United Nations Development Programme Ecologically Sustainable Cassava Plant Protection in South Ame rica and Africa: An Environmentally Sound Approach 1996 Annual report of Activities in South America Prepared by project personnelin Colombia and Brazil South American component of a global project involving: Centro Internacional de Agricultura tr~cal (CIA T) at Cali, Colombia; The lnternational Institute for Tropical Agriculture (liT A) in Nigeria, and the Empresa Brasileira de Pesquisa Agropecuária, Centro Nacional de Pesquisa de Mandioca e Frutas Tropic:ais (EMBRAP NCNPMF) at Cruz das Almas, Bahía, Brazil. March, 16, 1998 "Ecologically Sustainable Cassava Plant Protection in South America and Africa: An Environmentally Sound Approach" Technical Report on Activities conducted during 1996 by PROFISMA (Prot~io Fitossanitária Sustentavel da Mandioca): the South American component of a global UNDP project Project Ref. GL0/91/013 Table of Contents 1 EXECUTIVE SUMMARY ..••.............•...............•..•.....•.•....•.•.••.......................•.•....•.•................................. 1 2PROJECT COORDINATION .............•...............•...•...•..••••.••...•..•••••••••.•......•.......................................... 13 2 .1 COORDINATION AT CIAT, BY STEVEN LAPOINTE ...... . .. . ... .. ... .. . .. .. ... .. ....... ... . .... .. . .. ..... ..... .... ..... . .. . . ... ... ... 13 2.2COORDINATION AT EMBRAP NCNPMF, BY ARISTOTELES PIRES DE MATOS ...... ... .. .. .... .. .... .............. . ... 17 2 .3PHASE TWO PROPOSALS, BY LINCOLN SMITH ............. . ....... .. . . ...... ..... .... ........... . ...... . ...... .. ................ ...... 19 JTRAINING AND PARTICIPATORY RESEARCH ..•••..•....••..................................................•.............. 22 3 .1TRAIN!NG, BY BERNARDOÜSPINAPATIÑO .... ...... . .. .. . ........ .. .. .... ...... .. .... ....... .. .. .. .... . .. ............ .. .. . ...... .. .... 22 3 .2PARTICIPATORY EVALUATION OF TECHNOLOGY TESTING TRIALS ................. .. ... .. .. .. ..... .. .. .. .. .. ........ ........ 27 4STRATEGIC RESEARCH AT CIAT ..••..•....•............•..................................••......•............•...•.......•..••••.•• 52 4.1 BIOLOGICAL CONTROL OF CASSA V A GREEN MITE, BY LINCOLN SMITH .. .. .. .. ... .... ........... .. ............ .. ... ...... 52 4.2ENTOMOLOGY, BY ANTHONY C . B ELLOTII ... .. . .. ..... .. . .. .. . .. .. ..... ..... .. ...... .. ... .. . ... .... .. .... ... .. .. ...... ........ ...... 80 4 .3PLANT PATHOLOGY, BY ELIZABETH ALVAREZ AND JOHN B. LOKE ....... .... ........ .... ... . ................. .. .. .. ........ 84 4 .4VIROLOGY, BY LEE A. CALVERT .. . .. .. .. .. ...... ........... . ... .... ......... ........ .. ......... ...... .... .... .. .. .. ....................... 95 4 .5CROP MANAGEMENT, BY MABROUK EL-SHARKA WY, L UIS FERNANDO CADA VID, SARA MEJÍA DE T AFUR, AND MANUEL GERARDO CA YÓN .. .......... . . ........ . .. .. .... .. . ... .. .. .. . ... . .. ...... .. . .. .. .. . .. .. .. . ... . ........ ...... ... 100 SSTRATEGIC RESEARCH AT CNPMF ····-···············~····················································· .............. 108 5.1 BIOLOGICAL CONTROL OF TIIE CASSAVA GREEN Mm:, BY ALoYSEIA CRISTINA S. NORONHA ..... 108 5.2 BIOLOGICAL CONTROL OF TIIE CASSA VA MEAL YBUG {CM), BY JosÉ MAURICIO SIMóES BENTO ... ... ................... .............................................................................. ............ ...... 118 5.3 CASSAVA CROP MANAGEMENr, BY JOSÉ EDUARDO BORO ES DE CARV AUIO ...... ................ 121 5.4 PLANT PATIIOLOGY, BY CmoERU FvKUDA .............................................. ........................... 127 5.5 CASSA VA INTENSIVE DIAGNOSTIC SURVEY IN NORTHEAST BRAZIL, BY Jos E HUMBERTO ALMEIDA DE CERQUEIRA AND CARLOS EsTE V ÁO LEITE CARDO SO ........................... 138 6PARTICIPANTS AND INSTITUTIONS •••••.•.•..•.•••••.•••••••••••••••••••••.•••••••••••••••••••••.•...•..••....•••••••••••• 146 7 ACRONYMS .................................................................................................................................... 148 S PUBLICA TIONS ............................................................................................................................. 149 PROFISMA (Prote~io Fitossanitária Sustentivel da Mandioca): the South American component of a global UNDP project, "Ecologically Sustainable Cassava Plant Protection in South America and Africa: An EnvironmentaUy Sound Approach". T echnical Report on Activities conducted during 1996. 1 EXECUTIVE SUMMARY PROFISMA completed its fourth project year at the close of 1996, thus terminating GL0/91/013 . The project has made notable progress in basic and applied research in biological control, integrated pest management and integrated crop management in cassava, emphasizing participation of farmers in Northeast Brazil in the identification of high priority problems and the development and evaluation of technological solutions. Support of activities in Africa Three predatory phytoseiid mites (Neoseiulus idaeus, Typhlodromalus aripo and T. manihoti) imported from Brazil, via quarantine in Arnsterdam have become established in Africa. Two ofthese appear to be reducing cassava green mite (CGM) populations by 30 to 90% in the lowland humid tropics. T. aripo is dispersing rapidly, and has moved up to 200 km from an initial release site in southem Benin. Estimated benefits for 4 countries in West Africa are on the order ofUS $60 million per crop cycle. These predators do not appear to be effective in dry or subtropical regions, such as the East African plateau. This region has been targeted by liTA for finding new natural enemies in Latin America adapted to this climate. CIAT identified regions with climates homologous to target zones in Zambia in central Brazil and southem Mexico using GIS (geographic information system) techniques. A graduate student from Zambia has been working with a - Brazilian scientist to explore the Brazilian region to find predatory mites that would be suitable to release in the East African plateau. Nine Strains of the predatory mi te Typhlodromalus manihoti from several high altitud e sites in Colombia were sent to liTA, Benin vía quarantine at the University of Amsterdam (Mitox) for eventual release in Africa. 1 2 liT A is eager to release the fungal pathogen Neozygites c. f jloridana in Africa to help provide additional biological control of cassava green mite, especially in dry regions. This pathogen naturally causes epizootics in Northeast Brazil, that periodically kili virtually all of the cassava green mites in a field in one or two weeks. Characterization of the biology and taxonomy of Neozygites c.f floridana strains using molecular genetic techniques have continued. The strain isolated fro_m cassava green mite is highly specific to this genus, indicating that it would be safe to release in Africa. Preliminary experiments with RAPDs indicated that this would probably not be an efficient technology to use to identify Neozygites strains. Experiments with AFLP (amplified fragment length polymorphism) are showing more promising results. PROFISMA' s methodology for farmer participatory research was presented at liT A with the aim of facilitating the transfer and adaptation of these methods and experiences to the African countries with which ESCaPP is working. Although ESCaPP has been implementing its own strategy for training and farmer research (based on farmer field schools ), they became very interested in organizing a training event for Afiican scientists based on PROFISMA' s participatory research strategy. A proposal was jointly developed and submitted by liT A to IDRC-Canada seeking financial support for its implementation. Basic researcb at CIA T Acarology. In addition to the research mentioned above, the CIAT acarology program collected phytoseiid predatory mites in high altitude regions in Colombia, for possible release in high altitude zones in Afiica. Predatory mites (Typh/odroma/us tenuiscutus)from a dry region in coastal Ecuador, with climate similar to the target release zone in Northeast Brazil, were sent to CNPMF via quarantine at CNPMNEMBRAP A. The suitability of different prey and natural food sources for different phytoseiids was evaluated to leam more about the specificity of candidate species. The exploration database was analyzed to leam more about the geographic and climatic distribution of phytoseiid species to help select those most likely to succeed in various target release zones. The database was also analyzed to learn about the degree of host plant specificity of candidate phytoseiid mites. 3 A computer simulation model was developed to help plan mass-rearing of predatory phytoseüd m.ites. Model predictions were compared to experimental results to help "validate" the model. A taxonom.ic key to 30 phytoseüd species commonly found on cassava in northern South America was developed in collaboration with a Brazilian taxonom.ist. The microscope slide collection of tetranychid mites was reviewed to determine the geographic and mite host species range of Neozygites spp. In vitro culturing techniques were improved thanks to collaborative work by Brazilian and U. S. scientists working at the Boyce Thompson lnstitute (Ithaca, New York). Entomology. Field experiments were conducted in the llanos of Colombia to evaluate the efficiency and possibie interactions between three parasitoids (Aenasius vexans, Apoanagyrus diversicomis and Acerophagus coccois ) of the cassava mealybug (Phenacoccus he"eni). Mealybug populations were low, and parasitism rates of 15.5% and encapsulation of37.6% were observed. The parasitoid A. vexans had the highest percent parasitism, and total parasitism rates were highest in the field where all three parasitoid species were released. Experiments with a Y -tu be olfactometer indica te that the parasitoids A. vexans and A. diversicomis are attracted to volatiles from mealybug-damaged cassava leaves, suggesting that the plant produces synomones ( odors that attract parasitoids to help protect the plant from mealybugs). A survey of 100 field sites in Colombia helped determine the geographic and climatic distribution of 3 common species of whiteflies attacking cassava (Aieurotracheius socialis, Bemisia tuherculata and Trialeurodes variabilis). Ten species ofparasitoids were recovered from the three whitefly species, suggesting that sorne of these may be potential biological control agents. Seven of these were recorded for the first time in Colombia, and four of these are new undescribed species. Plant pathology. Farms at 34 sites in Colombia were visited, revealing that root and stem rots caused by Phytophthora spp. are a major problem, causing root losses of up to 80%. Cassava in fields at CIA T is severely affected by root rots despite chemical treatment of stem cuttings (planting material), planting on ridges, crop rotation, and the relative long dry seasons. Average root yield reduction caused by root rots was calculated to be 7. 1 T /ha per 4 growing cycle for the period 1981-1995 by analysis of 60 field experiments. From the survey, 75 Phytophthora isolates were isolated, and these are currently being analyzed for biological and genetic differences. New in vitro and greenhouse methods of inoculation were developed to test varietal resistance, and 22 cultivars were inoculated with 4 Phytophthora spp. isolates. We developed a PCR method to characterize cassava root rot pathogens, which has been adopted by CNPMF IEMBRAP A _scientists in Brazil. In vitro methods were developed to culture of Phytophthora species isolated from cassava, free ofbacterial contamination. We developed bioassays to study pathogen variability and to test cassava germplasm for resistance. Trichodenna spp., isolated from the rhizosphere of cassava plants were screened in vitro for antifungal activity against a range of fungal pathogens: Phytophthora drechs/eri, P. nicotianae, Dip/odia manihotis and Fusarium orysporum. Filtrates from líquid cultures of Trichodenna isolate 11 TSM-4 caused marked reduction in growth ofthe pathogen D. manihotis, and filtrates of 14PDA-4 were most effective in reducing the growth ofF. orysporum, suggesting promise for biological control. An effective thermotherapy method was developed to eliminate Phytophthora infections in cassava stem cuttings (planting material). This would permit stopping a major source of infection of newly-planted cassava fields. Fusarium so/ani and F oxysporum strains were obtained from 86 cassava samples obtained from different ecological zones in Colombia. Four methods of classification were tested for differentiation ofthe strains, including morphology, growth, pathogenicity and site- directed PCR (rDNA). Screening of resistance has been conducted on 51 cassava cultivars by inoculating them with 28 Fusarium strains. In a survey of cassava growing areas in Colombia, 55 strains of Dip/odia manihotis were isolated, 8 of which were highly pathogenic to cassava plants. Three inoculation methods were used for pathogenicity studies, including stem injection, stem wounding and root cylinder inoculation. In a search for genetic resistance, 20 cassava cultivars have been inoculated with 40 strains of D. manihotis. Virology. An epidernic of geminivirus is causing extreme\ y severe losses ( 50-100%) of cassava production in U ganda. W e discovered that the geminivirus causing the East African epidernic is distinct from either African cassava mosaic virus (ACMV) or East Africa cassava 5 mosaic virus (EACMV), and it may actually be a hybrid of these viruses. The discovery of the true cause ofthe epidemic (i.e., not whiteflies) has made it possible to begin solving the problem by focusing on the evaluation and distribution of cassava varieties resistant to this newvrrus. In surveys conducted in Northeast Brazil, the disease caused by cassava vein mosaic virus (CVMV) was found to be very pr_evalent, and often the majority ofplants in a field are infected with the virus. While there are indications that yield losses of 10-15% occur, further studies are needed to quantify losses due to CVMV. Such studies were not previously possible because of the difficulty of detection of the presence of the virus in symptom-less plants. A rapid method to detect the presence of CVMV was developed at CIA T using polymerase chain reaction (PCR) methods. This test is capable of detecting all isolates tested to date, even when visible symptoms are not present. This method has been transferred to laboratories in Brazil. Isolates of CVMV from six states in Brazil were characterized, and they differ significantly at the molecular level; however, this needs to be correlated with biological characteristics. Cassava frogskin virus has been expanding its geographic range in Latín America. It is now endemic in parts ofthe state ofBahia, Brazil. This probably occurred because ofthe major drought in 1992, that caused farmers to import cassava stem cuttings from neighboring states that border the Amazon region. Cassava frogskin disease (CFD) can cause the complete failure of a crop and is an important new pathogen that needs to be controlled. A microscopic analysis of the roots of CFD-affected cassava revealed very different structural features that are being evaluated as method to select CFD-tolerant cassava varieties, and the CIAT core germplasm collection is currently being screened. Whitefly transmitted geminiviruses are major problems in Africa and India. With the "B" biotype of Bemisia tabaci colonizing cassava in the Americas, increased efforts are being made to characterize the threat that geminiviruses pose to cassava. Agronomy. Long-term response of cassava to NPK (nitrogen, phosphorous, potassium) levels in acid infertile soils were measured at Santander de Quilichao, Colombia. After 13 consecutive cropping cycles without application ofNPK fertilizer, dry root yield was oniy 6 about 4 t/ha. With application of 100 kglha each ófNPK, root yields were dramatically increased (about 10 tJha dry root). The largest observed response was for K, indicating that K is the most limiting nutrient in these soils, which have high organic matter content (ca. 7% ). To sustain reasonable productivity, K fertilizer must be applied annually since most K uptake (> 60%) is removed in the harvested roots. Long-term effects of surface mulch (grass), fertilizer and tillage on cassava productivity in poor sandy soil were measured at Pivijay, in northem Colombia. Dry root yield was significantly increased by the application of either mulch or NPK fertilizer at a moderate levels. Conventional tillage increased yield only in the presence ofNPK fertilization without mulch. The combination of surface mulch and no tillage gave the highest root yield with and without fertílizer. These findings indicate that consecutive cultivation of cassava in these poor soils would lead to very low yields unless appropriate measures are followed to maintain soil fertility. lnformation management. We scanned and formatted 11 major documents (2,800 pages) including recent CIAT Cassava Program annual reports, CIA T Working Documents, scientific meeting proceedings and Cassava Newsletters to be included in a CD-ROM being published by liT A and the University of Florida, Gainesville. Basic and applied research in Brazil Acarology. Five shipments ofthe phytoseiid predatory mite Typhlodromalus tenuiscutus, originally sent from CIA T, were made from the EMBRAP A/CNPMA quarantine facility to EMBRAP A/CNPMF (2, 150 individuals ). A colony was established and mass-rearing. Mass rearing of Neoseiulus californicus was carried out in a screenhouse on Tetranychus urticae on jack bean plants, enabling the production of33,500 individuals in a one month period. 10,800 T. tenuiscutus were released at 2 sites in Babia, and 34,020 N. californicus were released at 4 sites in Babia and Pernambuco. 28 specimens of N. californicus were recovered during monthly monitoring of the release si tes, suggesting that it may be establishing. In experiments using jack bean leaf discs that had previously been immersed in neem (Azadirachta indica) extract, we found that neem extract caused a decrease in survivorship of the two-spotted spider mite (Tetranychus urticae) but not its phytoseiid predator N ca/ifomicus. Studies on neem's effect on cassava green mite are under way. 7 CNPMF has 21 isolates of the mite pathogen Neozygites cf jloridana from several regions, in Brazil; however nine ofthem have lost their viability. Sorne isolates have been stored for up to 20 months as "murnmies" and still remain viable, whereas others have lost viability in as little as 6 months. V ariation in viability appears to be partly caused by the quality of the mummy and the amount of time that it spent in the field before being collected. Evaluation of different conditions for storing murnmies indicates that freezing ( -1 0°C) is generally better than 4°C or 25°C. Isolates from Cruz das Almas and Piritiba do not sporulate at relative humidity lower than 95% (at 24°C). Neozygites isolates from T. urticae, collected in Cali, Colombia, and in Jaguariuna, State of Sao Paulo, Brazil, were pathogenic to both T. urticae and M tanajoa; on the other hand, the isolates from M tanajoa, collected in Brazil, were pathogenic only to M tanajoa, thus showing a very high specificity. The efficiency offield inoculation of Neozygites sp. to control CGM was studied by making ten inoculations (May 22 to June 24) using 40 mummified mites as source of inoculum. Although the first appearance of infected mi tes control and experimental fields occurred almost simultaneously, by August 2, the pathogen had spread all over the release field, whereas it was not found in the control field. The delay in initiation of an epizootic have may be due to the low mite population at the time (25 individualslleaf), low release rate orto climatic conditions. Entomology. A total of 111,610 parasitoids (Aenasius vexans, Apoanagyrus diversicomis and Acerophagus coccois) were produced, and 35,950 ofthem were released in six cassava fields in Babia and Pernambuco. A. diversicornis dispersed the most, reaching up to 304 km by April 1996. A. coccois dispersed more slowly, reaching 180 km from the releasing si~e by September/1995, while A. vexans has not dispersed from its release site as of November, 1996, despite the large number of releases. A negative correlation between number of parasitoids and mealybugs has been observed for both A. diversicornis and A. coccois in the State ofBahia, although A. coccois has a stronger effect on mealybug populations. Although mealybug populations fluctuate seasonally, the peaks are now about 20% the size of those observed befare establishment of the parasitoids. 