was planted on 23 March 2016 and harvested on 20 July. The second trial was planted on 10 October 2016 and harvested on 7 February 2017. In the first trial, peanuts were planted at a rate of 25 seed/row in single row plots that were 0.6 m wide and 2.4 m in length (~3 seed/0.3m). In the second trial, peanuts were planted at a rate of 3 seed/0.3 m in single row plots that were 0.6 m wide and 4.5 m in length. In both trials, each block was separated by a 1.5-m alley. Other yield-reducing factors were managed in order to mitigate confounding results from the experimental factors. As such, fields were fertilized with 50 kg/ha of 20-20-10 and seeds were treated with azoxystrobin, fludioxonil and mefenoxam (Dynasty PD®, Syngenta Crop Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Plots were not sprayed with fungicide, but were weeded on a biweekly basis and irrigated biweekly in the absence of rain. Data collection. Two to three weeks after planting, stand counts were made for each plot. Final leaf spot and rust severity ratings were taken immediately prior to digging. In the first trial, leaf spot severity was assessed with the Florida 1 to 10 scale (Appendix III). Rust severity was assessed with a modified 1 to 9 scale (Appendix III). In the second trial, five leaves per plot were sampled and the number of leaf spot lesions were counted. Rust severity per leaflet was estimated on a 0 to 100 % scale. Plots were sampled at 60, 75, 90, 105, and 120 DAP. For the number of leaf spot lesions and the percent rust on the leaflet, AUDPC values were calculated and standardized (stAUDPC) by dividing AUDPC values by the number of days between the first and final evaluation. Peanuts were manually harvested by first pulling the entire plant from the ground and removing all of the attached pods from the plant. Afterward, the soil in each plot was filtered through by hand to recover the remaining pods left in the ground. Pods were placed in large, green, mesh cabbage bags (Cady Bag Company, LLC. Pearson, GA) and washed after harvest in order to remove any remaining soil, and then placed on a large concrete pad to dry in the sun. The bagged pods were allowed to dry for a minimum of three days, and were moved under a shelter each night. After drying, bags were weighed. Statistical analysis. Means were calculated for final severity of stand count, leaf spot severity, rust severity and yield were subjected to analysis of variance with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). Results Figure 2.5.a. Plots from the first trial at MFK on 1 June 2016. Figure 2.5.b. Plots from the first trial at MFK on 21 June 2016. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 31 Figure 2.5.c. Trial 1: Tillman breeding line screening conducted at MFK during the spring/summer of 2016. Seed germination = stand count (number of plants emerged at ~ 3 weeks after planting divided by the number of seeds planted) UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 32 Figure 2.5.d. Trial 2: Tillman breeding line screening conducted at MFK during the fall/winter of 2016/2017. Conclusion: Some of these varieties show promise and should be multiplied and evaluated in multiple locations. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 33 Chapter 3.1 Inoculant × Fertilizer Interaction Studies Purpose: Determine whether rhizobia nodulation, fertilizer or a combination of the two factors are yieldlimiting factors for peanut production at two sites in Haiti. Experimental Design: Two inoculant/fertility trials were conducted during the spring of 2015. One trial was at the Meds & Food for Kids factory in the community of Quartier Morin, Haiti and the other at the Acceso research farm in the Central Plateau. Both trials were laid out in a split plot design with four replications. Presence or absence of liquid Bradyrhizobia (Optimize® liquid inoculant for peanut; Novozymes, Inc., Bagsvaerd, Denmark) was the main plot treatments, and granular fertilizer type was the subplot treatments. At MFK, there were four subplot treatments – diammonium phosphate (DAP) at a high rate (67.2 kg ha-1), DAP at a low rate (22.4 kg ha-1), monoammonium phosphate (MAP) at a high rate (67.2 kg ha-1) and an untreated control plot. Prior to planting, fields were disc plowed two to three times, and/or rototilled within two days prior to planting. On the day prior to planting, the various granular fertilizer treatments were applied (an even broadcast) and rototilled twice at the rates mentioned above to the respective plots. The liquid inoculant was applied in-furrow with a hand pumped backpack sprayer at a rate of 1.0 oz/1,000 feet of row (29.57 mL/304.8 m of row). Dates from planting to harvest, respectively, for MFK and Acceso were as follows: 26 March 2015 to 1 July 2015 and 29 April 2015 to 30 July, 2015. For all trials, the New Mexico Valencia A was planted at a rate of three seed/0.3m in two rows spaced 0.762 m apart in plots that were 1.5 m wide and 3.04 m in length. Blocks were separated by a 1.5-m alley. Other yield-reducing factors were managed in order to mitigate confounding results. As such, seeds were treated with azoxystrobin, fludioxonil and mefenoxam (Dynasty PD®, Syngenta Crop Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Plots were weeded on a biweekly basis, sprayed every 15 days (starting 30 days after planting) with fungicide (Muscle® ADV, Sipcam Agro USA, Inc.) and irrigated biweekly in the absence of rain. Data Collection: Two to three weeks after planting, stand counts were made for each plot. Final leaf spot and rust severity ratings were taken immediately prior to digging. Leaf spot severity was assessed with the Florida 1 to 10 scale (Appendix III). Rust severity was assessed with a modified 1 to 9 scale (Appendix III). Peanuts were manually harvested by first pulling the entire plant from the ground and removing all the attached pods from the plant. Afterwards, the soil in each plot was filtered through by hand to recover the remaining pods left in the ground. Pods were placed in large, green, mesh cabbage bags (Cady Bag Company, LLC. Pearson, GA) and washed after harvest in order to remove any remaining soil, and then placed on a large concrete pad to dry in the sun. The bagged pods were allowed to dry for a minimum of three days, and were moved under a shelter each night. After drying bags were weighed, and immediately afterwards, a 100-pod sample was shelled to obtain the moisture content of the kernels. Final weights (kg/ha) were adjusted to 10% pod moisture. A post-harvest test was made with the same 100-pod sample to evaluate the percentage of sound mature kernels (%SMK). Percent SMK was calculated by dividing the weight of the sound mature kernels by the total weight of the unshelled sample. Statistical Analysis: Final severity of stand count, yield and %SMK were subjected to analysis of variance with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). Each trial was analyzed separately and the model for each trial was a split plot design with inoculant and fertilizer considered as fixed effects, with replication and replication × inoculant as random effects. In all analyses, the Kenward-Roger option was used to adjust the degrees of freedom, and differences in the least square means were tested by Tukey’s multiple comparisons test. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 34 Table 3.1.a. Two-way analysis of variance for stand count, yield and %SMK for each location in Haiti. Location Fixed effect CP Stand count Yield (KgHa) %SMK F-value P-value F-value P-value F-value P-value Fungicide 3.05 0.1013 0.15 0.7279 - - CP Variety 1.13 0.3482 0.72 0.5088 - - CP Fungicide × Variety 0.01 0.9923 0.48 0.6326 - - MFK Fungicide 4.08 0.0899 0.01 0.932 0.97 0.3689 MFK Variety 0.91 0.4546 1.47 0.2618 0.59 0.6304 MFK Fungicide × Variety 0.24 0.8666 1.14 0.3641 0.99 0.4226 Table 3.1.b. Effect of inoculant and fertilizer on stand count, yield and %SMK at each location in Haiti. Location Fixed effect Treatment CP Inoculant Inoculated 71.9 az 532 a - Non-inoculated 67.4 a 509 a - High DAP 67.1 a 511 a - Low DAP 71.8 a 554 a - None 70.0 a 497 a - Inoculant Inoculated 89.0 a 3456 a 79.2 a Fertilizer Non-inoculated High DAP 95.4 91.0 a a 3430 3337 a a 84.8 86.7 a a High MAP 93.9 a 3450 a 78.0 a Low DAP 94.3 a 3578 a 81.7 a None 90.6 a 3411 a 81.8 a Fertilizer MFK z Stand count Yield (KgHa) %SMK Means within the same column with the same letters are not significantly different based upon Tukey’s honestly significant difference test. Conclusion: The data suggest that the crop does not respond to either inoculation or phosphorus fertilizer. It may be that adequate inoculum of Bradyrhizobia was present in the soil as evidenced by the level of nodulation on the untreated plots. However, this may or may not hold true for areas that have not had recent peanut cultivation and could require future research. Peanut roots with nodules containing Rhizobium bacteria from non-inoculant treated plots at MFK. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 35 Chapter 3.2 Foliar × Granular Fertilizer Interaction Trials Purpose: Determine if foliar applications of micronutrient foliar fertilizer, various rates of granular fertilizers, including high nitrogen granular fertilizers, or a combination of the two inputs can increase pod yield in high pH soils in Haiti. Experimental Design: Two fertility trials were conducted during the spring and summer of 2017 at the Meds & Food for Kids factory in the commune of Quartier Morin, Haiti. Both trials were laid out in a split plot design with four replications. Presence or absence of foliar fertilizer was the main plot treatments, and granular fertilizer type was the subplot treatments. Dates from planting to harvest, respectively, for each trial were as follows: 3 March 2017 to 1 June 2017; 11 April 2017 to 10 July 2017. For all trials, the local Haitian Valencia was planted in plots that were 1.2 m wide and 4.6 m in length; rows were spaced 0.6 m apart, and one untreated row was planted between each treatment plot. Blocks were separated by a 1.5-m alley. For the main plot, Nurish® (FERSAN, Santo Domingo, Dominican Republic) fertilizer was applied one time to all treated plots in the first trial; the concentrated fertilizer was diluted in water to a concentration of 1951 ppm. Solubor® 20.5% elemental boron soluble liquid organic fertilizer (Rio Tinto, Inc., London, United Kingdom) was also applied once to the both trials at a rate of 2.4 kg ha-1, so as to provide 0.6 kg ha-1 of elemental boron to the plants. The subplot consisted for four treatments; namely, 112.1 kg ha-1 20-20-10 N-P-K, 44.8 kg ha-1 diammonium phosphate (DAP), 112.1 kg ha-1 urea, and an untreated control plot. Table 3.2.a. Ingredients contained in the foliar fertilizer Nurish. Total Nitrogen (soluble) NO3 NH4 Water soluble organic Nitrogen Phosphorus P2O5 Potassium K2O 20% 5.94% 3.91% 10.15% 20% 20% Iron Zinc Manganese Copper Boron Molybdenum 1500 ppm 750 ppm 750 ppm 750 ppm 300 ppm 105 ppm The various granular fertilizer treatments were applied at the rates mentioned above to the respective treatments at 28 and 23 days after sowing for the first and second trials, respectively. All granular fertilizers were applied using the side dressing method by tracing a line perpendicular to each treatment row approximately 6 cm away from the row and spreading the fertilizer along this line by hand; after the fertilizer was spread, it was covered up with a thin layer of soil. Nurish® foliar micronutrient fertilizer was applied to the first trial at 45 days after sowing using a one-gallon hand-held pump sprayer, and was not applied to the second trial. This is because the research team had already obtained foliar boron fertilizer by the time the second trial was planted, and boron was judged to be the most plant-limiting micronutrient for peanut cultivation in the area and hence the most likely to have a positive effect on growth. To this effect, Solubor® foliar boron fertilizer was applied at 61 and 43 days after sowing to the first and second trials, respectively; the same one-gallon hand-held pump sprayer that was used to apply the Nurish® fertilizer was also used to apply the Solubor® fertilizer. Both foliar fertilizers were applied to the entirety of the rows making up the treatment plots. Other yield-reducing factors were managed in order to mitigate confounding results from the experimental factors. As such, seeds were treated with azoxystrobin, fludioxonil and mefenoxam (Dynasty PD®, Syngenta Crop Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Plots were weeded on a biweekly basis, sprayed every 15 days (starting 30 days after planting) with fungicide (Muscle® ADV, Sipcam Agro USA, Inc.) and irrigated biweekly in the absence of rain. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 36 Data Collection: Two to three weeks after planting, stand counts were made for each plot; average stand count across all plots was 71.4% for Soil Fertility Trial No. 1 and 91.8% for Soil Fertility Trial No. 2. Final leaf spot and rust severity ratings were taken immediately prior to digging. Leaf spot severity was assessed with the Florida 1 to 10 scale (Appendix III). Rust severity was assessed with a modified 1 to 9 scale (Appendix III). Peanuts were manually harvested by first pulling the entire plant from the ground and removing all the attached pods from the plant. Afterwards, the soil in each plot was filtered by hand to recover the remaining pods left in the ground. Pods were placed in large, green, mesh cabbage bags (Cady Bag Company, LLC. Pearson, GA) and washed after harvest in order to remove any remaining soil, and then placed on a large concrete pad to dry in the sun. The bagged pods were allowed to dry for a minimum of three days, and were moved under a shelter each night. After drying bags were weighed, and immediately afterwards, a 100-pod sample was shelled to obtain the moisture content of the