8 Plant pathology. Cassava witches' broom disease (CWBD), caused by a phytoplasm, is a very serious cassava problem in the "Serrada Ibiapaba", State ofCeará. An important achievement on the research of CWBD was the observation that elimination of infected planting material of susceptible varieties increased root yield. On farm trials to evaluate resistant germplasm were established in 60 rural communities, resulted in the selection of 2 improved clones that were preferred by 60% ofthe farmers to the local variety. Projected impact of adoption of these varieties, assuming adoption by 60% of the growers, implies replacement of about 3,000 ha of cassava and would represent 60,000 t increase in yield or about US$2.5 million increase in profits. A survey for the cassava root rot incidence was carried in three cassava growing regions ofthe State ofBahia. lsolations carried out under laboratory conditions showed that the pathogens involved in the cassava root rot development in the rural communities evaluated in the Bahía were Fusarium sp., Phytophthora sp. and Scyta/idium lignicola. A total of 179 varieties from the EMBRAPA/CNPMF Cassava Breeding Program were evaluated in on-farm experiments in Sergipe, Paraiba and Alagoas, permitting the selection of several varieties in each region that showed resistance to root rot. A farmer participatory trial showed that disease incidence is lower and root yield is higher when cassava is planted in ridges rather than on level ground. Field evaluation of 30 cassava hybrids, from the EMBRAP A/CNPMF Cassava Breeding Program, in the municipality of Umbauba, S tate of Sergipe, enabled the selection of five of them that were evaluated again in a on farm trial in a randomized block design, with three replicates. Eighteen cassava varietieslhybrids were evaluated for root rot incidence in a on farm trial carried out at the COP AL Quitéria, Municipality of Alagoa Grande, S tate of Paraiba, in a root rot naturally infested area. One hundred and seventy nine hybrids, from the EMBRAP A/CNPMF Cassava Breeding Program, were evaluated in on farm trials in the municipality ofVi~osa, State of Alagoas, where root rot incidence was relatively high. Due to erratic geographic and temporal incidence of cassava root rot, CNPMF decided to develop laboratory evaluation techniques to help complement field trials to identify cassava genotypes resistant to root rot pathogens. Several inoculation techniques were tested under laboratory and greenhouse conditions which generally showed that isoJates of Fusarium was not very virulent, but that isolates of P. drechs/eri and Scytalidium lignico/a expressed high virulence. Simultaneous inoculations of the antagonistic fungi Trichoderma polysporum, T. harzianum or T. pseudokoningii ~th the pathogen Phytophthora drechsleri were able to prevent it from producing normal sized lesions in sorne varieties of cassava, suggesting sorne possibilities for biological control of root rots. 9 Virology. The main objective in 1996 was to determine root yield losses dueto the cassava vein mosaic virus (CVMV). Such an experiment requires a large amount ofvirus-free planting material, but the PCR diagnostic method was inadequate for testing this many plants. Consequently, the production of CVMV -specific antiserum became the main objective since this is easier to perform. Several modifications on the CVMV purification technique, such as viral separation through polyacrylamide gel electrophoresis; however, all have failed to eliminate the presence of host protein in the purified virus. We evaluated 20 host plants as possible alternate hosts ofCVMV, but the virus multiplied onJy in cassava. Agronomy. Experiments on duration of manual weed control at 2 sites in Bahía indicate that it has a significant effect on yield onJy during the first 30 to 60 d after planting. Other experiments combining various forms of weed control and intercropping showed that weeding within double rows of cassava and planting pigeon pea, cowpea or jack bean between the double rows produced higher yields than leaving weeds between the double rows. Although weeding between the double rows tended to increase yield (though not significantly), this was at a 30% higher cost due to labor. A field trial on the effect of intercropping with jack bean or cowpea between cassava double rows to reduce erosion showed no difference because of a drought in which no soil loss by surface run-off was recorded. On farm trials in Acarau showed that adding bagana (palm leaf waste) mulch at 180 m3 /ha, or more, increased cassava root yield (25.1 t/ha) in comparison to fields without bagana (8.1 tfha). Another trial in Russas showed no effect, possibly dueto rainfall that occurred during the period of initial growth of the crop. 10 Farmer participatory research. The final training course "Participatory Methods for Evaluation ofTechnology Trials with Farmers" held at CNPMF/EMBRAPA in April was attended by 30 researchers and extension workers, representing 6 collaborating research and technology transfer institutions in_the states ofBahia, Ceará, Pernambuco and Paraiba, Brazil. The course improved their skills, abilities and attitudes for using two participatory methods in the evaluation of technology testing trials with farmers. The first series of farmer participatory research experiments at 18 COP ALs in 4 states of Northeast Brazil were completed and evaluated and the second series was initiated. The various experiments included evaluation of regional varieties for yield and resistance to root rot, cassava green mite and whitefly, planting on ridges to control root rot, use of chemical fertilizers, organic compost or legume companion crops, to improve soil fertility, and row spacing. A few experiments failed because of drought or because of inadequate follow-up by the technicians. However, most experiments produced results that increased yields by 1 O to 67%. In several cases the EMBRAP A scientists leamed of local varieties that performed better than their "improved" varieties, thus providing them with new germplasm to incorporate in their breeding programs. They also began to leam more about the traits other than yield that farmers value, which will contri bu te significantly to the relevance of future crop breeding efforts. Formation ofa second generation ofCOPALs (farmer research comrnittees) was initiated during 1996 in the S tate of Ceará through a project presented by local institutions to the Banco do Nordeste do Brasil (BNB). This project financed the establishment of COP ALs in six communities of cassava growers that had been identified during the participatory survey carried out by PROFISMA in 1994. Financia] support received from BNB included seed money to start the first COP AL FPR experiments, funds for farmer training events and for fue! and per diems for technicians assisting the COP AL. The COP ALs were autonomously organized by local institutions followed the same methodology taught by PROFISMA. Field days were held in Ceará and Pernambuco in 1996 with the aim of stimulating more support for COP AL activities at local and state level. The COP AL methodology for 11 enhancing fanner participation in research and technology transfer activities and the initial FPR results obtained by PROFISMA were presented to 160 fanners, technicians and decision- makers. PROSERTÁO is a rural development project executed by the Government ofthe State of Sergipe, Northeast Brazil, with financial support from IF AD through a 25 million dollars loan during a seven years period. PRQSERT ÁO staff approached CNPMF looking for technical assistance to sol ve the problem of root rots which plagues 95% of the municipalities in the project area. Ajoint proposal was developed in which PROSERTÁO funds activities by CNPMF that include: 1) the implementation of a training program in FPR & COP AL methodology for PRO SER T ÁO personnel, 2) establishment of COP ALs in sorne selected cassava growing areas, and 3) execution of fanner participatory activities to develop and transfer technologies to control cassava root rot. Socioeconomics. Intensive Diagnostic Survey was conducted in 18 cornmunities, involving 21 cassava growers in each cornmunity (9 COP AL members, 6 non-COP AL members, and 6 cassava growers from the neighborhood) in Northeast Brazil. Data were collected on land possession and usage, family composition, educationallevel, cassava production technology, and cornmercialization of cassava. Data were also collected on incidence of pests and diseases and the methods being u sed to control them. Over 50% of cassava growers had less than 5 ha. surveyed do not own the land they cultivate. About 29.9% ofthe land used as pasture, 29.9% is left under fallow, and 13.2% is natural vegetation, all of them general! y used for grazing. About 89% ofthe daughters, 81% ofthe sons, and 57% ofthe parents do not have any school education. The head ofthe family is predominantly aman older than 18 years; 52.5% of them are 50 years old or less. Approximately one third of the family members does not work in the rural cornmunity: they are either school age children or young people whose activities are other than rural labor. About 29%, besides working on their land, also sell their labor in order to make extra money. Most ofthe growers (64%) prepare the soil for planting only by manual hoeing. About 73% ofthe cassava growers use their own cassava stakes to start a new planting, and only 42% of them reported carrying out any kind of selection of stake quality prior to harvesting. Only 64% said that the stake must be 1 O to 20 cm long have a diameter greater than 2 cm. Farmers considered leafcutting ants, root rots, homworm, cassava green mite, and whiteflies to be the most important pest and disease problems. Although witches' broom disease is fairly uncommon, it can cause extremely high yield losses (72%). Farmers used pesticides only to controlleafcutting ants, homworms and anthracnose, and lime or ash to help control root rots. The majority of the cassava production in the Northeast (77%) is sold as cassava tlour, .. 19% as fresh roots, l . 6% as animal feed, and only O. 7% as dry chips. About a third of the aerial plant parts are used for feeding animals. Project management Due to anticipated reduction of funding in 1997, the CIAT coordinator in Brazil decided to eliminate bis own position, as well as those of sorne junior statf in Brazil and at CIAT. Project coordination by CIAT during 1997 was devolved to other staffin Cali, Colombia. 12 A concept note entitled, "Ecologically Sustainable Cassava Production Systems for Latin America" was prepared by CIAT and sent to 25 donors in May with the goal offinding additional donors to support a subsequent project based on the successes of this project. A second concept note entitled, "Ecologically Sustainable Cassava Plant Protection: A Global Strategy" was written jointly by CIAT and liT A in July to unifY the Latín American and African proposal concepts. The training component ofPROFISMA Phase 2 was prepared by CNPMF in collaboration with IPRA (CIAT program on farmer participatory research) personnel and was sent to the Chairman of the EMBRAP A Intemational Cooperation Department. 2 PROJECT COORDINATION 2.1 Coordination at CIA T, by Steven Lapointe The uncertainty of continued funding of this very successful project caused major disruption during 1996 and resulted in considerable effort directed to a search for continued support. The one-year extension provided by UNDP for 1997, while considerable and generous, given the current financial status ofthat agency, was insufficient to maintain all activities. After careful weighing of the altematives, the CIAT coordinator in Brazil decided to eliminate his own position, as well as sorne junior staffpositions in Brazil and at CIAT. This resulted in enough savings to allow retention of most project junior staff and the continuation of essential activities through 1997. 13 Project coordination by CIA T during 1997 was devolved to staff at Cali, Colombia. The Training Officer in Cruz das Almas assumed even greater responsibilities for coordination and execution of activities in Brazil. The refractory nature of the Brazilian national program staff at CNPMF to such large projects requiring el ose collaboration and sharing of responsibilities and executive authority with another organization (i.e., CIAT), was a major impediment that was only partially overcome through daily contact and the project's ability to work directly with state agencies and farmer communities. The placement of a CIA T scientist at CNPMF should be carefully considered in the future and should be contingent on careful delineation of responsibilitíes and authority for execution of such 1arge projects. CIAT and CNPMF should review recent experiences in an effort to improve institutional collaboration. Extemal Advisory Committee (EAC) Review The second and final extemal review of the African activities (ESCaPP project; Ecologically Sustaínable Cassava Plant Protection) took place during June, 1996 in Benin and Nigeria. The three Extemal Advisory Committee (EAC) members that reviewed PROFISMA in Cali in August 1995 (John Borden, Don Roberts, Asuncion Raymundo) were joined by Dr. Mohamed Dahniya. Bernardo Ospina (farmer participatory research trainer) and Steve Lapointe (Project Coordinator) represented the Latín American activities (PROFISMA; Portuguese acronym for Ecologically Sustainable Cassava Plant Protection). The EAC team 14 was split into 2 groups, one visited Nigeria (liTA headquarters and ESCaPP/Nigeria field sites) and the other toured ESCaPP/Benin field sites during the first 4 days ofthe review. The groups reunited in Cotonou for 2 days of presentations. ESCaPP presented impressive achievements. A large number of national program scientists and extensionists have been trained in production methods, and these, using local extensionists as multipliers, have ~stablished on-farm trials with grower participation. We visited severa! of these in Benin. The National Coordinator for ESCaPP in Benin, Norbert Maroya, was particularly impressive - organized, articulate, in touch with field activities and obviously motivated and comrnitted to the project. Overall, we were impressed with the enthusiasm and commitment of national program staff to the project - a marked contrast with PROFISMA, perhaps due to the fact that liT A maintains control to a greater extent than PROFISMA over finances and activities. By far the most impressive accomplishment of the project to date is the establishment, spread and measured impact of Typhlodromalus aripo (a phytoseiid mi te predator of cassava green rnite [CGM]). Field trials have demonstrated reductions in CGM numbers ofup to 90%, and econornic impact regarding increased root yields of 30%, corresponding to profits of about US$ 60/ha per crop cycle. Area-wide benefit will undoubtedly be a significant (in the $10-100 rnillions per year), and benefit will continue year-after-year because ofthe self- sustaining nature of classical biological control. This will no doubt represent an extremely high retum on the initial donor investment, causing widespread impact on food production and econornic development in the lowland humid tropics of sub-Saharan Africa, with no adverse environmental impact. Major accomplishments of PROFISMA Below are the major accomplishrnents of PROFISMA as the first phase of the UNDP project draws to a close: l. Biological Control l. Successful introduction and establishment of 3 species of parasitoid natural enernies for control ofthe cassava mealybug Phenacoccus herreni in Northeast Brazil. Two species 15 (Acerophagus coccois and Apoanagyrus [ =Epidinocarsis] diversicomis) are established and spreading rapidly in Bahia and Pernambuco causing reductions of mealybug populations. 2. Establishment of a phytoseiid predator mite (Neoseiulus ca/ifomicus) from northem S. America in Northeast Brazil for control ofthe cassava green mite (CGM). Establishment is tentative and spread is not yet documented. Protocols for exploration, collection, shipping, quarantine, rearing and release were developed for rapid, multiple introductions of additional species of natural enemies in 1996 and 1997. 3. Developed in vivo productíon methods for and methods for genetic characterization of the fungal pathogen (Neozygites c.fjloridana) for control ofCGM. Methods for isolating and growing Neozygites in artificial media are being developed to facilitate further study and exploitation of this fungus as an agent for biological control of CGM in Latín America and Africa. Genetic characterization allows taxonomic differentiation between fungus isolates to help select the best strains and to identify them for quarantine purposes and to monitor it after being released in Afiica and elsewhere. 4. Contributed to biological control of CGM in Africa in collaboration with ESCaPP through climate homologue mapping, and collection and characterization (biological and genetic) of pest and predator mites. Regions to explore in Minas Gerais, Brazil and southem Mexico were identified as ecological homologs for Zambian target release zone. D. Farmer Participatory Research in NE Brazil l . Completion of extensive and intensive participatory diagnostic surveys of cassava growers in NE Brazil. Major biotic and abiotic constraints to production in small-scale cassava production systems were identified through participatory surveys that emphasized farmers' perceptions and priorities. Approximately 1,600 farmers were contacted. 2. Trained a core of state extension agents and national program scientists in Farmer Participatory Research (FPR) methods. 3. Established 30 Local Agricultural Research Committees (COPALs) in 4 states of Northeast Brazil. 4. These COP ALs, with support from extensionists and national program researchers, will have planted and evaluated two cycles of experiments in their communities addressing pest and disease constraints identified by farmers. m. Areas of global coUaboration between llTA and CIA T l . Exploration and introduction of natural enemies {phytoseiid mites and Neozygites) from Brazil and northem South America into Africa for control of CGM. 16 2. Mapping of climate homologies between regions of South America and Africa for targeting natural enemy explorations. 3. Biological and genetic characterization of natural enemies (phytoseiid mites and Neozygites) . 4. Scientific exchange visits. 5. Training of African and South American technicians in mite taxonomy. 6. Sharing and compilation of information resources~ production of CD-ROM including databases, bibliographies, and "gray literature" technical documents. 7. Global project coordination and exchange. 17 2.2 Coordination at EMBRAPA/CNPMF, by Aristoteles Pires de Matos PROFISMA computer database A computer database called Participatory Research System (SPP AR), is currently in the final elaboration phase. Data from one rural community has already been entered to test it. The database runs in FOX-PROT for Windows versions 3.1 and 95, and it has image storage capability. Workshop on the cassava root rot disease On April 8 to 11, 1996, a workshop on cassava root rot diseases was held at EMBRAPA/CNPMF. Plant pathologists from Research Institutions and Universities of Northeast and West Central Brazil, EMBRAPA/CNPMF, and CIAT discussed research approaches and problems related to this group of diseases. As a result of the workshop, the following suggestions were elaborated: 1) Carry out pathogenicity tests in order to know what pathogen occurs in each cassava growing area; 2) Identify and characterize cassava root rot-causing pathogens; 3) Study root rot pathogen interaction regarding disease development; 4) Establish methods to evaluate cassava genotypes for resistance to root rot pathogens; 5) Evaluate accessions ofthe Cassava Active Germplasm Bank (CAGB) of EMBRAP A/CNPMF for resistan ce to root rot pathogens; 6) Continue running field evaluations of cassava genotypes for resistance to root rot; 7) Continue studies on biological control and effect of cultural control practices on root rot development. Weed management consultant During 1996 Dr. Robinson Antonio Pitelli, Professor ofWeed Management, Universidade Estadual Paulista, Jaboticabal, S tate of Sao Paulo, was a consultant for the activities related to: i) effects ofweeds on cassava yield, ii) effects of cover crops, including weeds, on cassava yield and arthropod populations. After one week evaluating our activities, Dr. Pitelli found 18 them well-planned and well-conducted. He stressed the interdisciplinary characteristics of our activities since a Weed Scientist, Entomologists, Plant Pathologists, Agronomists, Soil Scientists anda Plant Taxonomist are working together. His main suggestions for future actions were: i) survey of pathogens on weeds~ ü) take also into consideration, besides the relative importance, the index of importance value, when discussing recorded data~ iü) develop basic studies on nutrient uptake by weeds, weed response to mineral and organic fertilizers, and weed response to drought and nutritional stress. Farmer participatory research (FPR) consultant The last component of the FPR training program, "Participatory Methods for Evaluation ofTechnology Testing Activities with Farmers", was taught during April, 1996, by Mrs. Teresa Gracia and Mr. Luiz Alfredo Hemández, both from CIAT. This was a ten-day training session, that included theoretical classes and practical field work. Thirty researchers and extension workers, from six state institutions of four S tates, attended the course. Visiting COP ALs During 1996, two World Bank' s missions, together with EMBRAPA's representatives, visited the COP AL Cadete in Bahía. The first visit to the COP AL Cadete was a suggestion of the CNPMF Director ofResearch and Development since the World Bank representatives were interested in visiting small growers communities. The second visit was a consequence of the first one. In both occasions the visitors were very impressed with the level of grower participation on the research activities carried out by the COP AL. Following the first visit ofthe World Bank mission, an EMBRAPA consultant for technology transfer was recommended to visit EMBRAP A/CNPMF in order to better know the COP AL methodology implemented by PROFISMA After discussing with PROFISMA staff the theoretical bases of the farmer participatory research methodology adopted for the COP AL model, he visited the COP AL Colonia Agrícola Roberto Santos, located in the municipality of Inhambupe, and the COP AL Chapada, located in the municipality of Aporá, both in the State ofBahia. 19 By early December, 1996, the Coordinator of Agricultura! Development ofPROSERTAO (an IFAD-funded development project in Sergipe), together with 13 extension workers from that project, visited EMBRAP A/CNPMF in arder to maintain contact with the coordinators of PROFISMA and the IF AD-funded project conducted by EMBRAP A/CNPMF. He expressed an interest in visiting a COP AL in arder to observe in situ how the COP AL methodology has been implemented among cassavagrowers. The PROSERTAO team visited the COPAL Colonia, located in lnhambupe, State ofBahia and was favorably impressed. Graduate students (Institutional Relationship) Three graduate students from the Universidade Federal da Bahía, have been conducting their research at EMBRAPNCNPMF, working on PROFISMA activities as dissertation subjects, including: 1) effects of cover crops on physical and chernical properties of soil in cassava, 2) weed management in cassava growing areas, and 3) control of cassava root rot disease. PROFISMA-CNPMF Scientists are the majar professors of these students. 2.3 Phase Two Proposals, by Lincoln Srnith A concept note entitled, "Ecologically Sustainable Cassava Production Systems for Latín America" was prepared by CIAT and sent to 25 donors in May with the goal offinding additional donors to support a subsequent project based on the successes ofthis project. The donors contacted included: United Nations Development Programrne (UNDP), Carnegie Corporation ofNew York, The Ford Foundation, Intemational Development Research Centre (IDRC), World Bank, Intemational Fund for Agricultura} Development (IFAD), Henry P. Kendall Foundation, W. K. Kellogg Foundation, John D & Catherine T. MacArthur Foundation, The Pew Charitable Trusts, The lntemational Foundation, Inc., Agency for Intemational Development (USAID), W. Alton Jones Foundation, Inc., Jessie Srnith Noyes Foundation, Weeden Foundation, The Charles Stewart Mott Foundation, Andrew W. Mellan Foundation, The Rockefeller Foundation, Inter-American Development Bank, Ciba-Geigy Foundation for Cooperation with Developing Countries, Swiss Agency for Development and Cooperation (SDC), Swiss Centre for Intemational Agricultur-e (ZIL), The McKnight F oundation. . A second concept note entitled, "Ecologically Sustainable Cassava Plant Protection: A Global Strategy" was written jointly by CIA T and ITT A in July to unify the Latin American and Afiican proposal concepts. This concept note was sent to UNDP for consideration of possible funding of a second phas~ of the PROFISMAJESCaPP project beginning in 1997. 20 The training component ofPROFISMA Phase 2 was prepared by PROFISMA's Training Coordinator in collaboration with EMBRAP A personnel and was sent to the Chairman of the EMBRAP A Intemational Cooperation Department. The Plant Pathology and Entomology components ofPROFISMA Phase 2 were prepared by PROFISMNCNPMF staff, based on cassava growers' demands as identified by the participatory diagnostic surveys. These components will be discussed with the CNPMF cassava research team for further adjustments, before sending to donors. Visiting UNDP Headquarters The EMBRAP A/CNPMF Director of Research and Development, together with CIA T' s representatives, visited UNDP headquarters in order to discuss financia! support for the extension of the Project, starting in 1997. Arrangements for continuing PROFISMA activities during 1997 The project coordinator for PROFISMA is terminating bis position at the end of December 1996. This will substantially reduce project costs, helping us to retain other essential technical personnel during 1997. However, it will reduce the amount of supervision and coordination of activities between CIA T and CNPMF. It al so greatly decreases our capability to communicate wíth donors in the search for funding to support subsequent projects that will extend and solidify the progress that this project has obtained to date. A money-saving policy was implemented at EMBRAP A/CNPMF during 1996, based on two main actions: 1) EMBRAPA's financia! aid for operational costs, and administrative and scientific personnel; and 2) rationally reducing project personnel. Two field workers, the 21 project secretary, one lab technician, the Virologist and the Environmentalist were dismissed. In addition, EMBRAPA did not charge PROFISMA overhead costs to UNDP during 1996. Such procedures made it possible to maintain PROFISMA activities during 1997, though at reduced levels. The operational budget for 1997 will be provided by EMBRAP A, which is also expected to supply additional personnel when necessary. PROFISMA personnel will be paid using carryover from the 1926 budget plus bridging money provided by UNDP, as previously discussed with UNDP representatives. 22 3 TRAINING AND PARTICIPATORY RESEARCH 3.1 Training, by Bernardo Ospina Patiño Counes During 1996, the last component ofthe training model proposed by PROFISMA was implemented through a course on·''Participatory Methods for Evaluation ofTechnology Trials with Farrners" held at CNPMF/EMBRAPA in April, 1996. This last training stage was attended by 30 researchers and extension workers and was aimed at improving theirs skills, abilities and attitudes for using two participatory methods in the evaluation of technology testing trials with farrners. These trainees, representing 6 collaborating research and technology transfer institutions in the states ofBahia, Ceará, Pernambuco and Paraíba, were responsible, u pon finishing the course, for assisting the COP ALs and their communities in the harvest and evaluation ofthe trials installed during the 1995-96 growing season. Establishment of new COPALs Fonnation of a second generation of COP ALs was initiated during 1996 in the S tate of Ceará through a project presented by local institutions to the Banco do Nordeste do Brasil (BNB). This project financed the establishment of COP ALs in six communities of cassava growers that had been identified during the participatory survey carried out by PROFISMA in 1994. Financial support received from BNB included seed money for the first COPAL experiment, funds for training events for fanners, and fuel and per diems for technicians assisting the COPAL. Organization ofthe COPALs although autonomously conducted by local institutions followed the same methodology proposed by PROFISMA. One interesting aspect of this effort at the state level is that the six new COP ALs are working on the same line of research: testing the use of organic compost and earthwonns . . Field days Two fi.eld days were organized during this period with the aim of getting more support for COP AL activities at local- and state-levels. During these events, the COP AL methodology for 23 enhancing fanner participation in research and technology transfer activities and the initial results obtained by PROFISMA were presented to larger audiences of fanners, technicians and decision-makers. One event was held in the State ofCeará, organized by EMATERCE and hosted by the COPALs located in the communities Nova Veneza and Valparaiso in Ubajara. Total attendance to this event was near 100 persons including fanners from the six new COP ALs established this year .. The second field da y was organized in the S tate of Pernambuco under the coordination ofiPA and EMATER-Pe, the two local collaborating agencies. This event was hosted by the COPAL in the community Sao Bento do Una and was attended by 60 persons. Joint training activities PROFISMAIESCaPP Profisma' s training coordinator participated in 1996 annual evaluation ofESCaPP activities. During this meeting held at Cotonou, Benin, PROFISMA' s methodology for fanner participatory research was presented, and despite the fact that ESCaPP is implementing its own strategy for training and fanner research, they became very interested in organizing a training event for African scientists based on PROFISMA's participatory research strategy. The aim of this activity would be to facilitate transference and adaptation of these methods and experiences to the African countries with which ESCaPP is operating. A proposal was jointly elaborated and has already been submitted by liT A to IDRC-Canada seeking financia! support for its implementation. The proposal is based on a three to four weeks course on FPR and COP AL methodologies to be held both at CIAT (Colombia) and EMBRAP A/CNPMF (Brazil) with a target audience of 13 selected scientists from Benin, Cameroon, Ghana and Nigeria. These trainees are currently collaborating with ESCaPP. If the proposal is approved, the proposed period for this event is the first quarter of 1997. The importance ofthis activity lies not only in the fact that it was recommended by the UNDP- assembled Externa! Advisory Committee, but also for its characteristics ofbeing an inter- projects (ESCaPP and PROFISMA), inter-centers (CIAT, UTA, EMBRAPA), and ínter- continental training activity (Africa, South America). 24 Joint training activities PROFISMAIPROSERTÁO PROSERTÁO is a rural development project executed by the Government ofthe State of Sergipe, Northeast Brazil, with financia! support from IF AD through a 25 million do llar loan during a seven year period. This project includes an agricultura! development component, and initial information collected by a survey of the target area identified cassava root rot as the main constraint for agricultura! production. This disease was found in 95% of the municipalities included in the area of influence of the project. Given the magnitude of the problem and considering the fact that cassava is the main source of income for small-scale farmers in the region, PROSERT ÁO staff approached CNPMF looking for technical assistance to salve the problem. After initial discussions and exchange visits, a proposal was formulated that includes a) the implementation of a training program in FPR and COP AL methodology for PROSERTÁO personnel, b) establishment ofCOPALs in sorne selected cassava growing areas ande) execution ofparticipatory technology generation and transfer activities for control of cassava root rot. This last activities would be based on a combination of improved cultural practices and the use of resistant/tolerant cassava varieties and hybrids. This genetic material is being developed by CNPMF through the project "Development of Improved Cassava Germplasm for the Semi-arid Zone ofNortheast Brazil", financed also by IF AD and executed jointly by CIA T and CNPMF since 1992. The successful results obtained with this project coupled with PROFISMA' s experiences adapting participatory methods for farmer participation in research and technology generation and transfer activities to Northeast Brazil, could be fundamental factors in the execution of these activities. Additionally, the existence of training material airead y in Portuguese and the possibility of using sorne COP ALs as training sites will facilitate the implementation of this activity and the dissemination of the COP AL methodology in other regions of Northeast Brazil, a spin-off effect of the PROFISMA project. COPAL visits to EMBRAPA/CNPMF headquarters Visits to CNPMF research station were organized this year for the COP ALs located in the S tate ofBahia. The objective of this training activity was to allow farmer groups to become more familiarized with PROFISMA's activities, to increase their knowledge about specific • 25 tapies selected by them and to discuss altemative technological options that are currently available or under development at CNP:MF and that could be incorporated in future technology testing activities to be irnplemented by the COP ALs. For each visit, which lasted two days, PROFISMA staff prepared presentations, practical demonstrations and visits to laboratories and cassava fields, following agendas designed with collaboration ofEBDA personnel who are responsible for: technical assistance to the COP ALs. These agendas were designed to meet farmers' demands and priorities. A total of six COP ALS visited CNPMF during this year with participation of 160 cassava growers from the municipalities ofPiritiba, Alagoinhas, Inhambupe, Anguera, Aporá and Cruz das Almas. Participatory evaluation oftrials established by the COPALs in the 1995-96 growing season During 1996, the group of researchers and extension workers collaborating with PROFISMA in the States ofBahia, Ceará, Pernambuco and Paraíba, who are responsible for giving year-round technical assistance to the COP ALs, faced a great challenge with the harvest of the first technology testing trials established by the COP ALs. After covering the three training stages (diagnoses, planning and evaluation), this group oftechnicians finally was able to use the participatory methods for evaluation of technology trials with farmers. PROFISMA's strategy for this activity was defined as a combination oftwo methods: a) agronomic evaluation (the one commonly realized by researchers to determine yields, dry matter content, and other parameters) and b) technology evaluation by farmers (evaluation oftechnology trials using complementary techniques). The technique utilized was the so called Preference Ranking. For each evaluation the procedure was as follows: • The COP AL coordinated the activities and requested assistance from other interested farmers in the community. • A short meeting was held to discuss the activities, the organization and to assign responsibilities. • Farmers participated in the harvest of each treatment, each replication and each block. 26 • Cassava roots and aerial plant parts harvested for each replication and each experimental block were laid on the floor grouped by treatments, and farmers were asked to rate them (the best, the second, the last, etc.) using small pieces ofpaper with printed numbers. Each farmer represented one vote, and he voted on all the treatments. During their ranking of the different treatments, farmers were allowed to discuss among themselves about the criteria they were using. Technicians were asked to avoid any participation in this step. • After all the treatments have been ranked by the farmers, a final counting was made assigning points to each classification. For example, in an experiment with four treatments, the best treatment was assigned number 1 and was worth four points; the second, was assigned number two and was worth three points; the third, number three and was worth two points and the last one, was assigned number fourth and was worth one point. By adding the points assigned to each treatment, a final Preference Ranking, according to farmers criteria was obtained. • Farmers were invited to comment on this ranking (especially regarding the criteria they u sed). • Agronomic evaluation was conducted jointly by technicians and farmers. Each treatment was weighed (cassava roots and aerial plant parts). For dry matter determinations a sample was taken for each treatment, and the dry cassava content was determined in front of the farmers using the hydrostatic balance method. • A rapid analysis of the results was made on the spot and was u sed to stimulate the final discussion with the farmers. Raw, semi-processed data ofthe results ofthe experiment was left with the COP AL the same day. • A more complete report on each experiment was sent later to each researcher and extension worker assisting the COP AL in the interpretation of the results, its presentation to the larger community, and more important, the drawing of conclusions and proposals for the second experiment .. First experiences and results obtained by PROFISMA about the use of participatory methods for evaluation of technology trials with farmers are very encouraging. The technicians in volved, despite the fact that they are in most of the cases new practitioners, have 27 been able to adapt the methods to their own conditions and have stimulated farmers to make their own evaluations. Conversely, the farmer members of the COP ALs and the other fanners that collaborated with them managed to produce their own evaluation results and made a very important step forward in the process ofbuilding local, community level capacity to participate actively in technology generation and transfer activities. Another very important result of these activities was the strengthening of the collaborative relation between the technicians and the COP ALs. In the majority of the trials evaluated, results obtained with the agronomic evaluation were consistent with those obtained in the subjective evaluations made by the farmers. Their choices were supported by yield and dry matter content data. However, in sorne cases, it was clear that farmers' choices were not necessarily those options with the best agronomic results. This fact helped fanners to gain awareness about the importance of diversifying the technological options available for them and increasing the number and the frequency of technologies tested in their communities. In these cases, farmers reacted very enthusiastically and showed great interest to continue the testing and adaptation process with the technological option that showed good potential to be incorporated in their farming systems. Appendix 1 presents more detailed information on the participatory technology testing trials evaluated by the COP ALs during the 1995-96 growing cycle. 3.2 Participatory Evaluation of Technology Testing Trials 3.2.1 S tate of Babia COPAL: Colonia Agrícola Roberto Santos, Inhambupe The objective of the experiment was to evaluate six cassava varieties for resistance/tolerance to cassava green mites (CGM). This had been prioritized by farmers as the main constraint for cassava production in the region. The six varieties were selected locally by the farmers from genetic materials that they have planted over the years. This experiment did not include any variety from EMBRAPNCNPMF. Farmers in this community represent an atypical example of cassava growers since they are used to put heavy amounts of fertilizers in their crops ( cassava, citrus and passion fruit) and ha ve been doing it for many 28 years. The excellent results obtained in this experiment confirm this. Figure 3 .2.1.1 presents the results obtained for yield (ton/ha), dry matter content (ton/ha) and the preference ranking that farmers made of the treatments included in the experiment. Two of the varieties gave yields higher than 35 tons, four ofthem gave yields higher than 30 tons, figures that are well over state and national average yields for cassava. The local variety, used as check in this experiment (Platina) gave the low_est yield (27 tonlher) although farmers ranked itas 2nd in their preferences. The two varieties that gave higher yield were ranked by farmers in 3rd and 4th place. Finding out about the excellent performance of other varieties different from the local one was one of the main conclusions and lessons learnt by the members of this COP AL and their collaborating farmers. T o increase farmers knowledge about CGM and to create local capacity for making sound decisions about cassava crop management, farmers from this COP AL were trained to identify the CGM, and to malee field evaluations (damage score) ofthe level ofincidence ofthis pest. Farmers were then able to perform their own weekly evaluations on the pest attack and each variety was ranked according to the incidence of the pest. The information obtained by farmers was sent to the CNPMF entomologist assisting them who prepared a final evaluation ofthe CGM incidence during the cropping cycle, the cumulative damage score (Figure 3 .2.1.2). lt can be observed from this data the good tolerance of the local varieties to mite damage, the main reason for its popularity among farmers . However, this local variety did not give the best yield, and sorne of the introduced varieties that showed medium tolerance to mites were more productive. These results were enthusiastically discussed by the farrners who realized the possibility of introducing and adapting new, more productive cassava varieties into their farming systems. Results obtained in the experiment were discussed among COPAL members and other farrners and a second experiment was planned and planted. This new experiment include the top three varieties of the first trial as well as three new varieties recommended by EMBRAP A/CNPMF from its germplasm bank. 70 60 50 e 4o o ~ 30 20 10 o 40 35 30 ~ 25 ... &. 20 e: ::. 15 10 5 Voadera Cravelao 35.7 35.1 o -+-....1...-- Voadera Cravelao 58 El Farmers' preference ranking Bom Branca Platina Barrinha Jardim Leite 32.6 Bom Jardim O Root yield • Dry matter 1 31 .5 Branca Leite 28.2 Platina 27.0 Barrinha 29 Figure 3.2.1.1. Evaluation of six cassava varieties for resistance/tolerance to cassava green mite; Agronomic and Preference Ranking Evaluation at Colonia Agricola R. Santos, Inhambupe, Bahia. 30 • Voadeira · · + · · Cravelao 2000 • Bom Jardim • Branca Leite • Platina 6 Barrinha e 1800 760 o 1600 .. ... .... . ~2 u lit ..... cu 1400 .... ..... al ca 1200 E ca 1000 'tJ cu 91 > 800 ~ 21 ~ 600 ::::J 80 E 400 ::::J o 200 o o 11 53 60 68 76 81 90 96 104 112 127 144 156 175 No. of days after planting Figure 3.2.1.2. Evaluation of cassava green mite damage on six cassava varieties at COPAL: Colonia Agricola Rogerto Santos, Inhambupe, Babia. COPAL: Chapada, Aporá The objective ofthe experiment in this COPAL was to evaluate the effect oftwo cassava varieties and two "cultural" practices (planting method) on the control of cassava root rot, a disease that has become a major constraint for cassava production in various states of Northeast Brazil. Farmers had prioritized this problemas their principal production constraint. One ofthe varieties (Osso Duro) was recommended by CNPMF researchers as a good source of resistance based on experimental results obtained in past years. The other cultivar (Cernitério) was selected by fanners as one the best genetic materials they have in the region. One of the cultural practices was recommended by researchers (planting stakes in ridges of soil) and the other is commonly used by farmers (planting on tlat soil) . The experiment was 31 planted in complete randomized blocks with four repetitions, and fertilizers were not applied. Yields obtained in this experiment were extremely low indicating the need for more intense long-tenn work in this comrnunity aimed at improving their cassava-based production system. The variety recomrnended by researchers showed a very low performance, giving the lowest yields. Low yields by this variety were also obtained in experiments in other regions during 1996 and has been attributed to a_fé;lilure in the resistance that it had shown in the past. This material is being discarded by CNPMF as a source of cassava resistance to root rot. The cultural practices results were more encouraging since the recomrnended practice, planting in ridges, gave the best results for both varieties. The local variety planted in ridges gave 44.6% more yield than with flat planting, and the introduced variety produced 25.5% more when planted on ridges. This recommendation, which is not currently used by fanners in the region, could be the basis offuture technology adaptation work. Fanner evaluation ofthe experiment (preference ranking), was consistent with the agronomic results, selecting as their best choices those treatments that gave the best agronomic results (Figure 3.2.1.3). Future work in this community would be a great challenge for PROFISMA, collaborating institutions and for the COP AL itself The community is one of the poorer fanner groups, and they are using cassava varieties that ha ve extreme} y low productivity. It is surprising to notice that a few kilometers away there are other communities with varieties that are three or four times more productive. The future role of the COP AL could be crucial in the strategy of introducing and adapting better cassava varieties and improved cultural practices with the overall objective of improving cassava production, productivity and farmer in comes. Sorne cassava germplasm developed recently by CNPMF that is showing good resistanceltolerance to cassava root rot could be very important in future participatory research and technology transfer activities implemented through the COP AL in this region. Results obtained in the experiment were discussed among COP AL members and other farmers, and a second experiment was planned including improved varieties introduced from EMBRAP A/CNPMF germplasm bank. 20 18 cu .e ""' 16 G) Q. 14 0 -g 12 ""' cu > 10 cu 0 8 0 cu u 6 -o e 4 o .... 2 o 100 90 80 70 60 18.8 6.0 Cemitério + ridges planting 86 32 1 o cassava root yield • dry matter content 1 13.0 11.2 8.9 4.1 3.6 Cemitério + flat Osso duro+ Osso duro + flat planting ridges planting planting 1 O Farmers' preference ranking 1 66 e o u tn 50 40 30 20 10 o Cemitério + ridge planting . Cemitério + flat planting 37 Osso duro + ridge planting 31 Osso duro + flat planting Figure 3.2.1.3. Evaluation of varietal resistance/tolerance and cultural practices for cassava root rot control. Agronomic and preference ranking evaluation at Chapada, Aporá, Babia. 33 COPAL: Buril, Crisópolis The prioritized problem in this community was also the incidence of cassava green mite (CGM). The objective of the experiment was to evaluate six cassava varieties for a resistance to CGM. Four ofthe cultivars were directly introduced from the CNP"MF germplasm bank and had previously been identified through farmer participatory selection conducted in a germplasm development project financed by IFAD. The other two were selected by farmers from among the best locally produced material. The experiment was planted in randomized blocks with three repetitions. Planting distances were 1. O by 1. O m. One of the introduced varieties performed better than the local ones (Figure 3.2.1.4). The best introduced material produced 19% more roots yield than the best local variety. Two ofthe introduced materials showed lower dry matter (poorer quality) content than the local varieties, a factor to be considered in future breeding work. Farmers' preference ranking evaluation was not consistent with the agronomic data. They selected as the best two treatments the two local varieties (Platina and Jalé) that gave lower yields than the two best introduced materials (Maria Pau 118 and 128/08). F armers ranked three of the introduced materials as worse options although one of them, Maria Pau 118, was the most productive (Figure 3.2. 