kernels. Final weights were adjusted to 10% pod moisture. A post-harvest test was made with the same 100-pod sample to evaluate the percentage of sound mature kernels (%SMK). Percent SMK was calculated by dividing the weight of the sound mature kernels by the total weight of the unshelled sample. Statistical Analysis: Pod yield was subjected to analysis of variance with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). The model was a split-split plot design with trial, foliar fertilizer and granular fertilizer considered as fixed effects, and replication, replication × trial and replication × foliar fertilizer as random effects. The Kenward-Roger option was used to adjust the degrees of freedom, and differences in the least square means were tested by Tukey’s multiple comparisons test. Table 3.2.b. P values from the three-way analysis of variance for stand count, yield and %SMK. z Effect Stand countz Yield (kg/ha) %SMK Trial 0.0056 0.0067 0.0249 Fertilizer 0.9211 0.932 0.0018 Fertilizer × Trial 0.6612 0.5688 0.1052 Foliar Fertilizer 0.1381 0.6065 0.1877 Foliar Fertilizer × Trial 0.1705 0.8087 0.4659 Fertilizer × Foliar Fertilizer 0.7121 0.4554 0.7346 Fertilizer × Foliar Fertilizer × Trial 0.1691 0.8227 0.5202 Number of plants emerged at ~ 3 weeks after planting divided by the number of seeds planted. Fig 3.2.a. Plots from the first trial at MFK prior to harvest. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 37 Table 3.2.c. Effect of trial, granular fertilizer and foliar fertilizer for two field trials conducted at MFK. Effect Trial Stand county Yield (kg/ha) %SMK 1 94.2 az 3326.3 az 65.6 a 2 70.7 b 2711.2 b 60.2 b 20-20-10_100 lbs 85.2 a 3047.8 a 64.3 a Urea_100 lbs 84.5 a 3002.8 a 63.9 a DAP_40 lbs 84.0 a 2988.1 a 63.2 a Untreated 83.3 a 2974.1 a 60.2 b Treated 85.7 a 3024.8 a 63.4 a Untreated 82.7 a 2981.5 a 62.4 a Granular fertilizer Foliar fertilizer y Number of plants emerged at ~ 3 weeks after planting divided by the number of seeds planted. z Means within the same column with the same letters are not significantly different based upon Tukey’s honestly significant difference test. Conclusion: These data suggest that the crop does not significantly respond to foliar micronutrient or granular fertilization. However, these plots were completed on land that is known to have received previous fertility treatment and plants in areas with more deficient soil may respond differently. Chapter 3.3 2016 Fulvic Acid Trials Purpose: Determine the effect of fulvic acid on pod yield. It is hypothesized that fulvic acid would allow more nutrient availability in the high pH soils found in Haiti. Experimental Design: Plots were laid out in a random complete-block design with four replications. Treatments were as follows: 1. 2. 3. 4. 5. 55kg/ha Fulvic acid 70 kg/ha Fulvic acid 70 kg/ha Fulvic acid + 70kg/ha 20-20-10 120 kg/ha Fulvic acid Untreated The local Haitian Valencia was planted in plots that were 1.5 m wide and 3.04 m long with three rows spaced 0.4 m apart at a rate of three seed/30.5 cm. Each plot was separated by 1.2 m fallow, buffer zone, and blocks were separated by a 1.5 m alley. Plots were planted on 11 December 2015 and harvested on 10 March 2016. Fulvic Acid + 20-20-10 were weighed before and applied after sowing 2-5 cm away from seed. Seed and fertilizer depth were identical. Other yield-reducing factors were managed in order to mitigate confounding results from the experimental factors. As such, seeds were treated with azoxystrobin, fludioxonil and mefenoxam (Dynasty PD®, Syngenta Crop Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Plots were weeded on a biweekly basis, sprayed every 15 days (starting 30 days after planting) with fungicide (Muscle® ADV, Sipcam Agro USA, Inc.) and irrigated biweekly in the absence of rain. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 38 Results: Table 3.3.a. Effect of fulvic acid treatment on stand count and pod yield at MFK. Trt. Treatment description Stand count Yield 1 Fulvic Acid (Low) 55 kg/ha 0.89 a 2707.9 a 2 Fulvic Acid (High) 70 kg/ha 0.89 a 2205.1 b 3 Fulvic Acid (High) 70 kg/ha + 20-20-10 (70kg/ha) 0.89 a 2158.9 b 4 Fulvic Acid (Very High) 120 kg/ha 0.86 a 2214.3 b 5 Untreated 0.91 a 2445.0 ab Tukey’s HSD P=.05 Standard Deviation CV 0.155948 0.069157 7.76 436.37 190.97 8.14 Replicate F Replicate Prob(F) Treatment F Treatment Prob(F) 1.915 0.1811 0.235 0.9131 16.041 0.0002 5.839 0.0090 Fig 3.3.a. Applying a protective fungicide cover spray on the plots at MFK. Conclusion: Current data do not support the use of fulvic acid as an input for peanut production in Haiti. Further research could consider different products, application rates or trial locations. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 39 Chapter 4.1 Seed/Row Spacing Trials Purpose: The primary objective was to determine the best planting density for optimizing yield in runner and bunch-type peanuts. A secondary objective was to determine the effect of seed/row spacing on virus intensity. Experimental Design: A total of five seed/row spacing trials were conducted from 2015 to 2017. The first two were conducted at two locations during the summer of 2015. These locations included the MFK research site and a local university-owned property in Trou-Du-Nord. However, it should be noted that the first two trials were preliminary and did not include the 12-inch row-spacing, did not have a buffer area between plots, were not irrigated as heavily and had fewer records available. Therefore, we will only include the yield data as an appendix (see Appendix V). In 2016, two seed-spacing trials were conducted at the MFK research site in spring and fall, and the last trial was conducted at the same site in the summer of 2017. All trials were laid out in a split-split plot design, with five replications for Trial No. 1 and four replications for Trials No. 2 and No. 3. Variety was the main plot treatment, row spacing was the subplot treatment, and seed spacing was the sub-subplot treatment. Table 4.1.a. Factors and associated treatment levels for seed/row spacing trials in 2016 and 2017. Split-split plot design Factor Main plot Variety Subplot Row spacing (distance between rows) Sub-subplot Seed spacing (distance within row) Level Local Haitian runner Local Haitian Valencia 30.5 cm (12 in.) 45.7 cm (18 in.) 61.0 cm (24 in.) 3.3 seeds m-1 (1 seed/ft.) 9.8 seeds m-1 (3 seed/ft.) 19.7 seeds m-1 (6 seed/ft) For all trials, plots were 1.8 m wide and 3.0 m in length. Plots were separated by 0.61 m unplanted space and blocks were separated by a 1.5 m alley (with the exception of the 2017 trial which had a 0.61 m alley). In all plots, target rows were first marked with stakes and string, and hoes were used to create furrows with an average depth of 3.8 cm. The number of furrows made for 30.5, 45.7 and 61.0 cm row spacings was six, four and three, respectively. Plots were planted by hand, and uniformity was ensured by placing PVC pipes (marked with the appropriate respective seeding rates) within the furrow while planting. Dates from planting to harvest, respectively, for each trial were as follows: Trial No. 1: 11 March 2016 to 14 June, 2016 (Valencia) and 13 July 2016 (runner); Trial No. 2: 14 October 2016 to 14 January 2017 (Valencia) and 16 February 2017 (runner); Trial No. 3: 26 May 2017 to 24 August 2017 (Valencia) and 27 September 2017 (runner). Other yield-reducing factors were managed in order to mitigate confounding results from the experimental factors. As such, fields were disked two to three times prior to planting, and rototilled within two days prior to planting. 