1.4). The results were discussed with the COPAL and with other farmers ofthe community, and the planning of the second experiment resulted in the inclusion of these two promising clones, the introduction from CNPMF of two new ones and the inclusion of the two best local varieties as the check treatment. Members of this COP AL al so participated in training activities for evaluation of cassava green mite damage. The results oftheir evaluations are presented in Figure 3.2.1.5. It can be observed that Jalé, the preferred local variety did show the best resistance to cassava mites, justifying its local popularity. Three of the four varieties introduced from CNP"MF showed medium tolerance to mite damage, but two ofthem (Osso Duro and 47/19) gave the lowest yields ofthe experiment. The other variety, Maria Pau 118, was the most productive, and although it was ranked very poorly by farmers, it showed them the possibility of adapting it to their cassava systems since it produced 19% more that their local best variety. 20 18 16 14 ~ 12 ... 8. 10 e {:. 8 6 4 11.5 2 o-+-........___ 90 80 10 60 ! 50 o u 40 fl) 30 20 10 o Maria Pau118 38 r---- Maria Pau 118 15 128/08 67 ,.--- 128/08 o Root yield • Dry matter 1 14.7 13.25 12.9 12.45 Jalé Platina Osso Duro 47/19 82 O Fanners' preference ranking ...--- 68 .....- 44 r---- 16 n Jalé Platina Osso Duro 47/19 Figure 3.2.1.4. Evaluation of six cassava varieties for resistance/tolerance to CGM Agronomic and Preference Ranking Evaluation, COPAL: Buril, Crisópolis, Babia. 34 35 • 128-08 ··+· · Platina 2000 • Maria Pau 118 • 47/19 e 1800 • Osso Duro • Jalé o 1600 u •... •· ... CIJ 1400 .. . . •· .. IU .•... c:ft ca 1200 E ca 1000 'C IU > 800 ;: ca 600 "S E 400 :S (J 200 o o 11 53 60 68 76 81 90 96 104 112 127 144 156 175 No. of days after planting Figure 3.2.1.5. Evaluation of cassava green mite damage at Buril, Crisópolis, Babia. COPAL : Cadete, Cruz das Almas Located 3 km from EMBRAP A/CNPMF headquarters, this COP AL has a comparative advantage over the rest because its proximity allows closer contact between farmers and technicians (researchers and extension workers). It also has become an excellent training and demonstration ground for PROFISMA and in general for CNPMF. The farmer group decided to plant an experiment to compare two methods of fertilization: organic compost and chemical fertilizer. Both are commonly used in the region, and there are sorne farmers that complain about the inefficiency and the high costs of the chemical fertilizers . The two more popular cassava varieties were used, and the experiment was planted in two different fields. Data discussed in this report refers to only one of them because soil hardness has prevented harvesting the other one. The results obtained were very interesting. First, there seems to be a varietal difference in the response to fertilization (Figure 3.2.1.6}. One ofthe varieties (Cidade Rica) presented yield increases of 64% due to the use of organic fertilizer and of only 11% with the mixture of organic plus chemical. Chemical fertilizer alone with this variety gave the lower yield ofthe whole experiment. Results with the other variety (Cigana Preta) were even more interesting since the best yields were obtained without fertilizers . However, the other 36 variety with organic compost yielded 4 7% more. In relation to dry matter yield, the same tendency was observed, with the organic fertilization of the variety Cidade Rica showing the best yield (about 67% more dry matter yield per ha than the rest ofthe treatments). Farmer evaluation of the experiment (preference ranking}, was consistent with the agronomic results, selecting as their three best choices these treatments with the best the best agronomic results (Figure 3 .2 .1. 6). Based on these results, the main point of discussion among COP AL members, other farmers and technicians was about the convenience of using fertilizers, especially chemical in cassava production. Farmers argued that this fertilizer is easy to get and sometimes cheaper. They are now starting to question the economic viability of this practice. Building up in this initial experience, the COP AL has already planted a second experiment in which sorne improved varieties recommended by CNPMF were introduced. COP AL: Barra, Sáo Miguel das Matas, Babia Cassava farmers in this cornmunity diagnosed whiteflies (Aleurothrixus aepim) as the main constraint to cassava cultivation in the region. The objective of the experiment designed with this COP AL was to evaluate five cassava varieties for whitefly resistance/tolerance. Three of the varieties planted were introduced from CNPMF germplasm bank and were recommended for testing by the COP AL based on excellent results obtained in other experiments in similar semi-arid ecosystems. These three clones were also selected as promising varieties by farmer groups involved in another farmer participatory research project being executed by CNPMF and CIAT in Northeast Brazil. The other two varieties were selected by the farmers as their best locally available options. The experimental design was complete randomized blocks with five repetitions (five different farmer fields located within a distance of about 1 km). Although the main objective of the experiment could not be addressed to a great extent due to the absence of a severe attack of whiteflies during the experiment, the results obtained allowed farmer members ofthe COPAL, other farmers ofthe community and technicians sorne important initial conclusions. • .1: 27 ~ 22 Q. .. o e 17 • > 12 • .. .. 7 .. u o 2 e o 1- -3 10 "' 9 .e 8 ... 8. ... 7 " 6 = "' 5 E ~ 4 'O -o 3 e 2 o 1- 1 o 70 60 50 " ... 40 o u 24.8 14.4 cassava + organic fertlllzer 9.1 cassava + organic fertilizer 64.0 Root yields 16.8 16.6 CCidade Rica •Cigana Preta 14.2 15.1 cassava + organlc + chemical fertlllzer cassava wlthout fertilizer cassava + chemlcal fertillzer Dry matter yields 6.0 5.6 5.5 5.7 cassava + organic + cassava without fertilizer chemical fertilizer Farmers ' preference ranking 46.0 49.0 5.3 5.6 cassava + chemical fertilizer 1/) 30 20 10 o cassava + organic cassava + organic + cassava without cassava + chemical fertilizer chemical fertilizer fertllizer fertilizer Figure 3.2.1.6. Evaluation of fertilization methods on yields of two cassava varieties at COPAL Cadete, Cruz das Almas, Bahia. 37 38 First, two ofthe varieties introduced from CNPMF gave yíelds (roots and dry matter) 50% higher than the two best local materials; a fact that caught the interest of the farmers for further testing and adaptation of these materials to the region (Figure 3 .2.1. 7). Second, cassava yíelds with local varieties is extremely low in this region considering the fact that they commonly use chemical fertilizers_. Jhe farmers' main cash crop is tobacco, and they obtain the fertilizers directly from the tobacco companies in exchange for the product, at prices fixed unilaterally by these companies. This represents a very interesting topic for future research by the COP AL, which, besides the interest for new, improved varieties, expressed interest in testing organic and chemical fertilizers and their effect in the production oftobacco and cassava. They argued that this could help them becoming less dependent from the tobacco companies. Last but not least, this experiment served to confirm the importance of allowing farmers opinions to be incorporated in the development and dissemination of cassava germplasm. As shown with the results obtained, when the cassava varieties pass through a selection process during which farmers have the chance to evaluate them using their own criteria, the possibility of good performance and good acceptance by other farmers when these materials are taken into similar ecosystems is greatly enhanced. e o u U) 18 15.8 15.8 O Root yield • Dry matter 16 14 11.8 11.1 12 10.5 ca &:. '- 10 G) Q. e 8 o .... 5.4 6 5.2 3.4 4.1 3.5 4 2 o 128-8 189-11 192-13 Corrente Sao José 124.0 armers pre erence ran mg 122.0 OF 140 120 100 80 81.0 77.0 68.0 60 40 20 o 128-8 189-11 192-13 Corrente Sao José Figure 3.2.1. 7. Evaluation of five cassava varieties for resistance/tolerance to whiteflies: Agronomic and Preference Ranking Evaluation at Barra~ Sáo Miguel das Matas~ Babia. 39 40 COPAL: Umbuzeiro, Anguera, Babia Tbis COPAL is the only one ofthose established with PROFISMA's support in which the majority of the farmers are women. Cassava producing and processing in this community is done mainly by women who work on small pieces ofland, usually rented. Farmers prioritized low productivity and poor fertility of their soils as the constraints on which they would like to work. Considering the lack of economic resources of the group, it was decided during the planning stage that the main objective of the experiment would be to test the effect of planting cassava inter-cropped with legumes which could help to improve soil fertility. The two legumes chosen were cowpea (Vigna unguicul/ata) and jack bean (Canava/ia ensiformis). To facilitate crop management the double-row planting system was introduced. Farmers decided to use their best local variety, and the experiment was planted in three different farmer fields. Results obtained indicated the high degree of complexity when an attempt is made to modify a local, traditional farming system. In this case, the use of the two legumes associated with the cassava crop demanded certain specific management, especially the timing of the pruning, which was not clearly understood by the farmers, and even by the technicians giving technical assistance to them. Due to different problems, the technicians were not able to maintain a systematic, periodic presence in the community to advise farmers' decision-making on management of the experiment. As a consequence, these two legumes were allowed to grow beyond flowering stage. Competition with cassava was very severe, especially during a drought period that affected the region in the early stages of the experiment, and cassava yields were very low. The two treatments without legumes resulted in higher yields, and the introduced double row planting system was slightly better than the traditional, single row system. Farmers' preference ranking evaluation system was consistent with the agronomic results (Figure 3 .2.1. 8). 14 12 "' 10 .&:. ... 8 Gl ~ e 6 o .... 4 2 o 70 60 50 40 e o u (/) 30 20 10 o 12.2 4.4 Double row 46 Double row 11.6 4 Single row 65 Single row 41 ORootyield • Dry matter 1 4.3 3.2 1.5 Double row + Double row + jack cowpea bean o Farmers' preteren ce ranking 26 Double row + cowpea 3.2 r l Double row + jack bean Figure 3.2.1.8. Evaluation of cassava planting systems (double row, single row and legumes intercropping) at Umbuzeiro, Anguera, Babia. 42 The lesson leamt by the researchers and extension workers giving technical assistance and advice to the farmers was very irnportant. lt became clear that there is a need to maintain a systematic, periodic presence in the community to advise the COP AL farmers, especially in those moments in which they are forced to make decisions about crop management in farming systems and practices that are new to them. Another irnportant conclusion was that during the initial experiences of farmer groups. as researchers it does not seem appropriate to mix too many treatments (for example, planting distances, inter-cropping, etc.). It is better to initiate the work as simply as possible allowing farmers to gain experience little by little as the work becomes more complex. Farmers discussions after the harvest and evaluation oftheir experiment indicated their desire to introduce improved cassava varieties in their future activities. It is expected that genetic material which is currently available at CNPMF's germplasm bank and that has been especially developed for these semiarid, low soil fertility ecosystems will be fundamental in the continuation of the work by this COP AL. 3.2.2 S tate of Ceará COPAL: Nova Veneza, Ubajara, Ceará The community ofNova Veneza is composed by 120 families and is located in the Serra de Ibiapaba, a region characterized by a relatively good annual rainfall and moderate climate which allows year-round production of various cops such us sugar cane, fiuits, com, beans and cassava. The importance of cassava in the region is mostly for farmers ' own consumption of cassava flour (farinha de mandioca) since the other crops are more important as a cash and income source. Nonetheless, cassava is widely grown in the region, and the COP AL in this community identified as the main production constraint the lack of improved, more productive varieties and altemative, more efficient cropping practices, especially fertilization. Additionally, cassava production in the region has been affected drastically during the last five years by the disease known as mycoplasm witches' broom (WB) which caused asevere reduction in production, productivity and area planted. 43 Fanners included two local varieties in their experiment: one, Cruvela, is a local cultivar widely grown in the region and the other, Buja de Olho Roxo, was introduced as one ofthe varieties that shows sorne tolerance to WB. The experiment was planted with a randomized block design with six treatments and three repetitions. The two main characteristics of the experiment were the introduction of two legumes: Crotalaria and pigeon pea, as sources of green manure, and the use ofCompost, a practice that the Extension Service (EMATERCE) is promoting intensively in the region. The results of the experiment are presented in Figure 3.2.2.1. It can be observed that the introduced cultivar, Olho Roxo gave higher yields for all treatments (an average increment of 48% for the whole experiment), and all the treatments in which the compost was utilized al so gave higher yields ( an average increase of 128% for the variety Cruvela and of 41% for the variety 01ho Roxo. The overall best result was obtained by intercropping cassava with pigeon pea and using compost. This practice compared with the traditionaJ system of planting cassava alone gave a yield in crease of 225% for the variety Cruvela and of 69% for variety Olho Roxo ). Results obtained with dry matter content showed the same trend suggesting a beneficia) effect of the legume and the compost on the dry matter content ofthe cassava roots. It has been reported in the literature that pigeon pea is one ofthe few legumes that has a unique mechanism that allows them to access phosphate that is bound to caJcium and iron soil particles. These legumes release acids from their roots which react with caJcium-bound and iron-bound phosphate to release phosphate for plant uptake. AdditionaJly, their deep rooting characteristic has a positive effect helping water infiltration. Preference ranking evaJuation conducted by fanners was somehow different from the agronomic results. Fanners gave selected as their best choice, for both varieties, a treatment which was the third place in yields and dry matter content. Their choices for the worse treatment did coincide with the agronomic evaJuation (Figure 3 .2.2.1 ) . An interesting aspect of the work conducted by this COP AL is that they are now utilízing their experiences and skills as "farmer researchers" and have installed sorne simple experiments with other crops such as pepper and tomato. AdditionaJly, this COP AL is being used by EMA TER CE as a demonstration, training ground in which other comrnunities are invited for fields days on COP AL methodology, compost fertilization and other topics. In August 1996, this COP AL hosted with great competence the First S tate Day for COP ALs prometed by local collaborating institutions. 18 14 11 .2 10 l • ! 8 4 2 14.0 0+-....... - !! 6 5 ~ 4 !.3 ! 2 1 Cassava + pigeon pea+ compost 4.6 0+-....... - 30 Cassava + pigeon pea+ compost ~ 20 10 0+-....... - Cassava+ plgeon pea+ compost Cassava roots yield Cassava + compost Cassava + Crotalaria + compost Cassava + Crotalaria Dry matter yield Cassava + compost Cassava + Crotalaria + compost Cassava + Crotalaña Farmers' prererence ranking Cassava+ compost 36 Cassava + Crotalaria + compost Cassava + Crotalaria CVariety: Cruvela •variety: Oulho Roxo Cassava + Cassava alone pigeon pea Cassava + Cassava alone pigeon pea Cassava + Cassava alone pigeon pea 44 Figure 3.2.2.1. Evaluation of cassava intercropping with pigeon pea and crotalaria in double row planting with compost fertilization at Nova Veneza Ubajara, Ceara. 45 COPAL: Valparaiso, Ubajara, Ceará This COP AL is the only farmers research group supported by PROFISMA that is located in a land reform community. The group is composed by 65 families and practices collective production and processing of various crops such as cassava, coro, beans, and banana. Cassava is a very important crop for the group, which is attempting to beco me self sufficient and also uses surplus cassava flour for commercialization. The crop has been affected severely during the last few years by the mycoplasm witches' broom disease (WB) and the group has been actively involved in research work conducted by CIA T, CNPMF and local institutions aimed at finding sources of genetic resistance/tolerance to this disease. The objective defined by the COP AL for its first experiment was then related to this problem. They were getting very enthusiastic about sorne hybrids that have started to come out from these activities and decided to test the one they liked best, the hybrid 8709-02 (local name is Salamanta), anda local clone named Cabelo de Velha. The experimental design also included the use ofthe legumes pigeon pea and Crotalaria and compost as fertilizer. The total number of management treatments was six, with three repetitions for each. The results obtained in the experiment were completely different for the two varieties. The hybrid Salamanta confirmed that it has indeed a great potential in the region giving higher yields than the local clone in five of the six treatments. The average yield for this hybrid for the whole experiment was 79% higher than average yield with the local variety. Considering the effect of the compost on yields, for the hybrid Sal amanta, it can be observed that average yield for the three treatments that have compost was 20% higher than these respective treatments without compost. The trend for the local variety was totally opposite, with the treatments without compost giving an average yield 55% higher than these respective treatments that were fertilized with compost. This data seems to suggest that there is a genetic difference in the response to fertilizer because in both cases, the compost was prepared at the same time and applied at the same rates. The intercropping with pigeon pea was the best treatment for the local variety and was the second best one for the introduced hybrid, confirming the beneficial effects of this practice on cassava yields. The dry matter content results showed the same pattems as the yield data for both varieties. Comparing the local, traditional planting system with the best treatments for each variety, it can be observed that for the introduced hybrid the maximum increase in yield was 26.5% whereas for the local variety this was 46.8% (Figure 3.2.2.2). 46 Preference ranking evaluation conducted by farmers was contrasting for both varieties. In the case of the introduced hybrid, although the local planting system had the lowest yields, farmers ranked it as their second choice, only exceeded by the treatment that included compost only. With the local vari~ty, farmers's opinions generally matched the agronomic results, and their best two options were exactly those with highest yields (Figure 3.2 .2 .2). Although the introduced hybrid was ranked equally with the local variety when each was planted alone, it was rated lower than the local variety when cultivated with compost (with or without pigeon pea)- despite the fact that yields were higher by 67 to 77%. The discrepancy between farmer ratings and yield indica tes the importance of using additional measures of crop characteristics to evaluate and select germplasm. The main conclusion of this experirnent is the ratification of the great potential that this hybrid seems to ha ve for the cassava systems in the region. CNPMF is intensifying its work on further testing, multiplication and release among farmers, as well as maintaining germplasm development activities to find similar or better genetic materials. In early 1997, a Field Day is plannedjointly between CNPMF, local institutions and the COP AL to officially deliver this hybrid to cassava growers in the region. Cl40 .1:: ... 30 8. ., 20 ª e 10 Cassava root yields 31.9 30.4 {J. 0+-......__ ...... 13 11 Cl .1:: 9 ! 7 e (l. 5 3 Cassava + compost Cassava + pigeon pea+ compost Dry matter yield 10.8 10.3 1 +-...L.-- Cassava + compost Cassava + pigeon pea+ compost Cassava + crotalaria + compost Cassava + crotalaria + compost Farmers' preference ranking 100 80 ! 60 o ~ 40 20 o-+-......__- Cassava + compost Cassava + Cassava + pigeon pea + crotalaria + compost compost . OVariedade 8709-02 (Salamanta) •variedade: Cabelo de Velha Cassava alone Cassava alone Cassava alone Cassava crotalaria Cassava + pigeon pea Cassava crotalaria Cassava + pigeon pea Cassava crotalaria 78.0 Cassava + pigeon pea Figure 3.2.2.2. Evaluation of cassava intercropping with pigeon pea and crotalaria in double row planting with compost fertilization at VALPARAISO, UBAJARA, CEARA. 47 48 COPAL: Vila Moura, Acaraú, Ceará This community is located in the semi-arid region of Ceará, a drought-prone area in which limited rainfall gives farmers very few agricultural options, among which cassava is on top of the list. Cassava production in the region is practiced intensively with a growing period that in average exceeds 20 months. Farmers argue that cassava production with just one "winter" is not profitable (rainy season) and plants are usually allowed to grow during the second year. The COP AL defined as its experimental objective the testing of organic compost as fertilizer in cassava production mixed with two legumes Gack bean and velvet bean). The two legumes were intended to be used as mulch and the experiment included the use of compost and a planting arrangement in double rows to facilitate farmers' management ofthe experiment. Since the main objective was to increase farmers ' knowledge about the management ofthis new system, and especíally of the two legumes, the farmers decided to harvest the experiment after one year. The best two local varieties were used. The experiment was planted in randomized blocks with six treatments and three repetitions. A similar experiment was planted in the second year switching positions between the cassava and the legumes in each plot. Evaluation ofboth experiments did not include determinations of dry matter and farmers ' preference ranking evaluation. The results ofthe experimentare presented in Figure 3.2.2.3 . It can be concluded from the data obtained that cassava production in the region with just one growing cycle gives indeed very low yields. The best results with both varieties were obtained using compost without legumes. The use of velvet bean as mulch was practically null due to severe attacks ofleaf- cutting ants that destroyed it. The adaptation of jack bean to the dry clima te was very good, but it became evident that sorne management changes have to be introduced because it competes severely with the cassava crop. Probably the best option is to use just two rows in between the cassava plants and increase the spacing. Nonetheless, farmers were very impressed with the growth of this legume and especially with the concept of growing a plant to be, later on, cut and incorporated into the soil. The best yield in both experiments was obtained with the variety Fragoso, in the treatment cassava alone + compost (13.45 ton/ha), a yield which is around the average in the region for two cycles. Sorne conclusions were evident for the members of the COP AL based on the results ofthis experiment: 12 10 ca 8 .:: ... el) 6 Q. e o .... 4 2 o 14 12 10 ca .:: ... 8 el) Q. e o .... 6 4 2 o Cassava root yield 1995 experiment 9.2 8.8 Cassava + compost 13.5 8.6 cassava + compost Cassava alone Cassava + jack bean Cassava + velvet bean + compost Cassava root yield 1996 experiment cassava alone cassava + jack bean 10.1 cassava + velvet bean + compost OVañety: Fragoso • Vañety: Geraldo Lopes Cassava + jack bean + compost Cassava + velvet bean cassava + jack cassava + velvet bean + compost bean Figure 3.2.2.3. Evaluation of cassava intercropping with jack bean and velvet bean in double row planting with compost fertilization at Vila Moura, Acaraú, Ceará. 49 50 • In the region farmers are used to plant corn and beans intercropped with cassava and this malees the use of legumes for soil fertility recuperation more difficult to be adopted by farmers; • Compost use proved to be an attractive, cheaper option for farmers; • There is a need to study the econornic viability of planting varieties with short growth period (harvest in one cycle ), -using organic compost. This system gave higher yields in this experiment. COP AL: Lagoa Grande , Acaraú, Ceará The Farmers Association ofLagoa Grande has been collaborating with local research and technology transfer institutions during the last decade. Cassava research and development projects (processing, germplasm development, crop protection) executed in the State by EMBRAP A, CIAT and local agencies had all included this group as one of the principal collaborators and beneficiaries. The Ceará Integrated Project for example, executed in Ceará during the period 1989-92, allowed the group to install a dry cassava chip agroindustry. During the last seven years, this farmer group has been the only one in Ceará that has been able to maintain every year a reasonable volume of processed dry cassava chips. The reasons behind this success are various, but probably the principal one is the excellent organizational and administrative level that the group has achieved. For example, in 1996 the profits obtained by the group with cassava processing activities were used to buy one ton of cow manure for each member. Farmers use organic fertilizer systematically in their fields and cassava production and productivity levels in the community are very good. The COP AL decided to work on cassava varieties for human consumption (sweet varieties). The main argument to justify this experimental objective was that sweet cassava varieties are almost disappearing in the region although in past years they were very popular among farmers . The COP AL feels that there is great demand and good market opportunities for these varieties and decided to start an identificationlobservation with those materials available in the State with the aim of adapting and selecting the most appropriate ones for their ecosystem. 51 The observation field was installed in an area closed to the community that belongs to a local University (Universidad do Vale do Acaraú). The integration with this University could be a very important support for the COP AL future technology adaptation efforts. The experiment included 21 varieties collected in different regions of the S tate. The planting system was double rows intercropped with six different legumes. These two practices were new to the group. After 15 months.the experiment was harvested, and farmers performed an evaluation of the best cultivars based on criteria such as plant development, root production and especially palatability (cooking time and softness). After this evaluation, only 10 cultivars were selected for future research work. Sorne legumes used in intercropping with the cassava were harvested and the seeds were sold to farmers in the region. This activity could be an additiona1 source of income for the farmers. 52 4 STRATEGIC RESEARCH AT CIAT 4.1 Biological Control of Cassava Green Mite, by Lincoln Smith The cassava green mite (CGM, Mononychellus tanajoa) is one ofthe principal pests of cassava in Africa and Northeast Braz!l._ The principal objectives ofthe CIAT acarology unit are to find natural enemies (predators or pathogens) ofCGM, evaluate them for safety and suitability as biological control agents, and send them to EMBRAP A in Brazil and liT A in Benin for release. Corollary activities include conducting research to further develop methods to evaluate natural enemies, mass rear them and release them; to publish findings and disseminate information to national programs; and to assist Brazilian scientists and extension agents working on integrated pest management of CGM. Exploration for natural enemies Explorations in 1996 were conducted to find phytoseiid predatory mites in two target ecological zones: 1) hot semi-arid ( corresponding to NE Brazil) and 2) high altitude/subtropical (corresponding to the East African plateau). For the hot semi-arid target, we collected 5 species of phytoseiids in Manabí, Ecuador from 11 sites with 425-1500 mm annual precipitation, 6-9 dry months (<60 mm rain), 24-26°C average temperature, and 16-120 m elevation. For the high altitude/subtropical target, we collected 6 species of phytoseiids from 20 si tes in Antioquia, Boyaca, Caldas, Cundinamarca, Quindio, Risaralda, Santander, and Tolima, Colombia from sites with 425-1500 mm annual precipitation, 0-4 dry months, 14-26°C average temperature, and 495- 2216 m elevation. We establíshed 7 new colonies of phytoseiids from high elevations for evaluation. A culture of Typhlodromalus aripo from Palmira was established for biological investigations. This appears to be the most effective of the three mi te predators established in Africa, yet little is known about why it is so effective. Also there appear to be two kinds of "T. aripo" which differ in their biology and may actually be different species. 53 Cultures ofthe fungal pélthogen Neozygites sp. were collected from cassava green mite and two-spotted spider mite (Tetranychus urticae) at CIAT, Palmira and from CGM at Media Luna, Colombia. Fig. 4.1.1 shows the distribution of 685 sites explored for mite predators of cassava green mite in Latin America by CIA T and EMBRAP A. Figure 4.1.1. Sites in Latín America sampled for predators of cassava green mite. 54 Mite colonies maintained Table 4 .1.llists the colonies ofphytoseüd predatory mites maintained at CIAT for experimental evaluation. Three species of T etranychid pest mites (Mononyche/lus tanajoa, M. caribbeanae, Tetranychus urticae) were maintained to feed the predators and to use in experiments. Several species ofinseGts (thrips Corynothrips stenopterus, mealybug Phenacoccus he"eni, whitetly Aieurotracheius socialis) and the foliar fungal plant pathogen, Oidium manihotis, were also maintained for use in feeding experiments. Table 4.1.1 Phytoseiid strains maintained in laboratory colonies at CIA T, 1996. Origin Date -·--·--·-·---·--··-·-·-· ·····--·-··-·· ·~------ --···--· ··-----··--·-· -----·-·-·--·· ··· ······ -·· ··-····· -·-- ····--~· -- Species Country Dept./State Municipality Location Collected Typhlodromalus VENEZUELA Yaracuy Marin San Felipe Mar-95 manihoti 11 BRASIL Bahia Cruz das Almas Feb-93 11 COLOMBIA Ca u ca Cajibio Jun-95 " Antioquia Barbosa El Hoyo Jan-96 11 " Antioquia Copacabana Montañita Jan-96 11 Magdalena Medialuna Feb-96 11 Guajira Villanueva Jan-96 " Caldas Chinchiná San Gregorio Aug-96 11 .. Santander B/manga Bijagual Aug-96 11 .. Santander B/manga Los Colorados Aug-96 " Risaralda Sta. Rosa de Cabal UNISARC Aug-96 " Quindío Armenia Armenia Aug-96 T aripo COLOMBIA Valle Palmira CIAT Sep-96 T tenuiscutus ECUADOR Manabí Portoviejo Nov-94 " H Manabí Puerto Cayo Cantagallo Dec-95 T. limonicus BRASIL Sao Paulo Jaguariuna Jun-90 N. idaeus ECUADOR Manabí Rocafuerte Entrada a Danzarín Dec-95 N. californicus ECUADOR Manabí Portoviejo Nov-94 G. annectens ECUADOR Manabí Portoviejo Crucita/La sequita Dec-95 G. helveolus ECUADOR Manabí Puerto Cayo Cantagallo Dec-95 Euseius ho ECUADOR Manabí Rocafuerte Entrada a Danzarín Dec-95 55 Climatic matching with Zambia So far, none of the irnported predatory mites ha ve beco me established in the East African plateau. We suspect that this is because this target region is too hostile to the strains that were imported to IIT A in recent years. Locations in Zarnbia where IIT A had collected CGM were used as the basis for defining the climate ofthe target region (Fig. 4.1.2. Long term climatic data was associated with each of these sample sites using the average of the four nearest clirnatic stations, weighted by (1/distance2). Principal component analysis was then used to develop a statistical m o del of the clima te in the target region (Fig. 4 . l . 3). This model was u sed in the GENST AT prograrn to develop a map of the Americas showing the degree of climatic similarity to the target release region (Fig. 4.1.4). r- Angola Democratfc Republlc ot Congo (Zaire} .A.ÁA A A A A ,-----~Á ... A A ... Zambla \ ... \.A. ... A A A A A A A A A A A Zlmbabwe Figure 4.1.2. Sample sites in Zambia that served as the target for generating principal component analysis of climate (N=52). 300 'E2so .S 200 e g 150 ca ... :a. 100 o ~ 50 o.. o 26 () 24 o -- 22 ~ ~ 20 L. Q) ~ 18 ~ 16 14 -20 () o --~ 15 e ca L. ~ 10 ::l -ca L. ~ 5 E Q) r- o r- (x ± sd) ! ! f ! f - . f f • .... .... • ~ ~ 1 2 3 4 5 6 7 8 9 10 11 12 f f f f ! f f f ! f ! ! 1 2 3 4 5 6 7 8 9 1 o 11 12 f f f f f I f f ¡ t t ! 1 2 3 4 5 6 7 8 9 10 11 12 Month of year Figure 4.1.3. Climate ofZambian target region, based on 52 sample sites (precipitation, temperature, diurnal temperature range). 56 Probability Density • High • • • Low D No similarity . .,. 11 • • . . Figure 4.1.4. · Map of climatic similarity to target sites in Zambia used to identify bigb priority regions in tbe Brazilian central bigblands and Mexico for predator exploration. 57 58 Suitability of different prey for phytoseiids Phytoseiid species that are candidates for exportation as biological control agents were evaluated for the specificity of prey and food sources that they can consume. This is to assure that we select species that will not harm non-target hosts after being imported to Afiica or Northeast Brazil. The ability to use pollen or sugary exudates of cassava leaves, mealybugs or whiteflies is an advantage in helping to maintain predator populatíons when cassava green mites are scarce. Galendromus annectens and G. helveolus showed much higher longevíty in the presence oftetranychid prey (Mononychellus caribbeanae or Tetranychus urticae) than for the other food sources tested (Fig. 4.1.5). Neoseiulus idaeus was similar except that it could also utilize pollen from castor bean (Ricinus communis). Typhlodromalus tenuiscutus, Euseius ho and T aripo could use a greater varíety offood sources, including cassava exudate (a sugary liquid found on leaves), immature whiteflies (Aleurotrachelus socialis) and mealybugs (Phenacoccus herreni); both of which produce sugary exuda te. 50,-----------------------------r=============~-----¡ 45 40 - U) >-35 ro "O -30 >. -·~ 25 C) e ..Q 20 ::S 15 "O <( 10 5 o • D D • 111 N. idaeus G. annectens G. helveolus T. tenuiscutus Predator species M. caribbeanae T. urticae Pollen Exudate Whitefly Mealybug E. ho T. aripo Figure 4.1.5. Effect of different food sources on female adult longevity of several species of predatory phytoseiid mites. 59 Predatory mites must consume protein in order to produce eggs, so fecundity is an important measure of the suitability of different food sources to increase predator populations. Typhlodromalus tenuiscutus showed the highest specificity, laying substantial numbers of eggs only when fed on cassava mites,Mononychellus caribbeanae (which has previously been shown to be equal to M tanajoa in preference and suitability). Neoseiulus idaeus, Galendromus annectens and G. helveolus also had high fecundity when provided with the two-spotted spider mite, Tetranychus urticae (Fig. 4.1.6). Euseius ho was the only species ofthis group that could produce many eggs when fed only castor bean pollen. Typhlodromalus tenuiscutus and Euseius ho could also reproduce at low rates when held on cassava leaves infested with various developmental stages of immature whiteflies (Aleurotrachelus socia/is) . 50 45 • M. caribbeanae ~ T. urticae 40 D Palien -35 D Exudate en C> 0>30 • Whitefly Q) -~25 111 Mealybug "' §2o (.) Q) u.. 15 10 5 o N. idaeus G. annectens G. helveolus T. tenuiscutus E. ho Predator species Figure 4.1.6. Effect of different food sources on fecundity of several species of predatory phytoseiid mites. 60 Exportation of natural ene ... ies Exportations of natural enemies during 1996 included Typhlodromalus tenuiscutus from a dry Iocation in coastal Ecuador (Portoviejo, Manabí) which was sent to CNP.MFIEMBRAPA, Cruz das Almas, Bahia via quarantine at CNPMA, Jaguariúna, Sao Pauto Brazil (Table 4.1.2). Strains oiTyphlodromalus manihoti from severa! high altitude sites in Colombia (Barbosa, Cajibio and Copacabana, Bucaramanga, Armenia, Chinchina and Santa Rosa de Cabal) were sent to liT A, Benin via quarantine at the University of Amsterdam (Mitox). Climatic parameters associated with the sites from which the predators were collected were estimated from the average long-term data for the four nearest climate stations, weighted by the inverse ofthe distance squared (Table 4.1.3). To date, three species ofphytoseiids have been established in Afiica. Ofthese, T. aripo and T. manihoti are spreading rapidly and are reducing cassava green mite populations significantly. Typhlodromalus aripo is estimated to be increasing root yield by 3 0%, which is equal to about US $70 per season. Estimated benefits for 4 countries in West Afiica are on the order ofUS $60 million per year. Two species released in Northeast Brazil (T. tenuiscutus & N californicus) have been recovered in small numbers, the latter up to 4 months after release. Typhlodromalus manihoti and cassava green mite (CGM) were also sent to the University of Amsterdam, and CGM was sent to Boyce Thompson Institute, Ithaca, New York to establish colonies in support of collaborative basic research. 61 Table 4.1.2. Cultures of predatory pbytoseiid mites exported from CIA T in 1996. Date Colony Number Number o/o Sent Species Source Establisbed Sent Received Destination Survival 3/11/96 T. tenuiscutus Portoviejo, 12/95 400 + 184 + CNPMA, 46% Ecuador imms. 42 eggs Brazil1 4/23/96 T. manihoti Barbosa, l/96 910 432 Amsterdam2 47% Colombia T. manihoti Cajibio, 8/95 240 219 Amsterdam 91% Colombia 5/14/96 T. manihoti Guajira, 3/96 580 375 Amsterdam 65% Colombia T. manihoti Copacabana, 1/96 1,025 793 Amsterdam 77% Colombia 12/6/96 T. manihoti Bucaramanga, 8/96 480 Amsterdam ca. 60% Colombia (Los Colorados) " , Armenia, S/96 300 Amsterdam ca. 60% Colombia Chinchina, 8/96 120 Amsterdam ca. 60% Colombia , Sta Rosa de 8/96 180 Amsterdam ca. 60% Cabal, Colombia " , Bucaramanga, 8/96 180 Amsterdam ca. 60% Colombia (Bijagual) Sent to CNPMA quaranúne laboratory in Jaguariuna, Sao Paulo, for futher shipment to CNPMF, Bahia, Brazil. 2 Sent to quaranúne laboratory in Amsterdam, Holland, for futher shipment to liT A, Benin. 62 Table 4.1.3. Clima te associated with cultures of predatory phytoseiid mites exported from CIA T in 1996. Dry Montbs RB Precipitation Temperature --~~i~---·· Source ~titu~~-{1!!1. ___ Q!m) _____ eg_ ____ ~~~- Ye~~----~) ____ T. tenuiscutus Portoviejo, Manabí, 44 575 25 8 79 Ecuador T. manihoti Barbosa, Antioquia, 1399 2362 21 o 82 Colombia T. manihoti Cajibio, Cauca, 1763 2152 18 1 80 Colombia T. manihoti Villanueva, Guajira, 340 1043 27 5 74 Colombia T. manihoti Copacabana, 1844 1649 19 74 Antioquia, Colombia T. manihoti Bucaramanga, 983 1363 22 o 85 Santander, Colombia, (Los Colorados) " Armenia, Quindio, 1500 2430 19 o 79 Colombia " Chinchina, Caldas, 1360 2514 21 o 76 Colombia " Sta Rosa de Cabal, 1580 2558 19 o 84 Risaralda, Colombia " Bucaramanga, 977 1363 22 84 Santander, Colombia (Bijagual) Months with less than 60 mm precipitation. Clima tic distribution of phytoseiids Explorations for phytoseiids have been conducted in 14 countries in Latin America and the Caribbean with samples from over 1,097 sites. Over 87 species ofphytoseiids have been encountered and over 25 new species have been described. About 67% ofthe samples carne from cassava and 76% ofthe 2,943 phytoseiid records are also from cassava. Colombia, Ecuador and Venezuela account for 89% of these records, comprising at least 77 species of phytoseiids, with at least 44 in association with cassava. Because of the high number of samples and the geographic 63 diversity of this region we chose to use it for ecological analysis. The clima te of each sample site was estimated using long-term climatic data from the CIAT GIS (geographic information system) database (average of 4 nearest climate stations weighted by distance-2). Duration of dry season (no. ofmonths with <60 mm precipitation) and elevation (m above mean sea level) were considered the most useful climatic parameters available for classifying the sites. This choice was based on availability of information and usefulness for evaluating species to release in target climatic zones. Data are reported as incidence (number of samples of a climatic classification in which the species was present divided by the total of such samples in that classification) to adjust for sampling effort (Table 4.1.3). A higher number of species were encountered in Colombia than expected based on sample effort, which probably reflects the high variation in topography and climate found within the country. In general, we have few samples from high altitude dry sites, which limited the climatic analysis. The phytoseiid species found in most environments (wet and dry, high and low) were Typhiodromalus manihoti and T aripo (Table 4.1.4). Species found more commonly in dry regions were Euseius ho, Gaiendromus annectens, Typhiodromalus rapax, T tenuiscutus, and Neoseiulus idaeus. Species found at intermediate e1evations were Galendromus annectens, G. heiveolus, and Euseius concordis, while Typhiodromalus neotunus and Euseius naindaimei occurred at high elevations. Two species appeared geographically isolated: Typhiodromaius tenuiscutus and Neoseiulus ca/ifornicus, and they were strongly recommended for release in Northeast Brazil, where they do not occur. These are the species that were released there in 1995 and 1996. 64 Table 4.1.4. Distribution of phytoseiid species with respect to elevation and number of dry months based on long-term climatic data associated with coUection sites in Colombia, Venezuela and Ecuador. No. of Dry Montbs Elevation above sea level (meten ) Typhlodrornalus manihoti < 800 800-1200 1200-1600 > 1600 <3 0.54 0.25 0.25 0 .29 3-5 0.35 0.25 0.25 0.23 > 5 0.24 Neoseiulus anonyrnus < 800 800-1200 1200-1600 >1600 !-~ 1 0.05 0.15 0.15 0.07 0.06 0.13 0.13 0.06 0.07 Galendromus annectens < 800 800-1200 1200-1600 > 1600 !-~ 1 0.02 0.10 0 .10 0 .03 0.04 0.05 0.05 0.09 Typhlodrornalus aripo < 800 800-1200 1200-1600 > 1600 ~~ ~ 0.03 0.06 0.06 0.08 0.05 0.05 0.05 0.06 0.05 Galendromus he1veo1us < 800 800-1200 1200-1600 > 1600 m 0.10 0 .10 0.08 0.03 0.08 0.08 0.03 0.05 Neoseiulus idaeus < 800 800-1200 1200-1600 > 1600 ~~ 1 0.04 0.01 Euseius ho < 3 < 800 800- 1200 1200-1600 > 1600 d 0.04 0.01 0 .01 0.05 0.01 0.03 0.03 0.03 0.16 TOTAL SAMPLE Number of sites in each classification cell < 800 800-1200 1200-1600 >1600 Total !-~ 1 11 1 410 86 32 639 728 75 35 4 842 245 o 1 247 Total 1084 485 122 37 1728 65 Pbytoseiid associations witb cassava The previously described database of exploration records for phytoseiid mites was analyzed to leam more about the host plant associations ofphytoseüds. About 76% ofthe 2,943 phytoseüd records are from cassava, so relatively little can be said about associations with plant species other than cassava. Colombia, Ecuador an9 Venezuela account for 89% of these records, and were selected to restrict the geographic range for this analysis. In this subset, 81 .5% of the records are from cassava. The 2,084 records were classified as coming from either cassava or "other" host plant and analyzed by chi-square. Most ofthe host plants had only 1-3 records per phytoseüd species, so it is not worth looking at more than the aggregated "other" plant category. The 17 most frequently encountered phytoseiid species are listed in Table 4.1.5. Ifwe want to use phytoseüd species that have a close association with cassava, we should focus on the species in the upper part of the table. Typhlodromalus tenuisculus, N idaeus, Typhlodromips dentilis, T. manihoti, and G. helveolus had high associations with cassava. With respect to phytoseiids targeted for release in NE Brazil, of the recently collected species from the dry coast of Ecuador, this includes: T. tenuiscutus, N idaeus, G. helveolus and G. annectens. This strong association with the target plant favors the choice of these predators to introduce for biological control of cassava green mite. Euseius concordis, Amblyseius chiapensis, E. ho, and /phiseiodes zuluagai were negatively associated with cassava, which suggests that they may not be good candidates. lphiseiodes zuluagai hada high association with citrus, and this species has been reported on citrus in other countries such as Brazil. Table 4.1.5. Association of the 17 most common phytoseiid mites with cassava or other plants based on explorations in Colombia, Venezuela and Ecuador. Phytoseüd species Proportion of field samples (%) Cassava Other plants Number of samples Typhlodromalus tenuiscutus 91 3 • 64 Neoseiulus idaeus 96 4* 53 Typhlodromips dentilis 96 4* 75 Typhlodromalus manihoti 94* 6* 650 Galendromus helveolus 91 9* 102 Galendromus annectens 88 12 116 Neoseiulus anonymus 86 14 181 Typhlodromalus aripo 83 17 124 Proprioseiopsis cannaensis 82 18 57 Typhlodromalus peregrinus 82 18 49 Amblyseius aerialis 79 21 106 Typhlodromalus rapax 16 24 78 Phytoseiulus macropilis 10 30 43 Euseius concordis 59* 41* 132 Amblyseius chiapensis 51* 49* 47 Euseius ho 50* 50* 129 Jphiseiodes zuluagai 31* 69* 78 66 • Values significantly d.ifferent than expected (based on marginal values, i.e. 81.5% in cassava), indicated by the chi- square standardized deviation for the cell being greater than 2 or less than -2. Evaluation of Neozygites Natural outbreaks ofthe fungal pathogen Neozygites c.fjloridana in Northeast Brazil have been observed to rapidly kili off populations of cassava green rnite (CGM, M tanajoa). Although subsequent investigations by ITT A scientists have shown that apparently the same species occurs in Africa. The lack of virulent epizootics suggests that it would be worthwhile to introduce a more virulent strain from South America. Pathogenicity of Neozygites strains Cultures of the fungal pathogen Neozygites c. f floridana were collected from the cassava green rnite and two-spotted spider rnite (Tetranychus urticae) at CIAT and from CGM at Media Luna, Colombia. Other cultures obtained from Brazil and Africa were included in a laboratory evaluation ofhost specificity on three rnite species: M . tanajoa, M. caribbeanae and T. urticae. Single mi te mumnúes were placed in the center of a cassava leaf disk under conditions of high hurnidity (95%) and 16 hours of darkness to produce conidia. Twenty-five recen ti y emerged 67 female mites were placed on the disk to be exposed to the conidia for 2 d at 65% RH, 12 h photoperiod. Afterwards they were transferred to fresh leaf disks and were examined daily for signs ofinfection and mortality. None ofthe 3 strains fromM tanajoa infected T. urticae, and one ofthe strains from T. urticae failed to infectM tanajoa (Fig. 4.1.7). AH 5 strains infectedM caribbeanae. This confirms that ther~ _is substantial difference among the strains and that those isolated from Mononychellus appear to be more specific than those from T. urticae. Geographical distribution of Neozygites This pathogen is currently being studied for use as a biological control agent of CGM, but little is known about its geographic distribution or host range. We reviewed 10,782 specimens of tetranychid mites on microscope slides from our museum collection, representing 907 sample sites, to determine the host range and geographic range of this pathogen. These specimens were previously collected during foreign exploration for natural enemies of M tanajoa. Both conidia and hyphal bodies can be detected in the slides. Signs of Neozygites infection (presence ofhyphal bodies) was observed in 14 species oftetranychid mites from 11 neotropical, 2 African and 2 Asían countries, indicating that this pathogen may be much more important than was previously thought (Tables 4.1.6 and 4.1. 