20-20-10 N-P-K fertilizer was applied to each of the study fields at a rate of 112.1 kg ha-1. Prior to planting, seeds were treated with azoxystrobin, fludioxonil and mefenoxam (Dynasty PD®, Syngenta Crop Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Plots were weeded on a biweekly basis, sprayed every 15 days (starting 30 days after planting) with fungicide (Muscle® ADV, Sipcam Agro USA, Inc.) and irrigated biweekly in the absence of rain. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 40 Data Collection: Two to three weeks after planting, stand counts were made for each plot. Virus intensity was assessed as the number of 0.3 m sections of peanut plants with symptomatic infection (see Appendix VI for detailed virus symptoms). Virus ratings were made at 5 May and 8 December 2016, for the first and second trial, respectively. Ratings were not taken for 2017 trial. Peanuts were manually harvested by first pulling the entire plant from the ground and removing all the attached pods from the plant. Afterwards, the soil in each plot was filtered by hand to recover the remaining pods left in the ground. Pods were placed in large, green, mesh cabbage bags (Cady Bag Company, LLC. Pearson, GA) and washed after harvest in order to remove any remaining soil, and then placed on a large concrete pad to dry in the sun. The bagged pods were allowed to dry for a minimum of three days, and were moved under a shelter each night. After drying bags were weighed, and immediately afterwards, a 100-pod sample was shelled to obtain the moisture content of the kernels. Final weights were adjusted to 10% pod moisture. A post-harvest test was made with the same 100-pod sample to evaluate the percentage of sound mature kernels (%SMK). Percent SMK was calculated by dividing the weight of the sound mature kernels by the total weight of the unshelled sample. Statistical Analysis: Yield and %SMK were subjected to analysis of variance with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). Due to differences in seed spacing between trials, and to better understand the effect of each factor for a given environment, each trial was analyzed separately. For all three trials, the model was a split-split plot design with variety, row spacing, and seed spacing considered as fixed effects, and with replication and replication × variety as random effects. Due to significant variety × row spacing and variety × seed spacing interactions (α = 0.05), the effect of seed spacing was analyzed by variety and row spacing for the main response variable. The SLICE option in SAS was used to explore all two-way interactions. In all analyses, the Kenward-Roger option was used to adjust the degrees of freedom, and differences in the least square means were tested by Tukey’s multiple comparisons test. Figure 4.1.a. 2017 Seed- and row-spacing trial at MFK taken 5 June. 2017. Figure 4.1.b. 2017 Seed- and row-spacing trial at MFK taken 28 August. 2017. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 41 Table 4.1.b. P-values from the analysis of variance for virus, stand count, pod yield and %SMK for each trial and variety. Trial Variety 1 Runner Valencia 2 Runner Valencia 3 Runner Valencia z Effect Stand countz Kg/ha % SMK Between-row (B) 0.2868 0.0222 0.5072 Within-row (W) <.0001 0.0001 0.0002 B×W 0.3551 0.5495 0.1799 Between-row (B) 0.8325 <.0001 0.8847 Within-row (W) <.0001 <.0001 0.0424 B×W 0.8259 0.0006 0.0808 Between-row (B) 0.8122 0.439 0.3871 Within-row (W) 0.0511 <.0001 0.6163 B×W 0.3877 0.0047 0.9347 Between-row (B) 0.0260 0.0458 0.9809 Within-row (W) 0.0258 <.0001 0.7215 B×W 0.5390 0.0200 0.6823 Between-row (B) 0.1732 0.7546 0.0595 Within-row (W) 0.0113 0.2352 0.9072 B×W 0.0384 0.7218 0.4646 Between-row (B) 0.9861 0.0025 0.1895 Within-row (W) 0.0287 <.0001 0.0306 B×W 0.1181 0.5169 0.128 Number of plants emerged at ~ 3 weeks after planting divided by the number of seeds planted. Table 4.1.c. Effect of between-row spacing on pod yield for each within-row seed spacing for runner and Valencia market types for each trial conducted at MFK during 2016 and 2017. Yield (kg/ha) Trial 2 Variety Within-row spacing Between-row spacing Runner .3 meter (12 inches) 5081 a 1703 A 3557 a .46 meter (18 inches) 3492 b 1077 B 3479 a .6 meter (24 inches) 3967 ab 1083 B 3333 a .3 meter (12 inches) 5649 a 1746 A 3122 a .46 meter (18 inches) 4691 a 1805 A 3257 a .6 meter (24 inches) 5055 a 1854 A 3400 a .3 meter (12 inches) 6322 a 1931 A 2943 a .46 meter (18 inches) 5506 a 1950 A 3317 a .6 meter (24 inches) 5314 a 1987 A 3150 a .3 meter (12 inches) 4202 a 1553 A 2128 a .46 meter (18 inches) 2859 b 1275 A 2010 a .6 meter (24 inches) 2084 c 844 B 1899 a .3 meter (12 inches) 6900 a 2328 Ab 3304 a .46 meter (18 inches) 5955 a 2966 A 3059 ab .6 meter (24 inches) 4923 b 1949 B 2220 b .3 meter (12 inches) 7596 a 3691 A 3820 a .46 meter (18 inches) 7517 a 3519 Ab 3571 ab .6 meter (24 inches) 6086 b 2408 B 2740 b 3.3 per meter 9.8 per meter 19.7 per meter Valencia 3.3 per meter 9.8 per meter 19.7 per meter Trial 1 UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report Trial 3 42 Figure 4.1.c. Effect of seed spacing on total pod yield (kg/ha) for each market type and row spacing. Grouped bars for each row spacing treatment with the same letters are not significantly different based upon Tukey’s honestly significant difference test. Figure 4.1.d. Effect of seed spacing on final percent stand count for each market type and row spacing. Grouped bars for each row spacing treatment with the same letters are not significantly different based upon Tukey’s honestly significant difference test. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 43 Fig 4.1.e. Mean number of 0.3 m sections of peanuts with symptomatic viral infection. Error bars represent the standard error of the mean. Table 4.1.d. Effect of variety, between-row spacing and within-row seed spacing on percent virus incidence for runner and Valencia market types for two trials conducted at MFK during 2016. Variety Effect Between-row spacing Runner Within-row spacing Between-row spacing Valencia Within-row spacing Treatment % SMK Trial 2 Trial 1 Trial 3 .3 meter (12 inches) 71.0 a 66.0 a 59.3 ab .46 meter (18 inches) 70.5 a 62.4 a 61.6 a .6 meter (24 inches) 69.8 a 62.7 a 56.8 b 1 seed/30.5 cm (1 seed/ft) 67.8 b 64.4 a 58.9 a 3 seed/30.5 cm (3 seed/ft) 70.9 a 64.7 a 59.2 a 6 seed/30.5 cm (6 seed/ft) 72.7 a 62.1 a 59.7 a .3 meter (12 inches) 65.9 a 72.0 a 60.2 a .46 meter (18 inches) 66.6 a 72.2 a 59.7 a .6 meter (24 inches) 66.2 a 71.9 a 56.8 a 1 seed/30.5 cm (1 seed/ft) 64.1 b 71.5 a 56.0 b 3 seed/30.5 cm (3 seed/ft) 67.3 a 72.7 a 59.2 ab 6 seed/30.5 cm (6 seed/ft) 67.3 a 72.0 a 61.5 a Conclusions: These trials strongly suggest that the ideal planting density is not the same for the local runner and the local Valencia. Effect of planting density on virus intensity was inconsistent, partly due to limited disease pressure. However, these data suggest that in years with heavier virus incidence (e.g., 2016), higher planting densities could reduce the number of infected plants; more research is needed to substantiate this interpretation. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 44 Valencia. Overall, regardless of between-row spacing or within-row spacing, the Valencia variety yields consistently increased with increasing planting density (Table 4.1.c and Fig. 4.1.c). However, there is generally less of a yield gap between 30.5 and 45.7 cm row spacing than 45.7 and 61 cm row spacing, and 9.8 and 19.7 seed/m than between 3.3 and 9.8 seed/m (Fig. 4.1.c). Runner. We did not find the same consistency in the response to seed/row spacing treatments for the runner variety (Table 4.1.c and Fig. 4.1.c). However, yield in plots with 3 and 6 seed/ft within-row spacing were more often higher than plots with the 1 seed/ft spacing (Fig. 4.1.c). Row spacing did not have an effect on yield when planted at 3 or 6 seed/ft (Table 4.1.c), suggesting the within-row spacing is more important for the runner variety. Chapter 4.2 Planting Method Trials: Rows vs. Traditional Scatter Planting Purpose: Our recommendation has been to plant in rows, yet many growers in Haiti do not plant in this way. Therefore, the objective was to determine if there is a yield benefit to planting in rows vs. the traditional scatter method. The second objective was to determine if there is a yield benefit from increasing the planting density with the traditional scatter method for both runner and Valencia market types. The trial involved yield, but not labor time or other measures. This research does not address the question of total economic return of row- vs scatter-planting methods. Challenges for Interpretation: The number of seeds is not always the same for planting in rows vs. scatter. There is no feasible way to use the same amount of plot space to mimic both traditional methods and planting in rows and keep a uniform number of seeds per plot. Therefore, a compromise was made by selecting a plot size that would allow for making a direct comparison of the two planting methods with the same number of seeds in several (but not all) scenarios. Experimental Design: Two planting method trials were conducted at the MFK research site in 2017. Both trials planted at MFK were laid out in a split-plot design with four replications. Market type was the main plot (local Haitian Valencia landrace and local Haitian runner landrace) and planting method was the subplot (see Table 4.2.a for details). An additional trial with only the local Haitian Valencia landrace market type was conducted at the Acceso research site in Mirebalais in 2017. This trial was laid out in randomized completeblock design with six replications. Dates from planting to harvest, respectively, for each trial were as follows: MFK No. 1: 13 March 2017 to 11 June (Valencia) and 11 July (runner) 2017. MFK No. 2: 4 April 2017 to 3 July 2017 (Valencia) and 25 April 2017 to 4 September 2017 (runner). Mirebalais: 5 May 2017 to 9 August 2017. Table 4.2.a. Planting method treatments used in trials conducted at MFK during 2017. Treatment description English units Metric units 24 Inch Row (1 seed/ft) 61.0 cm Row (1 seed/30.5 cm) 12 Inch Row (1 seed/ft) 30.5 cm Row (1 seed/30.5 cm) 24 Inch Row (3 seed/ft) 61.0 cm Row (3 seed/30.5 cm) 12 Inch Row (3 seed/ft) 30.5 cm Row (3 seed/30.5 cm) 18 Inch Scatter (1 seed/divot) 45.7 cm Scatter (1 seed/divot) 12 Inch Scatter (1 seed/divot) 30.5 cm Scatter (1 seed/divot) 18 Inch Scatter (2 seed/divot) 45.7 cm Scatter (2 seed/divot) 12 Inch Scatter (2 seed/divot) 30.5 cm Scatter (2 seed/divot) UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report Total seed planted Total rows 30 60 90 180 30 60 60 120 3 6 3 6 45 For all trials, plots were 1.8 m wide and 3.0 m in length. Each plot was separated by a 0.6 m alley within blocks, and blocks were separated by a 1.5 m alley. For the plots planted in rows, target rows were first marked with stakes and string, and hoes were used to create furrows with an average depth of 3.8 cm. Fields were disked two to three times prior to planting, and rototilled within two days prior to planting. 20-20-10 N-P-K fertilizer was applied to each of the study fields at a rate of 112.1 kg ha-1. Plots were planted by hand, and uniformity was ensured by placing PVC pipes (marked with the appropriate seeding rate) within the furrow while planting. For the scatter-planted plots, individual furrows/divots were created with hoes in a randomized fashion while respecting the appropriate planting distance in accordance with treatment; furrows were made with an average depth of 3.8 cm. The scatter plots were then planted by hand by placing either one or two seeds into each individual furrow in accordance with the treatment specifications for each individual plot. Due to germination issues in the first trial, the runner market type in MFK No. 2 trial and the entirety of the Mirebalais trial were planted at double the desired planting rate; subsequently, after the stand count evaluation, plots with too many plants were thinned and plots with too few plants were replanted with transplanted plants. Other yield-reducing factors were managed in order to mitigate confounding results from the experimental factors. As such, seeds were treated with azoxystrobin, fludioxonil and mefenoxam (Dynasty PD®, Syngenta Crop Protection, Greensboro, NC) at a rate of 85 g of product per 45.4 kg of seed. Plots were weeded on a biweekly basis, sprayed every 15 days (starting 30 days after planting) with fungicide (Muscle® ADV, Sipcam Agro USA, Inc.) and irrigated biweekly in the absence of rain. Data Collection: Two to three weeks after planting, stand counts were made for each plot; average stand count across all plots was 70.9% for MFK Trial #1, 96.9% for MFK Trial #2, and 100% for the Central Plateau trial. Final leaf spot and rust severity ratings were taken immediately prior to digging. Leaf spot severity was assessed with the Florida 1 to 10 scale (Appendix III). Rust severity was assessed with a modified 1 to 9 scale (Appendix III). Peanuts were manually harvested by first pulling the entire plant from the ground and removing all the attached pods from the plant. Afterwards, the soil in each plot was filtered through by hand to recover the remaining pods left in the ground. Pods were placed in large, green, mesh cabbage bags (Cady Bag Company, LLC. Pearson, GA) and washed after harvest in order to remove any remaining soil, and then placed on a large concrete pad to dry in the sun. The bagged pods were allowed to dry for a minimum of three days, and were moved under a shelter each night. After drying bags were weighed, and immediately afterwards, a 100-pod sample was shelled to obtain the moisture content of the kernels. Final weights (kg/ha) were adjusted to 10% pod moisture. A post-harvest test was made with the same 100-pod sample to evaluate the percentage of sound mature kernels (%SMK). Percent SMK was calculated by dividing the weight of the sound mature kernels by the total weight of the unshelled sample. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 46 Statistical Analysis: Yield was subjected to analysis of variance with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). Because preliminary analyses indicated a significant variety × planting method treatments, varieties were analyzed separately. For both market types, the model was a split plot design with trial and planting method treatment considered as fixed effects, with rep and rep × trial as random effects. In all analyses, the Kenward-Roger option was used to adjust the degrees of freedom, and differences in the least square means were tested by Tukey’s multiple comparisons test. Fig 4.2.a. 2017 Planting Method Trial No. 1 at MFK taken 18 April 2017. Fig 4.2.b. 2017 Planting Method Trial No. 1 at MFK taken 5 June 2017. Table 4.2.b. Analysis of variance performed separately for runner and Valencia market types. Market type Runner Valencia Effect Trial Treatment Trial × Treatment Trial Treatment Trial × Treatment Yield Prob > F 0.0331 0.0821 0.9942 0.0008 <.0001 0.3070 %SMK Prob > F 0.0004 0.7282 0.1838 <.0001 0.4134 0.9737 Table 4.2.c. Effect of trial (location) on pod yield and %SMK for runner and Valencia market types. Market type Runner Valencia Effect (MFK) Spring 2017 (MFK) Summer 2017 Central Plateau (MFK) Spring 2017 (MFK) Summer 2017 Yield 2899.3 4044.2 1454.5 2800.5 1403.8 b a b a b UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report %SMK 58.9 b 90.6 a 61.9 b 88.9 a 47 Fig 4.2.c. Effect of planting method on total pod yield (kg/ha) for each market type. Planting method treatments are labeled as follows: width between row or scatter divot, planting method type, number of seed planted within the row per foot or per divot. Therefore, 24_Row_1Sd signifies, a 24-inch row planted at 1 seed/ft. Treatment with the same letters are not significantly different based upon Tukey’s honestly significant difference test. Conclusions: These results suggest that the traditional scatter planting method used in Haiti can provide similar yields to those obtained when the same/similar number of seeds are planted in rows. In these trials, higher yield for both varieties was more a function of increased plant density than the type of planting method utilized. However, as seen in chapter 4.1, plant density tends to impact yield more in the local Haitian Valencia than in the local Haitian runner. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 48 Appendix I Monthly Rainfall Data from the MFK Research Site Located Near Cap-Haïtien, Haiti, from 2015 to 2017 x > 0.254 mm Rain eventsy > 0.63 mm > 12.7 mm 336 7 5 5 2 176 6 3 3 2015 3 90 9 4 2 2015 4 2 1 0 0 2015 5 114 13 7 3 2015 6 128 14 6 3 2015 7 121 17 8 2 2015 8 17 6 0 0 2015 9 5 2 0 0 2015 10 65 6 4 1 2015 11 72 4 3 2 2015 12 1 1 0 0 2016 1 41 7 2 2 2016 2 455 11 8 6 2016 3 23 6 1 0 2016 4 135 12 7 3 2016 5 188 12 5 3 2016 6 112 5 5 2 2016 7 0 0 0 0 2016 8 98.9 2 2 2 2016 9 106.8 7 5 3 2016 10 221 16 10 4 2016 11 828 22 14 12 2016 12 11 5 0 0 2017 1 50 5 2 2 2017 2 15 4 1 0 2017 3 228 11 7 4 2017 4 81 10 3 2 2017 5 225 9 8 6 2017 6 126 4 4 3 2017 7 138 7 4 2 2017 8 1 1 0 0 Year Monthx Rain (mm)z 2015 1 2015 Where 1 = January and 12 = December. Number of days that received > 0.25, > 0.63 or > 12.7 mm of rain during each interval day after planting interval. Rainfall data was measured with two on-station Decagon rain gauges set to record at hourly intervals. y z Rainfall data was recorded hourly at the MFK research site with a Decagon ECRN-50 rain gauge (Decagon Devices, Inc., Pullman, WA). UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 49 Appendix II Soil Samples from Research Plots in Haiti Supplementary Figure II.a. 2015 soil sample results from research plots at Meds & Food for Kids (MFK) factory located outside of Cap-Haïtien (Quartier Morin), Haiti. Mehlich 1 mg/kg (ppm) LBC 1 Sample (ppm CaCO3/ pH) pH Equivalent 2 CaCl2 water pH Ca K Mg Mn P Zn 1 MFK – student field (control) N.A. 7.78 8.38 4409 37.3 583.3 37.41 41.3 0.46 2 MFK – student field (biochar) N.A. 7.59 8.19 3838 41.7 600.1 47.58 70.6 0.76 3 MFK – back field (west side) 529 7.10 7.70 3720 51.4 689.4 62.60 34.3 1.51 4 MFK – back field (east side) N.A. 7.70 8.30 3962 48.0 585.3 55.62 46.8 0.98 5 MFK – east side of drive (near septic) N.A. 7.82 8.42 4728 31.0 598.2 21.73 4.4 <0.21 6 MFK – west side of drive (near gate) 498 7.14 7.74 3807 47.8 543.2 47.44 143.8 1.19 7 MFK – front field A (w trees) 482 7.01 7.61 4480 105.5 478.7 53.62 501.3 2.98 8 MFK – front field B (w bananas) 398 7.14 7.74 3760 78.5 423.0 54.86 325.3 2.33 9 MFK – banana field 423 7.30 7.90 3366 91.0 523.1 60.88 106.0 1.67 Samples were taken in December 2014 and completed 10 March 2015. Soil Samples were analyzed by the University of Georgia Soil, Plant, and Water Laboratory. Supplementary Figure II.b. 2016 soil sample results from research plots at Meds & Food for Kids (MFK) factory located outside of Cap-Haïtien (Quartier Morin), Haiti and from research plots at the Acceso research farm located in the Central Plateau (Coupe Gorge). Sample pH (ppm CaCO3/pH) CaCl22 LBC 1 Mehlich 1 mg/kg (ppm) Equivalent Ca K Mg water pH Mn P Zn B MFK student field at 4-inch depth N.A. 7.62 8.22 3604 48.3 814 56.88 58.4 0.62 1.36 MFK student field at 10-inch depth N.A. 7.56 8.16 3283 40.0 832 57.25 59.3 0.58 1.37 MFK back east field at 4-inch depth N.A. 7.69 8.29 5747 91.9 1567 95.99 93.9 1.01 1.53 MFK back east field at 10-inch depth N.A. 7.65 8.25 3485 47.9 1004 57.84 44.9 0.62 1.45 MFK back west field at 4-inch depth 476 7.36 7.96 3319 55.7 1094 58.76 47.4 0.81 1.46 MFK back west field at 10-inch depth 511 7.35 7.95 3066 49.8 1166 60.75 32.4 0.85 1.55 MFK front field at 4-inch depth 389 6.99 7.59 4119 106.5 555 62.11 670.9 3.74 1.26 MFK front field at 10-inch depth 422 7.03 7.63 4033 99.7 589 51.89 539.1 3.35 1.16 Central plateau sample 1 at 4-inch depth 522 7.17 7.77 4347 32.8 121 14.36 1.7 0.65 0.56 Central plateau sample 1 at 10-inch depth 561 7.05 7.65 3975 28.2 126 19.18 1.2 0.67 0.43 Central plateau sample 2 at 4-inch depth 578 7.17 7.77 4644 31.7 124 14.00 1.5 0.74 0.41 Central plateau sample 2 at 10-inch depth 587 7.20 7.80 4098 29.8 117 16.91 1.0 0.64 0.40 Central plateau sample 3 at 4-inch depth 547 7.19 7.79 4362 36.6 123 15.92 1.9 0.75 0.44 Central plateau sample 4 at 4-inch depth 633 7.31 7.91 4532 32.1 120 14.29 1.6 0.63 0.35 Samples were taken in August 2014 and completed 20 October 2016. Soil Samples were analyzed by the University of Georgia Soil, Plant, and Water Laboratory. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 50 Supplementary Figure II.b. (Continued) 2016 soil sample results from research plots at Meds & Food for Kids (MFK) factory located outside of Cap-Haïtien (Quartier Morin), Haiti and from research plots at the Acceso research farm located in the Central Plateau (Coupe Gorge). Sample OM 3 Sand Silt Clay Soil Type MFK student field at 4-inch depth 3.31 44.0 25.9 30.1 Sandy Clay Loam MFK student field at 10-inch depth 3.39 44.0 25.8 30.2 Sandy Clay Loam MFK back east field at 4-inch depth 3.52 37.9 27.8 34.2 Sandy Clay Loam MFK back east field at 10-inch depth 3.60 37.9 27.8 34.2 Sandy Clay Loam MFK back west field at 4-inch depth 4.65 29.8 33.9 36.2 Clay Loam MFK back west field at 10-inch depth 4.86 25.8 34.0 40.3 Clay MFK front field at 4-inch depth 3.88 61.8 20.0 18.3 Sandy Loam MFK front field at 10-inch depth 3.92 57.8 21.9 20.3 Sandy Clay Loam Central plateau sample 1 at 4-inch depth 5.87 24.1 26.0 49.9 Clay Central plateau sample 1 at 10-inch depth 5.94 24.1 24.0 51.9 Clay Central plateau sample 2 at 4-inch depth 5.90 26.1 26.0 47.9 Clay Central plateau sample 2 at 10-inch depth 5.99 22.1 24.0 53.9 Clay Central plateau sample 3 at 4-inch depth 5.94 26.1 26.0 47.9 Clay Central plateau sample 4 at 4-inch depth 5.96 24.0 26.0 50.0 Clay Samples were taken in August of 2014 and completed 20 October 2016. Soil Samples were analyzed by the University of Georgia Soil, Plant, and Water Laboratory. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 51 Appendix III Leaf Spot and Rust Rating Scales Used to Assess Foliar Disease Severity in Haiti Supplementary Figure III.a. Florida 1 to 10 scale used to assess leaf spot severity. Score Florida 1 - 10 Rating Scale (Chiteka et al., 1988) Description 1 No disease 0 2 Very few lesions (only on the bottom part of the canopy) 0 3 Numerous lesions on bottom and a few lesions on upper canopy 0 4 Severe lesions on bottom; intermediate middle; moderate top ~5% 5 Bottom defoliated; severe middle; intermediate top ~ 20 % 6 Bottom and middle defoliated; top heavy ~ 50 % 7 Bottom and middle heavily defoliated; top severe pressure ~ 75 % 8 90% defoliated ~ 90 % 9 99% defoliated: Very few leaves remaining and those covered with lesions ~ 98 % 10 Plants dead ~ 100 % % Defoliation Supplementary Figure III.b. Modified ICRISAT 1 to 9 scale used to assess rust severity. ICRISAT Peanut Rust 1 - 9 Scale (Subrahmanyam et al., 1995) Score Description % Severity 1 No disease 0 2 Lesions sparsely distributed largely at lower leaves (a few lesions on the bottom leaves only) 1-5% 3 Many lesions on lower leaves, necrosis evident; very few lesions on middle leaves; no lesions on top 6 - 10 % 4 Numerous lesions present on lower and middle leaves; severe necrosis on lower leaves. A few lesions on top leaves 11 - 20 % 5 Severe necrosis of lower and middle leaves; lesions on top leaves but not severe 21 - 30 % 6 Extensive damage to lower leaves. Lesions densely present on middle leaves with necrosis; lesions also on top leaves 31 - 40 % 7 Severe damage to lower and middle leaves; lesions densely distributed on top leaves 41 - 60 % 8 100% damage to lower and middle leaves; lesions on top leaves with severe necrosis 61 - 80 % 9 Almost all leaves withering; bare stems present 81 - 100 % References: Chiteka, Z., Gorbet, D., Shokes, F., Kucharek, T., & Knauft, D. (1988). Components of resistance to late leafspot in peanut. I. Levels and variability-implications for selection. Peanut Sci. 15:25-30. Subrahmanyam, P., McDonald, D., Waliyar, F., Reddy, L., Nigam, S., Gibbons, R., Rao, V. R., Singh, A., Pande, S., & Reddy, P. (1995). Screening methods and sources of resistance to rust and late leaf spot of groundnut. Information Bulletin No. 47. International Crops Research Institute for the Semi-Arid Tropics. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 52 Appendix IV 2010-2011 ICRISAT Trials Purpose. Evaluate the performance of foliar disease resistant breeding lines developed by the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) for suitability for use in Haiti. Experimental design. Three field trials were conducted at three different locations in northern Haiti between 2010 and 2011. In 2010, one trial was conducted in the community of Bois Rouge and another trial was conducted at the Université Chretienne du Nord d’Haiti (UCNH) in Limbé, Haiti. In 2011, an additional trial was conducted in the community of Isle Adam. All trials were laid out in a randomized complete block design with four replications. Treatments consisted of 15 advanced breeding lines from ICRISAT and at each location a known check was included. These were Tamnut OL06 (Bois Rouge), Tifguard (UCNH) and the local Haitian runner and the local Haitian Valencia (Isle Adam). Table IV.a. ICRISAT breeding lines evaluated in trials conducted in Haiti during 2010 and 2011. Identity Branching Habit Botanical Type Seed Color ICGV 99027 ICGV 99028 ICGV 99029 ICGV 99030 ICGV 99031 ICGV 99032 ICGV 99033 ICGV 99036 ICGV 99046 ICGV 99050 ICGV 99051 ICGV 99052 ICGV 99053 ICGV 99054 ICGV 99057 Sequential Sequential Sequential Sequential Sequential Sequential Sequential Alternate Sequential Alternate Alternate Alternate Sequential Sequential Sequential Spanish Spanish Spanish Spanish Spanish Spanish Spanish Virginia Spanish Virginia Virginia Virginia Spanish Spanish Spanish Tan Tan Tan Tan Tan Tan Tan Red Red Tan Tan Tan Tan Tan Red At each location, peanuts were planted in two rows in plots that were 1.2 m wide and 3.04 m long. Seeding rate was not recorded but is assumed to be three seed/0.3 m. Plots were planted on 16 June 2010 at Bois Rouge and 30 October 2010 at UCNH. Date of planting was not able to be determined for the trials at Isle Adam. Plots were weeded several times throughout the season, did not include insecticide or fungicide application and were not irrigated. Prior to digging, final leaf spot severity was assessed with the Florida 1 to 10 scale (Appendix III) and final rust severity was assessed with a modified 1 to 9 scale (Appendix III). UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 53 Results Table IV.b. Effect of variety at Bois Rouge, 2010. Trt. Variety Leaf spot Rust 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ICGV 99027 ICGV 99028 ICGV 99029 ICGV 99030 ICGV 99031 ICGV 99032 ICGV 99033 ICGV 99036 ICGV 99046 ICGV 99050 ICGV 99051 ICGV 99052 ICGV 99053 ICGV 99054 ICGV 99057 TN 2.6 2.6 2.6 2.5 2.5 2.5 2.2 2.5 2.6 2.0 2.2 2.6 3.1 2.8 2.1 6.7 Tukey’s HSD P=.05 Standard Deviation CV Replicate F Replicate Prob(F) Treatment F Treatment Prob(F) B B B B B B B B B B B B B B b a 1.5 0.6 20.4 0.1 0.0429 14.868 0.0001 1.8 2.0 1.8 2.3 1.8 1.7 1.7 1.6 1.5 1.7 1.6 1.8 1.8 2.4 2.0 8.1 Kg/ha b b b b b b