7). Neozygites infection was also observed in Tetranychus kanzawai specimens from China and Indonesia. The highest frequency of infection, as indicated by the presence ofhyphal bodies, was found in M caribbeanae andM tanajoa (Fig. 4.1.8). Significantly low frequencies occurred in 0/igonychus peruvianus, which is common on cassava but which is generally protected beneath silk webbing, and Tetranychus sp. (which is probably mainly T. urticae, which cannot be identified to species without males). Although it is not certain that only one species of Neozygites is represented by these results, they are in agreement with the previously described laboratory pathogenicity tests for N c. f jloridana indicating that Mononychellus is very susceptible compared to other tetranychids. 100,-------------------------------------------------------, 90 _ Mononychellus caribbeanae u ~ 80- ~ 70- U) aJ 60- ::J u ·s; :0 e: u Q) ~ U) m ::J u ·s; :0 e: -o ~ o u ~ ~ U) m ::J u ·s; '6 e: - o ~ o 50- 40- 80- 70- 60- 50- 40- 30 - 20- 10- 10(D 90 80- 70- 60- 50- 40- 30- 20 - 10- o Mononychellus tanajoa T T ···J···· .... ... . ·.:.:¡:¡¡¡::¡¡¡ 1 1 mtaf mtcd Tetranychus urticae 1 mtaf mtcd ........ ........ ........ .... .... . ...... . . ... ...... ········· .. .... ... ......... .. .... ... ......... . ........ II\jJ mtml mtml 1 Strain of Neozygites (x ±95%CI) T tuaf tuct T tuaf tuct T ••• • o • • • .... . ... 1 · ··:¡~¡· ··· ....... . f}f¡{ ::::::::::::::::: ::::::::::::::::: • tuaf tuct 68 Figure 4.1. 7. Pathogenicity of five strains of Neozygites c.f. jloridana on 3 species of tetranychid mites associated with cassava (mt- from M tanajoa; tu- from T. urticae; af- from Benin, Africa; cd - from Cruz das Almas, Babia, Brazil; mi - from Media Luna, Magdalena, Colombia; ct- from CIAT, Valle de Cauca, Colombia). Table 4.1.6. Species of Tetranychid mites held in the CIA T collection that show signs of infection b Neo •tes s . as indicated b the resence of h hal bodies. ___ __;S;.¡;p'""ec-'-i~s Samples Species Samples -·-- Eutetranychus banksi 1 Tetranychus canadensis 1 Mononychellus caribbeanae 38 Tetranychus cinnabarinus 1 Mononychellus mcgregori 5 Tetranychus desertorum 1 Mononychellus planki 1 Tetranychus marianae 3 Mononychellus tanajoa 43 · Tetranychus mexicanus 0/igonychus gossypii 5 Tetranychus tumidus 4 0/igonychus peruvianus 4 Tetranychus urticae 1 Tetran chus s . 6 Total 115 Table 4.1.7. Countries from which specimens ofTetranychid mites held in the CIAT collection show signs of infection by Neozygites sp. as indicated by the presence of hyphal bodies. ·······-·-···-·-..f.~l!~~-!).' ___________________________ §~P.~.~~----·--- ····-·--··········---~~~-~-~9'---·········--············ ··· ········§-~P.~.~~---···· ·········· Neotropics Africa Brazil 19 Benin 3 Colombia 38 Mozambique 1 Cuba 12 Ecuador 11 Asia Mexico 10 China 3 Nicaragua 1 Indonesia Panama 3 Paraguay 1 Peru 3 Trinidad 4 Venezuela 14 Total 124 69 .-.10 ~ o - "O C1) 8 u 6 .!! e e 4 o t o 2 a. e a. o (1) ~ (1) :::! (1) q)llJ .e;::! "5 ·& ~fi¡ ;:,...lll e: ·-e: e: o:::. o lll o,2 e:~ :.::::Q) ~ OCl. + (1) (1) :::! q) :::! :::::: lll .e: q) t:: ~ - .e:: lll (.) q) e: ~ :>..Q ~ §:9 e: >..:;: ~ o lll ~ (.) 70 700 600 500 (/) C1) a. 400 E ro (/) 300 -o ..... C1) 200 ~ E ::S e 100 ro - o o ._ ci (1) (1) (1) (1) (1) ~ :::! (1) :::! ·- .e:: (1) .e:: (1) 2~ ~:§ :::! (.) e: .e:: :::...q) ~e: (.) e: "O ~ § :::... ~ ~ e:~ e: o Cl. q)~ o ..... lll e: f.:. g> ~ (.) ~ o Figure 4.1.8. Proportion of samples containing individuals infected by Neozygites sp. (as indicated by the presence of hyphal bodies) in the 11 most common species of tetranychid mites in the CIA T collection (+ , higher than expected; - , lower than expected; chi-square test with standardized deviations greater than 2). DNA characterization of Neozygites Strains of Neozygites c.fjloridana have been shown to differ substantially in their pathogenicity (see above); however, they can not be distinguished by morphology. Field research conducted by liT A in Benin and CNPMF in Brazil indicate that strains from Northeast Brazil are more virulent than those in Africa. However, it is critica! to develop methods to identify strains of Neozygites before neotropical strains ofthe fungus can be released in Africa. Characterization ofDNA using PCR-based techniques should be able to permit the identification of individual strains; however, this has never been done with this genus of fungus. Despite the obstacles that are normally encountered in such pioneering research, substantial progress has been made. In vitro culturing methods developed by Don Roberts (Boyce Thompson lnstitute, USA) and Luis Leite (Instituto Biologico, Brazil) have been successfully 71 adopted and modified. We can now grow strains of the fungus from both T. urticae and M tanajoa (Roberts & Leite could only grow a T. urticae strain) on artificial media (Grace's Insect medium + yeast extract + lactoalbumin). Problems with bacterial contamination have been resolved by adding antibiotics to the media (Fig. 4.1.9). This has removed a major obstacle to our work, now permitting us to produce ~~fficient quantities ofuncontaminated DNA. Nevertheless, the growth rate of strains isolated from M tanajoa still grow very slowly (Fig. 4 . 1.1 0). Because the Neozygites strains isolated from M tanajoa appear to be more specific in host range than those from T. urticae, this suggests that growth rates may be improved by further refinement of the growth media, perhaps by induding compounds found in cassava leaves. Futher work will be done in this area. Five strains of Neozygites c. fj/oridana were multiplied in vitro for DNA extraction. Preliminary evaluations using RAPD (random amplified polymorphic DNA) and AFLP (amplified fragment length polymorphism) techniques have been performed. The initial RAPD trials using 5 primers (A2, A3, B2, B9 & D2; Operon Technologist, La Jolla, CA) were able to distinguish Neozygites c.f floridana from a distantly related fungus (Phaeoisariopsis griseo/a, bean angular spot), but not among T urticae strains of Neozygites. The AFLP method generally provides many more bands (DNA fragments), thus increasing the power to detect differences. An unreplicated AFLP trial showed strong differences among 5 Neozygites strains, while the pattems for different samples from a single batch of one strain were clearly identical. Confirmation of these results will mean that we can readily identifY individual strains of this pathogen, thus opening the way to make releases in Africa. 72 160 (X± SE) 140 Strain T. urtícaeCIAT1 • Antibiotics, TS :::::--120 E ~ No antibiotics, T '-~100 1 Antibiotics, TS o --{}- No antibiotics, T ~ _. 80 en Q) -g 60 .o co .c. 40 a. ~ I 20 o 1 9 16 23 Days Figure 4.1.9. In vitro multiplication of a Neozygites cf.jloridana strain isolated from T. urticae (TuCIATl) with and without antibiotics (streptomycine + tetracycline) in two difTerent tissue culture media (T, TNMFH + lactalbumin + yeastolate; TS, TNMFH + lactalbumine + yeastolate + fetal bovine serum). 160 Strain M. tanajoa CIAT1 (x ±SE) 140 ::::-120 E • Antibiotics, TS --<>---- No antibiotics, T '-M~100 • Antibiotics, TS o --o-- No antibiotics, T ""'""" 80 ..._ en (].) "C 60 o .e ca 40 .e c.. >-I 20 o o 1 9 16 23 Days Figure 4.1.10. In vitro multiplication of a Neozygites cf. jloridana strain isolated from M tanajoa (MtCIATl) with and without antibiotics (streptomycine + tetracycline) in two different tissue culture media (T, TNMFH + lactalbumin + yeastolate; TS, TNMFH + lactalbumine + yeastolate + fetal bovine serum). Collaborative research on Neozygites 73 Strains of Neozygites c.f jloridana (Zygomycetes: Entomophthorales) and cultures of cassava green mites were sent to Boyce Thompson Institute (BTI), Ithaca, New York several times to support collaborative strategic research. Don Roberts and Luis Leite (a Brazilian graduate student) successfully developed an in vitro method ofrearing a Tetranychus urticae strain of Neozygites. This was critical to help CIAT scientists obtain sufficient quantities of uncontaminated DNA to allow them to start using molecular methods to characterize strains of the fungus. The BTI scientists also investigated the process ofinvasion ofhost mites by the fungus. This strategic research should help us understand the differences in host specificity of different strains of this pathogen. Samples of Neozygites c. fjloridana were sent to Siegfried Keller at the Swiss Federal Res. 74 Station for Agronomy, Zurich for taxonomic identification. So far, none ofthe strains from Africa, Brazil or Colombia have shown any morphological characters that distinguish them. This emphasizes the need to develop molecular methods to identify strains. A strain of Neozygites sp. from T. urticae was released to Biocontrol, SA and Coinbiol, SA, in Palmira, Colombia for possible development as a commercial biological control agent. Phytoseiid mass rearing techniques Mass rearing of Typhlodromalus tenuiscutus Experiments were conducted to help optimize the use of the Mesa-Bellotti method of rearing phytoseiid predators to assist mass-rearing of predators in NE Brazil. This system consists of using small plastic containers with a screen shelf Fresh mite-infested leaves are placed in the bottom ofthe container to feed the phytoseiid predators. After 2-3 days additional fresh, rnite- infested lea ves are added on top of the shelf The predators then move from the old lea ves to the new leaves, which are removed in 2-3 days, when the next batch offresh mite-infested leaves are added. This system was designed specifically for predators that require Mononychellus prey, which must be grown on cassava leaves. This particularly applies to Typhlodromalus tenuiscutus which we are now releasing in NE Brazil. The experiment on colony production was initiated with 50, 100 and 200 females in Mesa-Bellotti trays, and adult females were harvested weekly for 8 weeks, starting at week 3. An experiment was conducted to measure the effect of initial number of T. tenuiscutus females used to establish Mesa-Bellotti mass-rearing cultures on the number of adult female progeny harvested. After the third week of rearing, all adult females were counted and removed from the rearing system. It can be seen that this method results in a continued decrease in production of the culture, which suggests that not enough females are retained by this harvesting method to sustain levels of oviposition (Fig. 4. l . 11 . ). The highest production of females from Mesa-Bellotti cultures was from those initiated with 1 00 females, the least from those started with 50 females 75 (Fig. 4.1.12a). Those initiated with 200 females fluctuated from week to week the most, suggesting possible crowding effects. Production of adult female T. tenuiscutus in the Mesa- Bellotti mass-rearing system per number of females used to initiate the culture was highest at the lowest density tested (50 females; Fig. 4 .1.12b) suggesting the effect of competitive interactions at the higher densities. 600 "'C 500 Q) ....... en ~ 400 ro .e ~ 300 ro E Q) 200 u.. 100 o 3 4 5 6 7 initial no. of females • 50 A 100 D 200 8 9 Weeks sin ce initiation of the culture Figure 4.1.11. Effect of initial number of T. tenuiscutus females used to establish Mesa- Bellotti mass-rearing cultures on the number of adult female progeny harvested. 500 ~400 r- o Q) =300 8 (/J ~200·r­ E tf 100 (X± 95% IC) 8 "'C Q) ~ 6 t) 60%) is removed in the harvested roots . 101 Table 4.5.1 Effect of NPK rates on dry root yield and dry plant top (t/ha) of two cassava varieties during the 1995-1996 season (13th cycle) at Santander de Quilichao, Cauca, Colombial. Fertilizer treatment (kglba) M Col1684 CM 91-3 N p K Dry root Dry top Dry root Dry top o o o - 4.4 2.0 3.9 2.3 50 50 50 9.1 3.5 7.8 3.9 o lOO lOO 10.4 3.3 5.2 2.9 50 100 lOO 10.1 4.4 7.8 3.8 lOO lOO lOO 10.5 4.3 9.2 3.9 lOO o lOO 9.8 4.0 8.9 4.-l lOO 50 lOO 13.0 4.2 8A 3.8 lOO lOO o 7.5 3.0 5.2 3.1 lOO lOO 50 10.3 4.0 7.6 4.1 LSD 5% (Tukey) for comparison between NPK treatments for dry root 3.75, for dry top 1.57. Long-term effects of surface mulch, fertilizer and tillage on cassava productivity in poor sandy soil in northern Colombia Table 4.5.2 presents data on root yield and plant top biomass as affected by plant surface mulch ( dry grass ), NPK fertilizer and type of tillage on sandy soils at Media Luna, northern Colombia. Both dry root yield and top biomass were significantly increased by the application of either mulch or NPK fertilizer at a moderate leve!. On the other hand, conventional tillage increased yield only with NPK application in absence of mulch. The combination of surface mulch and zero tillage gave the highest root yield with and without fertilizer. These findings indicate that consecutive cultivation of cassava in these poor soils would lead to very low yields unless appropriate measures are followed to maintain soil fertility . The choice between application ofNPK fertilizer or surface plant mulch would depend on various factors related to the socio-economic conditions of the local farmers . However, the potential economic return due to fertilization is high in this case. Moreover, the application of fert il izer would ensure enough quantity of planting material with high quality. 102 Table 4.5.2 Dry root yield and dry plant top (tlba) of cassava (cv. M Col 1505) at Media Luna, Magdalena, Colombia during the season 1995-1996 (8th cropping e ele. Treatment Conventional tillage Conventional tillage + muJch2 Zero tillage Root 4.3bc3 5.9ab 3.3c Fertilized 1 Top 2.4b 3.6ba 2.3b Zero tillage + mulch 6. 9a 4.2a 1 Fertilized with 330 kg/ha 15-15-15 NPK (50 N; 21.6 P; 41 .7 K kglha). 2 Mulch, 12 tlha dry grasses applied annually. Unfertilized Root Top l.2b l.4b 4.6a 2.7a l.4b 1.2b 4.9a 2.8a 3 Values in columns followed by the same letters are not significantly different at 5% by Duncan' s test. Table 4.5.3 contains data on the response of cassava to both plant rnulch and NPK fertilizer in the soils of Santander de Quilichao. Although these soils contain greater organic rnatter than the sandy soils of northern Colombia, rnulch application increased root yield even in the first year of the tri al. Again, these findings indicate the beneficia! effect of plant residues as rnulch. F ertilizer application also increased root yield in these soils which have been under cassava cultivation for several years. Both plant top biornass and root dry matter content were also enhanced by applications of surface mulch and NPK fertilizer. When native weeds were repeatedly cut, without uprooting, and used for mulch, yields were depressed in absence ofNPK fertilizer. This finding suggests that weeds significantly compete for nutrients in this case. Previous research (Cassava Program Annual Reports, 1987- I 992) indicated that using post-emergence herbicides at an advanced stage of weed growth enhanced cassava productivity as compared with no weed mulch. 103 Table 4.5.3 Dry root yield and dry top (tlha) and root dry matter (%) (cv. CM 523-7 and CM 507-37) at Santander de Quilichao, Cauca, Colombia. First season {1995-1996). CM 523-7 CM 507-37 Unfcrtilizcd Fertilizcd 1 ---·· ------- Root dry Root dry Treatment Root Top matter (% Root Top matter (% ) ---· -~---- Without mulch 9.7 3.9 40.9 12.2 5.3 40.8 Brachiaria mulch 2 JI. O 5.7 41.3 13.8 5.5 41.1 Weed mulch 3 7.1 2.8 40.7 13.4 5.1 41.3 ------- Mean 9.3 4.1 41.0 13.1 5.3 41.1 ------- ·- -· ·- - -- -· ---- -··---·- ---- ---- LSD 5% (Tukcy) 2.65 1.06 2.91 1 Fertilized with 500 kg/ha 10-20-20 NPK (50 N; 44 P; 83 K kg!ha). 2 Mulch of Brachiaria decumbens, 12 l/ha dry applied annually. Root 10.3 12.6 8.3 10.4 3 Mulch of di)' weed, 4.44 l/ha; top biomass of in situ weeds were used through repeated cuttings. Unfcrtilized Root dry Top matter (%) Root - 3.9 34.8 13.9 4.4 36.8 16.3 ' 2.8 33.4 13.3 3.7 35.0 14.5 -- ------ --·-- - ·-- - -- Fertilized Root dry Top matter (%) 5.7 36.9 7.5 38.2 5.1 36.3 6.1 37.1 104 Effects of quality of planting material and water stress on cassava productivity Figures 4.5. 1, 4.5.2 and 4 .5.3, present data on leafphotosynthesis, stomatal conductance and midday leaf water potential as affected by stake quality and water stress. In this tri al, planting material ofM Col 1684 were obtained from the long-term NPK trial at Quilichao on the 12th cropping cycle. Mother plants with the following NPK levels were used: O, O, O (TI); 50, 50, 50 (T2); 100, 100, 100 (T3); and 100, 100, O (T4) kg/ha N, P, K. The trial was conducted in the field lysimeter at Quilichao, where half of the experimental area was deprived of water at 65 days of planting by covering the ground with plastic sheets for 3 months. Leaf net photosynthetic rate (Fig. 4.5.1) was similar in both the control and the water-stressed crops, irrespective ofthe origin of the planting material. Apparently, the stake quality did not affect leaf photosynthesis. Stomatal conductance was significantly reduced by water stress in all stake origins. However, this effect was observed only at 60 days after stress was initiated (Fig. 4.5.2). The same trend was observed in midday leafwater potential (Fig. 4.5.3). These data indicate that cassava tolerates prolonged water stress, and its lea ves maintain a high level of photosynthesis during the water stress period. Moreover, cassava conserves water by partially closing its stomata during prolonged water stress. Evaluation of crop/soil management practices in farmers fields in northern Colombia In collaboration with CORPOICA, several field trials were established in April-May 1996 in various sites to evaluate the effects of the application of mulch and fertilizer on productivity of the cassava/maize intercropping system in farmers' fields . These trials are located in Los Palmitos, Dept. of Sucre; in Plato, Dept. of Magdalena; Pueblo Nueva, Dept. of Cordobá; Pivijay, Dept. of Magdalena; Chiriguana, Department ofCésar; and in El Carmen de Bolívar, Dept. ofBolivar. These trials will be harvested in Feb./March 1977. In addition to the research mentioned above, the Associate, Luis F. Cadavid, had participated in two training workshops for specialists of CORPOICA in regions 2 and 3. He prepared and dictated two courses on integrated crop/soil management. The first course was in Barranquilla, Atlantico in May 1996, and the second was in Turipana, Montería, Cordobá in August 1996. 35 M Col1664 (T1) 30 25 "' "' 'E 20 ~ u o 15 E 2: Cl) 10 ~ M Col1 684 (T4 ) u ~ 30 .e e: >- "' 25 o o .e Q. Qj 20 z 15 10 60 75 90 105 120 135 150 165 60 75 90 105 120 135 150 165 180 Days after planting Figure 4.5.1 Effects of the quality of planting material and water stress on leaf photosynthesis. Arrow, Initiation of stress. Stakes were obtained from mother plants on the l31h cycle of a long-term NPK trial that received annually the following treatments: T 1=0,0,0 NPK; T2=50,50,50 kg!ha NPK; T3;;:100,100,100, kg/ha NPK; T4=100,100,0 kg!ha NPK. (bars, ±sd). 105 -0.8 -0.9 -1 .0 -1 . 1 -1 .2 ñl -1 .3 a. ~ -1 .4 ñi -1 .5 e: IU o -0.8 a. .... -0 .9 IU (ij ~ -m -1 .0 IU -1 . 1 ....J -1.2 -1 .3 -1 .4 -1.5 -1.6 M Col 1684 (T,) M Co11684 (T3) _.._ Contro l -0- Stress M Col 1684 (T,) 60 75 90105120135150165180 75 90105120135150165180 Days after planting 106 Figure 4.5.2 Effects of the quality of planting material and water stress on leaf conductance. Arrow, initiation of stress. Stakes were obtained from mother plants on the l31h cycle of a long-term NPK trial that received annually the following treatments: T 1=0,0,0 NPK; T 2=50,50,50 kglha NPK; T3=100,100,100, kglha NPK; T4=100,100,0 kglha NPK. (bars, ±sd). .I/) N 1.2 0.9 0.6 .E 0.3 o E ~ O.O ~---------------------4------M __ C_oi_1_68_4_(_T~2) ______ ~ e ro t> :J -g 0.9 o u (ij Q,) ...J 0.6 0.3 M Col 1684 (T3) _.,_ Control -0- Stress M Col1684 (T4 ) o. o -+-----r----r----r--.-----.---..---.---+--.....---,----.---.----.---.------r-----1 60 75 90 105 120 135 150 165 60 75 90 105 120 135 150 165180 Days after planting Figure 4.5.3 Effects of the quality of planting material and water stress on midday leaf water potencial. Arrow, initiation of stress. Stakes were obtained from mother plants on the 13'h cycle of a long-term NPK trial that received annually the following treatments: T 1=0,0,0 NPK; T2=50,50,50 kg!ha NPK; T3=100,100,100, kglha NPK; T4=100,100,0 kg!ha NPK. (bars, ±sd). 107 5 STRA TEGIC RESEARCH AT CNPMF 5.1 Biological Control of the Cassava Green Mite, by Aloyseia Cristina S. Noronha lntroduction of phytoseiid mites 108 Individuals of Typhlodromalus tenuiscutus McMurtry & Moraes were collected in Puerto Cayo, Manabi Ecuador anº sent from CIA T to the "Costa Lima" Quarantine Laboratory, El\1BRAP A/CNPMA, municipality of Jaguariúna, State of Sao Paulo. During 1996, five shipments of T tenuiscutus were made from EMBRAP A/CNPMA to EMBRAPA/CNPMF anda total of2,150 individuals were sent, 75.48% ofthem arrived alive. As stated in PROFISMA' s Annual Report 1995, T tenuiscutus shipments from CIA T during 1996 were with recently established colonies. Maintenance rearing Using an adaptation ofMcMutry & Scriven's phytoseiid rearing method, it was possible to obtain a monthly average production of about 2,500 individuals of N californicus per tray onjack bean (Canavalia ensiformis) detached leaves, under laboratory conditions, at EMBRAP A/CNPMF (Table 5. l . 1 ) . Mass rearing T. tenuiscutus rearing was performed under laboratory conditions on M tanajoa on detached cassava leaves. Mass rearing of N californicus was carried out in a screenhouse. Individuals were fed on Tetranychus urticae on jack bean plants. Such a methodology enabled the production of33 ,500 individuals of N californicus starting from a single colony, in a one month period. Release and monitoring of phytoseiid mites Phytoseiid mi tes were released in cassava fields located in the S tates of Bahía and Pernambuco, as shown in Table 5.1.2. Releases were performed late aftemoon, by attaching phytoseiid infested cassava lea ves to cassava branches. A total of 10,813 individuals of T. tenuiscutus and 34,020 individuals of N californicus were released during 1996. 109 Table 5.1.1. Maintenance rearing of Neoseiulus californicus, on detached leaves of Canavalia ensiformis, under laboratory conditions. Number Female Evaluation Colonies Initial N° Date Larva e N!m~h+male Female Total 01 100 16/Sept/96 215 973 970 2,518 16/0ct/96 83 490 1,006 1,579 - 15/Nov/96 340 1,205 2,021 3,566 02 100 18/Sept/96 178 1,064 882 2, 124 16/0ct/96 151 1,103 1,536 2,790 15/Nov/96 11 1 960 1,819 2,890 03 100 25/0ct/96 12 1,203 1,960 3,175 27/Nov/96 168 885 1,399 2,452 Table 5.1.2. Field releases of phytoseiid mites. S~ecies S tate Munici(!ality Field lndividuals Relea ses r tenuiscutus Bahia Juacema Juacema 600 1 Jaguari Jaguari 10,213 50 Total 10,813 N californicus Bahia C. Almas EAUFBA 3,000 1 lchu Ichu 8,070 2 Jaguari Jaguari 19,950 Pernambuco Petrolina Ad 3,000 4 Total 34,020 Monthly monitoring recovered 1, 085 mi tes from all cassava fields where releases were performed. One hundred and forty-two slides were examined under microscope enabling preliminary identification of28 phytoseiid mites, resembling N californicus; slides were sent to Dr. G. de Moraes to confirm the identifications. Effect of neem, Azadirachta indica A. Juss, on mites This study has been carried out aimed at developing an alternative method for the cassava green mite control. Preliminary results showed a decrease on survival of r urticae fed on 110 jack bean Ieaf discs previously immersed in neem extract (Table 5.1.3). On the other hand, N californicus survival ability was not seriously affected by neem extract (Table 5.1.4). Although preliminary, these results are very promising. Studies on the effect ofneem M tanajoa are under way at the Entomology Laboratory, EMBRAPA/CNPMF. Table 5.1.3. Survival ability of Tetranychus urticae on jack bean leaf discs immersed in neem aqueous ·extract. . ............ ~~·ry·~~~~ .. ~.~-~~.~.9' .. e:.'!l ........................ Treatment Neem treated leaf disclimmediately after immersion Neem treated leaf disc/1 h after immersion Distilled water treated leaf disc/immediately Distilled water treated Ieaf disc/1 h after immersion 24 h 48 h 55 18 28 90 90 24 86 86 Table 5.1.4. Survival ability of Neoseuilus ca/ifornicus on eggs of Tetranychus urticae/ j ack bean leaf discs immersed in neem aqueous extract. Treatment Neem treated leaf disc + T urticae eggs Neem treated leaf discs + T urticae eggs (1 h later) Distilled water treated leaf di se + T urticae eggs Distilled water treated leaf disc + T urticae eggs (1 h later) Neem treated leaf discs Neem treated leaf di ses ( 1 h later) Distilled water treated leaf disc Distilled water treated leaf di se ( 1 h later) Survival ability (%) 24 h 48 h lOO 70 80 60 100 80 lOO 90 70 80 80 80 100 lOO lOO lOO In vitro selection and production of Neozygites sp. by Italo Dellalibera Jr. Studies were carried out aiming at selecting the most efficient isolates of Neozygites sp. for controlling CGM in semiarid zones ofNortheast Brazil, as well as for sending to Africa. The bank of Neozygites sp. isolates contains currently 21 isolates from severa! regions; however nine of them lost their viability, i.e. they do not sporulate any more. Since the in vitro maintenance ofthis pathogen is not yet possible and there is very little information on 111 its storage, all the isolates have been multiplied in vivo (Table 5.1.5). Sorne isolates have been stored for up to 20 months and remain still viable (sporulating). The isolates from the municipalities oflpirá, Cruz das Almas, and Sao Miguel das Matas, State ofBahia, stored for 31, 28 and 28 months, respectively, are still viable, while those from Piritiba and Senhor do Bonfim, State ofBahia, and Caruaru and Exu, State ofPemambuco, lost their viability after storing for 29, 20, 6, and 20 months, respective! y. Although the cause of loosing the viability of sorne isolates is unknown, it seems that the quality ofthe collected mummy, and the period of time that the infected mite remains in the field before being collected may constitute determinant factors on the storage period. Table 5.1.5. Neozl_g_ites S(!. isolates held at CNPMF. Number lsolate Place of collection Collection date Last multiQlication BINO! Piritiba, Bahía May/1111993 Mar/30/1996 2 BIN04 Ipirá, Bahía Aug/27/ 1993 Mar/30/1996 3 BIN08 Cruz das Almas, Bahía Nov/05/1993 Mar/31/1996 4 BIN09 Sao Miguel das Matas, Bahía Nov/22/1993 Apr/03/1996 5 BIN10 Tianguá, Ceará Nov/11/1994 Apr/14/199 5 6 BIN11 Rio Real, Bahía Mar/14/1996 Apr/01 / 1996 7 BIN12 Cristinápolis, Sergipe Jun/ 14/1 994 F eb/1511996 8 BINI3 Porto Real do Colégio, Jun/ 114/1 994 Apr/0 1/1996 Alagoas 9 BIN14 Teotonio Vilela, Alagoas Jun/14/ 1 994 Apr/0 1/1996 10 BIN15 Vitória de Santo Antao, Jun/15111994 Apr/0 111996 Pernambuco 11 BIN1 6 Gravatá, Pernambuco Jun/1511994 F eb/2611996 12 BIN18 Caruaru, Pernambuco Jun/ 1511994 Apr/16/1995 13 BIN19 Croatá, Ceará Jun/18/ 1994 F eb/15/ 1996 14 BIN20 Itapagé, Ceará Jun/18/1 994 Mar/25/ 1996 15 BIN2 1 F1oriano, Piauí Jun/22/ 1994 Mar/26/ 1996 16 BIN22 Araripina, Ceará Jun/2311994 Mar/23/ 1996 17 BIN25 Itaberaba, Bahía Jun/29/ 1994 Apr/02/ 1996 18 BIN26 Lambari, Minas Gerais Apr/03/1996 19 BIN28 Mandiocaba, Paraná Oct/23 /1995 20 BIN29 Jaguariúna, Sao Pau1o* Mar/11 / 1996 21 BIN30 Cali, Colombia* *lsolate from T urticae. 112 The conidial production of six isolates multiplied in vivo, under laboratory conditions, in May, 199 5 and in March, 1996 was evaluated in November, 1996. The results showed that the isolates 20, 22 and 23 produced a higher number of conidia when stored up to 8 months than when the storage period was 18 months. On the other hand, no effect of storage period was observed in isolates 4, 15, and 21 (Table 5.1.6). Table 5.1.6. Conidial production of six Neozygites sp. isolates, stored for eight and eighteeo months at 4°C/5% RH (total number of conidia produced per single mummyfied M tanajoa, after 24 hours uoder dark cooditions aod 100% RH). Total conidia produced per CGM mummy1 Isolate 8 month storage period 18 month storage period 4 569±95 a 417±210 a 15 356±70.9 a 430±261 a 20 1284±231 a 132±229 b 21 393±195a 197±3 19 a 22 748±133 a 0±0 b 25 963±191a 261±312 b 1 Values followed the same letter, in the same column, do not differ significantly (P=0.05). Dueto indications of storage effect on Neozygites sp. isolates, a study was carried out in order to evaluate severa! storage conditions on the viability oftwo isolates ofthe pathogen, one from Piritiba and the other from Cruz das Almas. Treatments were as follows : -1 0°C, 4°C/5% relative humidity (RH), 4°C/21% RH, 4°C/37% RH, 25°C/SO% RH. Relative humidity levels of 5%, 21% and 50%, were obtained by using glycerol solutions, while 3 7% correspond to the RH inside a refrigerator. As shown in Figure 5. l . 1, both isolates lost their viability after one month of storage period at 25°C/50% RH. The isolate from Piritiba showed no significant difference related to sporulation after eight month storage period except for 4°C/21% RH storage conditions. On the other hand, the isolate from Cruz das Almas showed no effect of storage period on sporulation when stored under -1 0°C. It is interesting to call the attention for the fact that there was no decrease in sporulation ofboth isolates, stored under -1 0°C, after the eight-month storage period, thus indicating that 113 freezing mummyfied mites seems to be a very good storage condition, while under 25°C/50% RH the pathogen viability decreased rapidly. This is a very interesting observation since 25°C and 50% re1ative humidity are not uncommon environmentaJ conditions in cassava growing areas ofNortheast Brazil . 1500 ;>.., § 1200 ::S ~ 900 ·-~ ·a 600 o u 300 15oo T ;>.., 1200 - E 1 E ::S ~ e: o u a Oday a Oday O -100C 040C e 5% RH. • 40C e 37o/o R.H. O 4°C e 21o/o R.H. • 250C e 50% R. 7 days 1 rronth O -100C O 40C e 5o/o RH. • 40C e 37o/o R. H. O 40C e 21o/o R.H. • 250C e 50% R.H. 7 days 1 rmnth Isolate from Cruz das Almas a b 2 rronths 4 rronths 8 roonths Storage period Isolate from Piritiba a ab ab 2 rronths 4 rronths 8 roonths S torage period Figure 5.1.1. Viability of two Neozygites sp. isolates stored under several conditions. 114 Phenotypic characterization of Neozygites sp. isolates Since no information is available in the literature regarding to phenotypic characterization of Neozygites sp., basic research was conducted aiming at developing the methodology necessary to carry out that activity. Studies were initially directed to determine the effect ofrelative humidity on sporulation, and length oftime of conidial production of Neozygites sp. isolates. It was observed that the isolates from Cruz das Almas and Piritiba do not sporulate relative humidity lower than 95% at 24°C (Figure 5.1.2). Consequently , 95% RH was chosen to be used in further studies of selecting Neozygites sp. isolates. Exploratory experiments, regarding the selection of Neozygites sp. isolates, consisted in evaluating conidial production of 12 isolates at 24°C and l 00% RH, to be la ter compared with results obtained at 24°C and 95% or less RH. Although significant differences were detected among isolates, a high variation in conidial production by mummies of the same isolate was al so observed, as evidenced by the standard deviation (Table 5. l . 7). Such a variation may be due to the fact that the mummyfied mites had been stored for 18 months befo re using in the experiment. Isolates stored for shorter period of time will be used in further studies of selecting Neozygites sp. isolates . .----.2~ T r: lsolate from lpir.í E Í l lt 200 - 1 ' 1 12 h r~IIL { ~ 2~- lsolate from Cna cllu Almas O l8h 1.035 1,04 1,045 1.05 1.035 1,04 1,045 1,05 H2S04 '25<200 and 3= >200 miteslleaf. 118 5.2 BIOLOGICAL CONTROL OF THE CASSA VA MEALYBUG (CM), by José Mauricio Simoes Bento Mass rearing and releasing parasitoids of the cassava mealybug Apoanagyrus (=Epidiniocarsis) diversicornis, Acerophagus coccois and Aenasius vexans, all parasitoids ofthe cassava mealybug, Phenacoccus herreni, were introduced from CIAT, through the "Costa Lima" Quarantine Laboratory, EMBRAP A/CNPMA, to EMBRAP Al CNPMF where those parasitoids were mass-reared. A total of 111,610 parasitoids were produced, 35,950 ofwhich were released in three cassava fields in the State ofBahia and three cassava fields in the State ofPemambuco, showing high leve! ofincidence ofthe cassava mealybug (Table 5.2.1). Table 5.2.1. Number of cassava mealybug parasitoids produced under laboratory conditions, at EMBRAPA/CNPMF, and released in cassava fields located in the S tates of Babia and Pernambuco. ······ ······ ······· ······· ·-~-~2.~ ................................................... J.~.?.~ .......................................................... ~.22.~ ............................ . Species Released Released Released Produced BA PE Produced BA PE Produced BA PE E. diversicornis 6.190 3,010 310 5,235 3,720 5,901 640 A. cocois 9,950 1,950 11,566 500 750 53 .229 11,100 4,050 A. vexans 9,043 2,980 2,000 10,496 3,020 1,900 Total 16,140 4,960 310 25,844 7,200 2,750 69,626 14,760 5,950 Parasitoid dispersion A. diversicornis showed higher dispersion ability than A. coccois andA. vexans, reachíng 130 km from the releasing site by January 1995, 234 km by September 1995, and 304 km by ApriV1996. A. coccois dispersed slowly, reaching 180 km from the releasing site by September/1995, while A. vexans has not dispersed in the State ofBahia as ofNovember 1996 des pite the large number of releases airead y performed. Parasitoid releases of A. coccois andA. vexans started in the State ofPernambuco in August 1995, and by November 1996 a total of9,010 indíviduals had been released. lndividuals have already been recaptured 40 km away from the release site. 119 Biological control of P. he"eni Paraitoid monitoring has been carried out every two weeks, since the first release back in 1994. Results show a significant decrease on number of CM per plant dueto parasitoid action. A positive correlation between number of parasitoids and CM control has al so been observed for both A. diversicornis andA. coccois in the S tate of Bahía (Figure 5.2.1 ) . Field observations have demonstrated ~hat A. diversicornis has high dispersa! and survival abilities, but it is less aggressive than A. coccois which demonstrates a very high ability to control the cassava mealybug, especially when at high populations. Such a situation has been observed in Itaberaba, S tate ofBahia, where A. coccois, probably because of its ability to attack early stages (instars) ofthe pest, has shown a better biocontrol effect than A. diversicornis. Biological and economical effects of exotic parasitoids Studies on biological and eco no mica! effects of introduced parasitoids for the cassava mealybug control started by October, 1996. Data have been collected every two weeks and results will be presented in the 1997 Annual Report. Hyperparasitism After releasing the exotic parasitoids, A. diversicornis, A. coccois, and A. vexans, a significant increase of hyperparasitoid populations, mainly Pachyneuron sp. and Prochiloneurus sp, was observed. These hyperparasitoids have been able to reduce the parasitoid populations when they reach high levels under field conditions. o. :; ~ E ::> e C1) "' "' ~ 12 10 8 6 4 2 o 18 15 12 9 6 3 o + T ' ;, J AS ONDJF M A M 1994 ltaberaba ¡ t~ 1 ! ~ i ¡:' /" - ¡ + T ;1"+. + Avarage Number of CM/ Shoot np • Average Number of A. divers1eornlsl Shoot b ¡. 1 ¡ 1 + 1 1 1 1 JJA S 1995 ·1· ONDJ FMAM JJA 1996 - Average Number of CM/Shoot t1p *' Average Number of A coecolsl Shoot t1p 1 o/ . ' ¡:]\ l 1 f . .;:. l· ~~ + '¡ ¡- SON DJ 1994 FMA M JJ A S ON 1995 DJF MAM J JA 1996 SO N 4 3 2 o Figure 5.2.1. Population dynamics of cassava mealybug (CM) Phenacoccus herreni (second to fourth instars) in cassava fields where Apoanagyrus (=Epidinocarsis) diversicornis and Acerophagus coccois were released from 1994 to 1996 (Arrow indicates first release). 120 121 5.3 CASSA VA CROP MANAGEMENT, by José Eduardo Borges de Carvalho Effect of weeds on cassava yield Weed competition in the cassava crop may cause yield reduction as high as 40%, for both roots and aerial plant parts, if weeding is not performed at the right time during the crop cycle, mainly in ecosystems where the water availability is deficient in the soil. Weeds also act as source of food and shelter for arthropods, either pest or beneficia! organisms, as shown in Table 5.3.1 and in previous PROFISMA's Annual Reports. Table 5.3.1. Relative distribution of various classes of arthropods on weeds commonly found in cassava fields in the municipalities of Itaberaba and Piritiba, S tate of Babia. Cassava Other Natural Other pests crops pests Enemies Insects Weed s~ecies {%} {%} {%} {%} Blainvillea rhomboidea 55.5 33 .3 11.2 Acanthospermum hispidum 14.3 57.1 28.6 Diodia teres 22.2 55 .5 22.3 Croton lobatus 25 .0 25 .0 25.0 25 .0 Croton gladulosus 16.7 33.3 16.7 33.3 Passiglora cincinnata 16.7 33.3 33.3 16.7 Mitracarpus hirtus 16.7 50.0 33 .3 Polygata violacea 20.0 60.0 20.0 Borreira sp. 57.1 20,6 14.3 Eupatorium ballataefolium 40.0 60.0 Wissandula subpeltata 57.1 14.3 28.6 Cassia rotundifolia 16.7 33.3 33.3 16.7 Setaria vulpiseta 14.3 57.1 28.6 Acathospermum austra/e 25 .0 50.0 25.0 Sida cordifolia 33 .3 33 .3 33.4 Eupatorium laevigatum 42.8 28.6 28.6 Solanum erianthum 62.5 12.5 25.0 Average 21.5 33.0 22.6 22.9 Studies on weed effects on cassava yield, carried out in past years in Itaberaba and Piritiba municipalities, State ofBahia, showed that in Itaberaba the cassava root yield was very low, and consequently no significant difference for treatment was observed. In the 122 experiment installed in Piritiba, the control treatment (without weed control) produced only 3.35 metric tons/ha, a root production sígnificantly lower than the other treatments evaluated in that region (Table 5.3.2). Although stilllow, root yield in Piritiba was higher than in Itaberaba. Such a low root yield, and no significance for treatment in the experiment installed in Itaberaba, seems to be due to a very drastic decrease in rainfall, starting from September, similar to prevíous observations (PROFISMA' s Annual Report 1995). Besides the low rainfall, an unpredicted homworm attack may have also affected the cassava root yield in the experiment installed in Itaberaba. The results from the effect of weeds on cassava yield give a strong indication that weeding can start 30 days after cassava germination and continue up to five months. Table 5.3.2. Joint analyses of cassava root yield as affected by weed competition (1995/1996 growing seasons). Treatments Weed control from germination For 90 days For 120 days For 150 days For 180 days For 2 10 days Weed control beginning 30 days after germination For 60 days For 90 days For 120 days For 150 days For 180days Yield (t/ha) 1 Itaberaba 2.09 a 2.82 a 3.41 a 3.02 a 2.96 a 2 .54 a 2.72 a 2.43 a 3.06 a 2.36 a Piritiba 8.42 a 8.68 a 9.16 a 8.39 a 7.60 a 7.91 a 9.53 a 8.33 a 8.67 a 8.44 a Weed control throughout crop cycle 2. 71 a 8.22 a Without weed control 0.70 a 3.35 b Values fo llowed by the same letter in the same column, do not differ significantly (P=0.05), according to Tukey' s test. Effect of cover crops on cassava yield, arthropod population and erosion control Studies ha ve been carried out aiming at evaluating the effect of cover crops on soil fertil ity/yield, eros ion and arthropod populations in the ecosystems of Cruz das Almas, 123 Itaberaba and Piritiba. It is well known that, dueto its initial slow growth, a large portian of a cassava field is exposed to the impact of rain which causes soilloss by surface run-off. Regarding erosion control, a very low soilloss resulted from planting either jack bean or cowpea between cassava double rows (PROFISMA' s Annual Report 1995). Another tria! was planted in the same area during the 1996 growing season in arder to continue this type of evaluation. Due to a very drastic ~rought in Piritiba, no soilloss by surface run-off was recorded. Raining resumed by late December 1996, and results will be presented in the 1997 Annual Report. Table 5.3.3 shows data related to the effect of cover crop on arthropod populations in cassava fiel d. Results show that cover crops, either legume or weeds, act as source of food and shelter for both natural enemies and pests. The high amount of natural enemies found on weeds (22.6%) suggests the importance ofweed management on setting up biocontrol strategies for cassava pests. Table 5.3.3. Frequency of arthropods on cover crops in cassava fields in the municipalities of Cruz das Almas, ltaberaba, and Piritiba, State of Bahía . .................................................... .:ft~q~~~-~Y.J.~(~L ................................................. . Arthropods Legume Weeds Cassava pests 16.0 21.5 Other crop pests 75.0 33 .0 Natural enemies 9.0 22.6 Others 0.0 22.9 Total 100.0 100.0 Regarding to root yield as affected by cover crops, the joint analyses ofthe 1995-1996 growing season presented in Table 5.3.4 show that weed control over entire crop enabled the highest root yield in Cruz das Almas, and in Piritiba. On the other hand, weed control within double rows/weed between double rows, resulted in the lowest yields in both ecosystems, probably due to competition for water during period of low water availability in the soil. No significant decrease on cassava root yield by neither pigeon pea (Cajanus cajan), jack bean nor cowpea was observed in Piritiba. In Cruz das Almas, jack bean depressed cassava root 124 yield, but this behavior seems to be due to other problems that caused yield reduction in replicate two rather than competition by the cover crop. Although showing the highest root yield, weed control over entire crop raised production costs over 30% as compared with treatment four (weed control within double rows/jack bean between double rows). Table 5.3.4. Joint analysis of eassava root yield for the 1995/1996 growing seasons, in Piritiba and Cruz das Almas. . ............................ ~.~-~.!.Y.~.~!~ .. (~Q.~.~).~···········-· ······ Treatments Weed control over entire crop Weed control within double rows; weeds between double rows Weed control within double rows; pigeon pea between double rows Weed control within double rows; jack bean between double rows Weed control within double rows; cowpea between double rows Standard error(%) Cruz das Almas Piritiba 9.93 a 8.10 a 2.22 e 3 .54 b 7.27 ab 6.04 ab 3.97 be 6.24 ab 6 .10 abe 7.51 a 47.44 29.61 1 Values followed by the same letter, in the same column, do not differ significantly (P=0.05), according to Tukey ' s test. During the 1995/1996 growing season a very drastic drought occurred in Itaberaba and, due to that, the cover crop experiment installed in that ecosystem was completely lost. Effect of bagana mulching on cassava yield and weed control. By Genário Marcolino de Queiroz This study aims at determining the lowest amount ofbaganalha able to increase significantly the cassava root yíeld, to maintain soil moisture, and to reduce weeding frequency over the cycle of the cassava ero p. In order to reach these objectives, experiments ha ve been carried out in the cassava growing areas of Acaraú and Russas, State of Ceará. On farm trials installed in Acaraú showed that adding bagana at 180 m3 /ha, or more, increased significantly the cassava root yield (25 .05 ton/ha) in comparison with the 125 treatments with bagana at 45m3 /ha (10.49 ton/ha) and without bagana (8.11) (Tab1e 5.3.5). Regarding weed ineidenee in the eassava field with bagana treatment, a predominanee of weeds was able to sprout out of the roots, sueh as "Catanduba" ( around 90% ), "Cipauba", "Catingueira" and "Cipó-pau" (about 10%). Table 5.3.5. Effect of the bagana mulching on the cassava root yield, a erial part and starch content at Acaraú, Sta te of Ceará. Yield {ton/ha} 1 Starch content Bagana levels (m3/ha) Root Aerial part (%) 315 28.98 a 37.10 a 24.76 ab 270 23 .52 a 30.46 ab 25 .34 a 225 22.12 ab 26.82 abe 24.99 ab 180 25 .05 a 32.73 ab 25.05 ab 135 18.64 abe 21.18 bed 25.34 a 90 17.14 abe 14.70 ed 22.82 b 45 10.49 be 10.89 d 23.99 ab o 8.11 e 8.41 d 24.89 ab Values followed by the same letter, in the same eolumn, do not differ signifieantly (P=0.05), aeeording to Tukey's test. The results obtained from the on farm tria! installed in Russas, in February 1995, showed no effeet ofbagana mulching on the eassava root yield (Table 5.3.6). This might be dueto the rainfall that oeeurred from January to July 1995, eorresponding to the period of initial growth ofthe erop (Table 5.3.7), orto soil eharaeteristies as well. It is interesting to mention that previous experiments installed in Russas, using bagana as mulehing for cassava fields, were lost due to drought and/or high soil temperatures (previous PROFISMA Annual Reports). Table 5.3.6. Effect of bagana mulching on the cassava root yield, aerial part and starch content at Russas, Sta te of Ceará. 126 Bagana levels Cassava yield (ton/ha) Starch content m 3 /ha Root Aerial part 315 11.46 23.47 270 225 180 135 90 45 o 10.93 9.98 13.45 13.85 12.16 7.71 15.53 24.40 21.20 27.62 27.69 23 .97 23.43 24.76 (%) 24.34 25.46 25 .05 24.92 26.37 28.40 26.43 24.36 The experiment installed in Acaraú, in February 1996, shows normal development; on the other hand the experiment in Russas presented problems related to germínation probably due to high soil temperature in the bagana covered plots. Both trials will be harvested by July August, 1997. Table 5.3. 