b b b b b b b b b a 0.2 0.1 12.1 0.5 0.6316 22.413 0.0001 1274.4 1440.0 1296.4 1237.6 1352.8 1351.5 1295.2 1721.4 1671.8 1278.0 1502.2 1278.8 1052.1 1357.1 1446.7 245.5 ab ab ab ab ab ab ab a a ab ab ab b ab ab c 610.2 237.7 18.3 0.2 0.0446 7.652 0.0001 Table IV.c. Effect of variety on leaf spot, rust, and pod yield at UCNH, 2010. Trt. Variety Leaf spot Rust Kg/ha 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ICGV 99027 ICGV 99028 ICGV 99029 ICGV 99030 ICGV 99031 ICGV 99032 ICGV 99033 ICGV 99036 ICGV 99046 ICGV 99050 ICGV 99051 ICGV 99052 ICGV 99053 ICGV 99054 ICGV 99057 Tifguard Tukey’s HSD P=.05 Standard Deviation CV Replicate F Replicate Prob(F) Treatment F Treatment Prob(F) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 a a a a a a a a a a a a a a a a 0 0 0 0 1 0 1 3.7 3.5 3.2 3.5 3.2 3.5 3.7 2.0 3.5 3.2 2.7 3.0 3.7 4.2 3.2 6.0 0.3 0.12 5.93 1.25 0.3029 11.326 0.0001 bc bc bc bc bc bc bc d bc bc cd cd bc b bc a 1529.3 1227.8 1288.6 1573.7 1785.9 1320.1 1160.6 2172.9 2246.0 1754.4 1670.4 1299.4 945.3 884.3 1704.5 759.2 abc abc abc abc abc abc bc ab a abc abc abc c c abc c 1034.0 401.9 27.58 1.39 0.2587 4.533 0.0001 UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 54 Fig IV.a. Planting plots at UCNH in 2010. Table IV.c. Effect of variety on leaf spot and rust at Isle-Adam, Haiti, 2011. Trt. Variety 1 ICGV 99027 2 ICGV 99028 3 ICGV 99029 4 ICGV 99030 5 ICGV 99031 6 ICGV 99032 7 ICGV 99033 8 ICGV 99036 9 ICGV 99046 10 ICGV 99050 11 ICGV 99051 12 ICGV 99052 13 ICGV 99053 14 ICGV 99054 15 ICGV 99057 16 Local runner 17 Local Valencia Tukey’s HSD P=.05 Standard Deviation CV Replicate F Replicate Prob(F) Treatment F Treatment Prob(F) Leaf spot 3.5 3.7 3.7 3.0 4.0 3.7 3.7 4.0 4.7 2.7 3.5 3.0 4.5 3.7 4.0 7.0 7.0 bc bc bc c bc bc bc bc b c bc c b bc bc a a Rust 1.8 1.5 1.8 2.3 1.5 1.5 1.1 1.5 1.5 1.5 2.0 1.5 1.3 2.0 1.5 2.8 8.0 cde cde cde bc cde cde e cde cde cde bcd cde de bcd cde b a 0.13 0.03 4.62 0.82 0.20 9.86 0.312 0.5843 15.632 0.0001 0.045 0.8348 122.348 0.0001 Kg/ha Not available - UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 55 Appendix V Summer 2015 Seed- and Row-Spacing Trial Results Experimental Design: Everything was the same as those conducted in 2016 and 2017 (see Chapter 4.1) with the exception that the 2015 trials did not include the 12-inch (0.3 m) between-row spacing treatment and did not have a 2-foot (0.6 m) border between plots within the same block. Otherwise, plots were maintained in a similar fashion. The only known difference is that plots at Trou-Du-Nord were irrigated by hand and less frequently than at MFK. Lastly, there are no stand-count records, but we assume that the germination was similar across treatments. Statistical Analysis: Yield was subjected to analysis of variance for each trial with PROC GLIMMIX (SAS 9.4 Institute, Cary, NC). The model was a split-split plot design with variety, row spacing, and seed spacing considered as fixed effects, and with replication, replication × variety and replication × between-row spacing as random effects. The SLICE option in SAS was used to explore all two-way interactions. In all analyses, the Kenward-Roger option was used to adjust the degrees of freedom, and differences in the least square means were tested by Tukey’s multiple comparisons test. Table V.a. Analysis of variance results for pod yield for two trials conducted in Haiti in 2015. Location MFK Trou-Du-Nord Effect F-value P-value Variety (V) 42.44 0.0019 Between-row spacing (B) 1.59 0.2614 V×B 0.48 0.496 Within-row spacing (W) 24.87 <.0001 V×W 1.69 0.2043 B×W 0.73 0.4927 V×B×W 0.17 0.8464 Variety (V) 14.43 0.0115 Between-row spacing (B) 0.77 0.4392 V×B 5.97 0.0214 Within-row spacing (W) 5.48 0.01 V×W 0.75 0.481 B×W 1.31 0.2864 V×B×W 0.26 0.7767 UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 56 Table V.b. Effect of variety, between-row spacing and within-row spacing on pod yield at two locations in Haiti during 2015. Location Effect Variety MFK Yield (kg/ha) Local runner 2486 a Local Valencia 1242 b 18 1909 a 24 1618 a 1 Seed/ft 1276 b 3 Seed/ft 1931 a 6 Seed/ft 2203 a Local runner 1316 a Local Valencia 637 b 18 963 a 24 870 a 1 Seed/ft 784 b 3 Seed/ft 865 b 6 Seed/ft 1132 a Between-row spacing Within-row spacing Variety Trou-Du-Nord Treatment Between-row spacing Within-row spacing Table V.c. Simple effect of between-row spacing on pod yield for each variety at two locations in Haiti during 2015. Trial Variety Local runner MFK Local Valencia Local runner Trou-Du-Nord Local Valencia Between-row spacing Yield (kg/ha) 18 2762 a 24 2237 a 18 1319 a 24 1170 a 18 1235 a 24 1403 a 18 751 a 24 540 b Appendix VI Virus Symptoms on Peanut in Haiti Virus Rating Method: Virus intensity was assessed as the number of 0.3 m sections “hits” of peanut plants with symptomatic viral infection per foot of row. Primary Symptoms: Ringspots Dense clustering of stunted leaves at the terminal Chlorosis/mosaic Extreme stunting Death of terminal leaves UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 57 Local Haitian runner with extreme stunting and chlorotic leaves. Local Haitian Valencia leaves with ringspots and mosaic. Local Haitian Valencia leaflets with ringspots. Local Haitian Valencia with dense cluster of stunted leaves at the terminal. Local Haitian Valencia with death of terminal leaves. Confirmation of tospovirus in Haiti with Agdia immunoStrip® for tomato spotted wilt virus (TSWV). Test strips are known to have cross reactivity with tomato chlorotic spot virus (TCSV) which has previously been reported in Haiti (Adegbola et al., 2016). References: Adegbola, R., Fulmer, A., Williams, B., Brenneman, T., Kemerait, R., Sheard, W., Woodward, J., Adkins, S., & Naidu, R. (2016). First report of the natural occurrence of tomato chlorotic spot virus in peanuts in Haiti. Plant Dis. 100:8, 1797. UGA Cooperative Extension Bulletin 1499 • Haiti Peanut Research Report 58 This publication was made possible through support provided by the U.S. Agency for International Development (USAID) Office of Agricultural Research and Policy in the Bureau for Food Security under the terms of Award No. AID-ECG-A-00-07-0001. The opinions expressed herein are those of the author(s) and do not necessarily reflect the views of USAID. extension.uga.edu Bulletin 1499 October 2018 Published by the University of Georgia in cooperation with Fort Valley State University, the U.S. Department of Agriculture, and counties of the state. For more information, contact your local UGA Cooperative Extension office. The University of Georgia College of Agricultural and Environmental Sciences (working cooperatively with Fort Valley State University, the U.S. Department of Agriculture, and the counties of Georgia) offers its educational programs, assistance, and materials to all people without regard to race, color, religion, sex, national origin, disability, gender identity, sexual orientation or protected veteran status and is an Equal Opportunity, Affirmative Action organization.