7. Month January February Mar eh April May June July August September October November December Total Rainfall (mm) at Russas, State of Ceará. Year 1993 1994 45.2 75.9 0.0 79.4 76.0 160. 1 34.6 129.6 4.4 127.4 0.0 115 . 1 13.2 24 .0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 41.0 173.4 752.5 1995 45.6 104.7 129.0 329.2 213.4 45 .6 38.6 0.0 0.0 0.0 0.0 0.0 906.1 1996 74.2 147.4 250.0 190.3 48.6 0.0 25 .6 2.0 738.1 127 5.4 PLANT PATHOLOGY, by Chigeru Fukuda 5.4.1 Cassava Witches' Broom Disease Cultural Control of Cassava Witches' Broom Disease The cassava witches' broom disease (CWBD), caused by a phytoplasm, is a very serious cassava problem in the "Serrada Ibiapaba", State ofCeará, growing area. An important - achievement on the research of CWBD was the observation that "cleaning-up" ( elimination of infection) infected planting material of susceptible varieties enabled high root yield. Despite the effectiveness ofthe cleaning-up technique, it is only a matter oftime for the disease-free susceptible planting material to become re-infected by the CWBD pathogen under field conditions, thus requiring continuous cleaning-up procedures. Genetic control of CWBD The study on genetic control of CWBD started in 1986, as pan of the EMBRAP Al CNPMF Cassava Breeding Program, aiming at identifying cassava genotypes resistant/ tolerant to that disease. By 1993, this study began to be an EMBRAP A/CNPMF Cassava Breeding Program!PROFISMA joint activity. From 1986 to 1993 eight hundred and seventy cassava clones from the EMBRAP A/CNPMF Cassava Breeding Program were tested, through farmer participatory methods, in the Serra da lbiapaba growing area, where there are about 5,000 ha planted with cassava, most ofthem planted with the cassava variety Cruvela, highly susceptible to the CWBD pathogen. After severa! years of evaluation, seven promising clones were selected. In 1994 these seven clones were then evaluated in comparison with the local variety Cruvela, susceptible to the pathogen, and the CWBD resistant variety Bujá, was recommended by the State ofCeará Agricultura! Research Institution (EPACE). On-farm trials were installed in 60 rural communities, located in seven municipalities in the Serra de Ibiapaba cassava growing region. Evaluation at harvest time, carried out by cassava growers in 53 rural communities enabled the selection of the clones 8709-2 and 8911-16 for harvesting at both 12 and 18 months after planting, clone 8952-06 only for harvesting at 18 months and clone 8 7 40-1 O for harvesting at 12 months after planting. 128 Results of a simplified phytosanitary diagnostic survey, carried out during 1994 in the Serrada Ibiapaba cassava growing region, showed that one ofthe most important factors involved in the CWBD spread throughout the region ís the sharing of planting material, a very common practice among cassava growers. Such a behavior gíves an indication that the selected CWBD resistant clones would be dispersed within rural communities by sharing planting material. A very strong support to this assumption is the observation that, the stem of several selected clones had already been removed by cassava growers prior to harvest and, according to them, used for either replacing their own field or donating to other cassava growers. It is interesting to mention that growers' preferences for the evaluated cassava clones/varieties were ~ 60% for the clones 8709-2, 8952-6 and 8911-16 as compared with 15% to 35% for the local variety Cruvela. Increase in adoption level ofthe selected clones is expected through action of the EMA TER CE by ditfusion methodologies. Considering 60% adoption ofthe CWBD resistant clones, this would imply in a replacement of about 3,000 ha ofthe susceptible variety Cruvela. Considering the root yield potential ofthe selected clones, around 20 metric ton/ha, such replacement would representa 60,000 metric ton yield. Considering the average price ofUSD $41.80/cassava root metric ton, this would give an estimated USD $2,508,000.00 production value increase. 5.4.2 Root Rots, by Chigeru Fukuda and Aristoteles Pires de Matos Cassava root rot disease survey in the State of Babia A survey for the cassava root rot incidence was conducted in three cassava growing regions ofthe State ofBahia. Evaluations, performed in nine rural communities oftive municipalities, showed that the disease incidence varíes from region to region. The highest levels of root rot inciden ce, 50%, were observed in the rural community Saco, municipality of Cachoeira, and in a cassava field located in the municipality of Irará. In the rural community Chapada, municipality of Aporá, the disease incidence was 29%. Very low root rot incidence, 1%, was observed in the rural community Buril, munícipalíty of Crisópolis. Depite growers information on the occurrence ofthe cassava root rot disease in the rural community Umbuzeiro, only one infected plant was found . This is probably dueto the 129 prevalent dry season during the survey. Isolations carried out under laboratory conditions showed that the pathogens involved in the cassava root rot development in the evaluated rural communities in the State ofBahia were Fusarium sp., Phytophthora sp., or Scytalidium lignicola. Genetic control of the cassava root rot Field evaluation of 30 cassava hybrids, from the EMBRAP A/CNPMF Cassava Breeding Program, in the municipality of Umbauba, S tate of Sergipe, enabled the selection of five of them that were evaluated again in an on-farm tria! in a randomized block design, with three replicates. Data in Table 5.4.2.1 show that hybrids 90.175-7 and 90.117-5 expressed 11 .54% of root rot infection, 90.185-1 showed 14.29% of infection while the hybrids 90.113-2 and 90.115-5 showed over 30% ofroot rot infection. Besides showing field tolerance to root rot disease, the hybrid 90.175-7 showed root weight of 1.12 kglplant, a little lower than the susceptible hybrid 90. 113-2, which yielded l. 24 kg of root/plant. All the other hybrids produced root weights varying from 0.58 to 0.94 kglplant. Table 5.4.2.1. Reaction of cassava hybrids to natural infection of root rot pathogens, 12 months after pJanting, in Umbauba, S tate of Sergipe. lnfected plant Total root weight Average root weight Hybrid (%) (kg) (kg/plant) 90.113-2 31.03 26 1.24 90.115-5 39.29 17 0.94 90.117-5 11 .54 14 0.58 90.175-7 11.54 28 1.1 2 90.185-1 14.29 20 0.80 Eighteen cassava varieties/hybrids were evaluated for root rot incidence in a on farm tria! carried out at the COPAL Quitéria, Municipality of Alagoa Grande, State ofParaiba, in an area naturally infested by root rot. As shown in Table 5.4.2.2 the hybrid 148-2 expressed field tolerant reaction to root rot pathogen with 9. 5% of infected plants and only 2% of infected roots. Moderate infection was observed in the hybrids 46-03 and 128-8, respectively 23.8% and 38.1%, while all the other evaluated genotypes showed high root rot infection, including the local varieties. The hybrid 148-2, besides expressing tolerant reaction to the cassava root rot, also showed a good root yield, 2.6 kg!plant, higher than all the other evaluated varieties/hybrids. 130 Table 5.4.2.2. On-farm evaluation of 18 cassava genotypes for root rot incidence, in the rural community "Engenho Mares", municipality of Alagoa Grande, S tate of Paraiba. - Infected root Infected plant Total root Average root Genotypes (%) (%) weight (kg) weight (kglplant) 179-4 51.0 100.0 12.6 0.60 32-06 89.0 100.0 1.2 0.06 A Bravo Branco 67.0 100.0 11.7 0.56 Bujá 100.0 100.0 0.0 0.00 M. México.59 100.0 100.0 0.0 0.00 Maria Pau 62.0 100.0 6.3 0.30 Monge* 100.0 100.0 0.0 0.00 Passarinha 100.0 100.0 0.0 0.00 Osso Duro* 83.3 90.5 4.4 0.21 Isabel de Souza 56.0 80.9 21.0 1.00 Olho Verde 29.2 80.9 10.0 0.48 Fío de Ouro 11 .0 61.9 26.0 1.24 Paraibana* 50.0 61.9 9.6 0.46 128-8 22.0 38.1 29.8 1.42 100-08 55 .0 23 .8 17.6 0.84 46/03 20.0 23 .8 21.8 1.04 Cedinha* 48.0 23 .8 21.0 1.00 148-2 2.0 9.5 54.0 2.60 * Local varieties One hundred and seventy-nine hybrids, from the EMBRAP A/CNPMF Cassava Breeding Program, were evaluated in on-farm trials in the municipality of Vi¡;osa, S tate of Alagoas, where root rot incidence was relatively high. Thirty hybrids (16.76%) showed no root rot infection and were selected for further evaluations under field conditions while the others expressed severa! degrees of root rot incidence. High susceptibility to root rot pathogens was expressed by seventy hybrids that showed over 50% of infected roots (Table 5.4.2.3). 131 The trial installed in the COP AL Chapada was aimed at evaluating the effect of planting systems on the behavior oftwo cassava varieties regarding root rot incidence. It was observed that disease incidence is lower when cassava stakes are planted in ridges in comparison with planting in holes. The results also show that planting the susceptible local variety Cemitério in ridges, despite having both plant and roots infected by root rot pathogens, yielded the highest ro9t weight, l . 122 kglplant. Such a good behavior of Cemitério explains the cassava growers preference for this variety. Table 5.4.2.3. Field evaluation of cassava hybrids, 1992 family, for resistance to root rot pathogens, in the municipality of Vi~osa, Sta te of Alagoas. Class of root rot Plants/class incidence Number Percentage No infected roots 30 16.76 0.1 to 10% 27 15.08 10.1 to 25% 20 11.17 25 .1 to 50% 32 17.88 Over 50% 70 39. 11 Total 179 100.00 In 1995 a high incidence of cassava root rot was observed in the rural community Chapada do Aporá, municipality of Aporá, State ofBahia. Two on-farm trials were installed in that rural community aiming at evaluating the behavior of cassava varieties and promising hybrids regarding root rot incidence. As shown in Table 5.4.2.4, despite showing 45% infected plants and 15 .2% infected roots, the local variety Cemitério hada very good performance regarding to root yield, l . 44 kglplant, just behind Paraibana, that yielded l . 52 kg/plant , while Olho Verde showed the lowest root weight, only 0.285 kg/plant. As can be concluded from data in Table 5.4.4, root rot incidence was very low throughout the experiment, except for the local variety Cemitério. It is possible that the root rot incidence in Cemitério may be dueto the use of infected planting material rather than infection from contaminated soil. 132 Table 5.4.2.4. Reaction of cassava genotypes to root rot pathogens under field conditions in rural community of Chapada do Aporá, municipality of Al!orá, Sta te of Babia. Infected Total root Average Starch lnfected root plant weight RW content Genoty(!eS {%} {%} {kg} {kg/(!lant} {%} Rosinha 0.0 0.0 22.20 1.11 28.98 128/8 0.0 0.0 14.40 0.78 28.75 Cedinha 0.0 0.0 25 .00 1.25 28.41 Paraibana 0.0 0.0 30.40 1.52 26.61 Milagrosa 0.0 0.0 22.70 1.14 30.28 Olho Verde 0.0 0.0 5.70 0.29 26.84 Olho de 0.0 0.0 24.60 1.37 23 .68 Porco Cemitério 15.2 45 .0 28.80 1.44 A very low root rot incidence was observed in the tria! where a total of25 cassava hybrids from the El\!1BRAP A/CNPMF cassava breeding program were under field evaluation (Table 5.4.2.5), despíte the high incidence ofthe disease in the previous growing season. Such an erratic cassava root rot incidence in the same area, from season to season. has been a problem for evaluating/selecting cassava germplasm for root rot resistance under field conditíons. Due to that it was decided, parallel with field evaluations, to develop evaluation techniques, under controlled conditions, aiming at identifying cassava root rot resistant/tolerant genotypes. 133 Table 5.4.2.5. Reaction of the local cassava variety and hybrids, from the EMBRAP Al CNPMF Cassava Breeding Program, to root rot pathogens under field conditions, rural community of Chapada do Aporá, municipality of Aporá, S tate of Babia. lnfected root Infected plant Average root weight Varietylhibrid {%} {%} {kg/~lant} 076 2.6 5.0 1.060 056 0.0 0.0 0.471 576 0.0 0.0 1.820 537 26.7 31.58 0.505 478 0.0 0.0 1.158 260 5.0 5.26 0.568 896 0.0 0.0 0.978 554 0.0 0.0 1.856 967 0.0 0.0 0.481 9660.0 0.0 0.0 0.775 858 0.0 0.0 0.374 859 0.0 0.0 1.594 1183 0.0 0.0 0.835 1189 0.0 0.0 0.826 538 0.0 0.0 1.100 142 9.5 16.67 1.578 116 0.0 0.0 1.325 812 0.0 0.0 1.620 053 0.0 0.0 0.500 645 0.0 0.0 2.142 613 8.3 6.25 0.694 649 0.0 0.0 0.800 604 0.0 0.0 0.442 387 0.0 0.0 0.450 598 0.0 0.0 0.920 Cemitério 0.0 0.0 2.071 Studies on cassava root rot under controlled conditions In order to establish a feasible method for evaluating cassava germplasm for resistance to root rot disease caused by Fusarium sp. severa] inoculation techniques were tested under 134 laboratory and greenhouse conditions. No disease development was observed when healthy planting material was planted in artificially contaminated soil prior to planting. Although showing disease development, the pathogen inoculation in branches preved to be time- consuming sínce it takes about 60 days from planting to recording data. Stake inoculation, under laboratory conditions, either by adding a conidial suspension to a wound or by transferring a disk of colony to wounded tissue, gave the best response since evaluation may be performed one to two weeks after inoculation. Severa! isolates of Fusarium sp. were obtained from infected cassava tissue collected in cassava growing areas in Northeast Brazil. Pathogenicity tests carried out with these isolates inoculated in stakes of susceptible varieties showed that, although pathogenic, the virulence ofthe Fusarium sp. isolates was not high. On the other hand, isolates of Scytalidium lignicola expressed high virulence, inciting higher lesion development than Fusarium sp. (Figure 5.4.2.1). S. lignicola was also able to incite disease development even when inoculated at low inoculum concentrations, such as 103 conidia/mL (Figure 5.4.2.2). 1,4 1,2 0,8 0,6 0,4 0,2 o BGM 0859 BGM 0032 BGM 009 1 Figure 5.4.2.1. Reaction of three cassava genotypes to stake inoculation with Fusarium sp. and Scytalidium lignico/a. 135 Preliminary results regarding reaction of30 cassava genotypes to inoculation with Fusarium sp. showed that the average area oflesion, one week after inoculation, was very small, varying from 0.29 cm2 up to 1.57cm2, while the control variety Correnteza, that shows high susceptibility to root rot pathogens under field conditions, showed only 0.45 cm2 average area of lesion, indicating again the low virulence of Fusarium sp. to the cassava crop. Although preliminary, the pathogenicity tests perforrned up to now seern to indicate that, in Northeast Brazil, S. lignico/a rnight be a more important pathogen to the cassava crop than Fusarium sp., and P. drechsleri seems to be even more virulent than S. lignicola. Studying the effect of inoculum concentration on disease development incited by S. lignicola in stakes ofthree cassava genotypes showed that the pathogen was able to cause sorne infection even at very low concentrations such as 103 conidia/mL. Increasing inoculum concentration increased lesion development up to 105 conidia/mL. Inoculum concentration of 1 06 conidia/mL did not significantly increase lesion development (Figure 5.4.2.2). 1,4 1,2 0,8 0,6 0,4 0,2 o Disease index . BGM 0032 C'JBGM 0091 IIIBGM 0859 1:::)1:::) " Figure 5.4.2.2. Effect of inoculum concentration of Scytalidium lignicola, on disease development of three cassava genotypes. 3 2 o 136 Biological control of cassava root rot disease The effect of the antagonistic fungi Trichoderma spp. was evaluated on the biocontrol of the cassava root rot caused by either P. drechsieri or S. lignicoia. As shown in Figure 5.4.2.3 the isolate T 11, Trichoderma poiysporum, when simultaneously inoculated with P. drechsieri, was able to antagonize the pathogen, as expressed by the significantly smaller size of the lesion in root ofboth "::arieties Maria Pau (BCG 118) and Manteiga (BGM 801). The isolates T 25, T harzianum, and T 26, T. pseudokoningii, did not show antagonistic effect to P. drechsieri in inoculated cassava roots. Data in Figure 5.4.2.4 show that the isolates T 11, T poiysporum, T 25, T harzianum, and T 26, T. pseudokoningii, significantly decreased the size of lesion, when simultaneously inoculated with S. lignicoia in cassava stakes, variety Manteiga, thus indicating antagonistic effect to the pathogen. On the other hand, no antagonistic effect was observed when inoculations were performed in stakes of the cassava variety Maria Pau. Disease severy score (1-- ·<· IBGM 801 -BGM 118 Figure 5.4.2.3. Effect of three Trichoderma spp. isolates, simultaneously inoculated with Phytophthora drechsleri, in cassava roots, under laboratory conditions. Disease index scorre 5 4 3 CJBGM 801 -BGM 118 Figure 5.4.2.4. Effect of Trichoderma spp. isolates, simultaneously inoculated with Scytalidium lignicola in cassava stakes, under laboratory conditions. 5.4.3 Virology, by Celia Camara do Vale, Jose Albercio Lima, Chigeru Fukuda 137 The main objective of the virology activity during 1996 was to determine root yield losses due to the cassava vein mosaic virus (CVMV). An experiment to study such a kind of problem requires a large amount ofvirus-free planting material, but the PCR technique did not perform well as a diagnosis method for large numbers of plants. Due to that, the production of CVMV -specific antiserum became the main objective since this technique is reliable and more easy to perform. Severa! modifications on the CVMV purification technique, such as viral separation through polyacrylamide gel electrophoresis, were made aiming at overcoming the presence of host protein in the purified virus. Regardless of all modifications, host plant protein was still present in the " purified" virus preparation. It was then decided to study the CVMV host range in order to identify a host from which the virus could be successfully purified. The following plants were evaluated: Amaranthus caudatus, Capsicum annun, Cassia 138 occidentalis, Canavalia ensiformis, Chenopodium amaranticolor, e mura/e, e quinoa. Citrullus vulgaris, Cucumis anguria, e me/o, Gossypium hirsutum, He/iantus annun, Lactuca saliva, Lycopersicum escu/entum, Nicotiana benthaminana, N rustica, N glutinosa, N tabacum, Sesamum orienta/e and Manihot escu/enta. Severa! inoculation techniques were tested, including injection of purified virus into plant tissue, and grafting. Syrnptorn development was observed only i_n cassava plants inoculated by grafting. This study will continue by evaluating other potential host plants. 5.5 CASSA VA INTENSIVE DIAGNOSTIC SURVEY IN NORTHEAST BRAZIL, by Jose Hurnberto Almeida de Cerqueira and Carlos Estevao Leite Cardoso Land property and usage About 59% ofthe surveyed cassava growers in Northeast Brazil have land property title, and the average area for them is 12.9 her; 16.7% ofthe cassava growers, average land area around 14 her, ha ve land property document other than title; other 18.9% own their land, 5, 7 her average area, but without property title; 17.1% ofthe surveyed growers are sharecroppers with an average land area of l. 5 her; and 3.2% are renters with 1.4 her average land area. The large arnount of cassava growers without land property title, over 40%, constitutes a big problem regarding to agricultura! credit. The average Iand area of the cassava growers surveyed in Northeast Brazil, varies from 1.9 to 72 her (Table 5.5.1). Figure 5.5.1 shows land area distribution within the surveyed population. It was found that over 53% ofthe growers have land area smaller than 5 her and for almost 70% the land area is smaller than 10 her, and only 3.6% have land area larger than 60 her. Table 5.5.1. Land area of the surveyed cassava growers in Northeast Brazil. Average Freguenc~ Laod extracts land area Absolut Relative Accumulated {her} {number) {%) {%) Smaller than 5 her 1.9 150.0 53.4 53.4 Larger than 5 her, smaller than 1 O her 6.8 44.0 15.7 69.0 Larger than 1 O her, smaller than 20 ~er 12.6 45 .0 16.0 85.1 Larger than 20 her, smaller than 30 her 24.4 13 .0 4.6 89.7 Larger than 30 her, smaller than 60 her 48.6 19.0 6.8 96.4 Laq~er than 60 her, smaller than 90 her 72.0 10.0 3.6 100.0 Total is higher than 1 00% because sorne growers ha ve are as in more than one land usage. 15,7% Smaller than S ha 53,3% i 10-20 ha 16,0% 60 - 90 ha 3,6% 6,8% 20-30 ha 4,6% 139 Figure 5.5.1. Distribution of average land area of cassava growers in Northeast Brazil. Data in Table 5.5.2 show that all surveyed cassava growers use part oftheir land, 3.9 her average area, to plant either annual or perennial corps. lt is interesting to call the attention for the land used as pasture, 29.9%, left under fallow, 29.9%, and with natural vegetation, 13 .2%, all of thern generally used for grazing. This suggest a high irnportance of grazing for the surveyed cassava growers. 140 Table 5.5.2. Land usage and average land area in each kind of use by cassava growers in Northeast Brazil. Relative participation1 Average land Land usage (%) area (her) Either annual or perennial crops 100.0 3.9 Pasture 29.9 5.0 Fallow 29.9 7.4 Native vegetation 13.2 10.2 Forest 11.4 16.4 Non productive area 9.6 0.7 Lowland area 7.8 7.0 1 Total of relative participation is higher than 100 because all the cassava growers surveyed have their land occupied with several uses. Rural family Table 5.5.3 shows family composition in rural communities in Northeast Brazil where cassava is one ofthe main crops. Over 96% ofthe families are composed by parents son, and daughters, while 3.1% of them include al so other relatives. The educational level of family members in the surveyed rural communities is very low. The majority ofthe youth does/did not attend school, that results in 89.02% ofthe daughters and 80.64% ofthe sons without any school education. It was found that 57.2% ofthe parents and 76.8% ofthe fami ly members including parents did not attend school, while 23 .2% are either elementary or high school educated. The head ofthe family is predominantly aman older than 18 years; 52.5% ofthem are 50 year old or less, and 70.7% are around 35 year old. Most ofthe cassava growers, 86.4%, live in the rural community while the remaining 13 .6% live in other places. Approximately one third ofthe family members, that is 35.8%, does not work in the rural community: they are either school age children or young people whose activities are other than rural labor. A little high percentage, 36.5%, work only in theír land, and 28.7%, besides workíng in their land, also sell their labor force in order to make extra money and increase the family income. Table 5.5.3. Family composition and educationallevel of the cassava growers in Northeast Brazil . 141 .................................. !~.~-~.tr .. ~~~P..~.~~~~-~~---·-········ ······ ·····-············· ···-·· ··· ······ ········--~-~~-~~.!~~-~~L!~~-~.t-···-···········--···· School educated Not school educated Members Freguenc! {%} Number {%} Number {%} Head 318 15.6 182 57.23 136 42.77 Wife 301 14.7 196 62.12 105 34.88 Daughter 674 33 .0 600 89.02 74 10.98 Son 687 33 .6 554 80.64 133 19.36 Son in law 14 0.7 9 64.29 5 35 .71 Sister 6 0.3 4 66.67 2 33 .33 Brother 4 0.2 4 100.0 Mother 12 0.6 5 41.67 7 58 .33 Granddaughter 6 0.3 3 50.00 3 50.00 Grandson 16 0.6 8 66.67 4 33 .33 Daughter in law 5 0.2 2 40.00 3 60.00 Father 0.0 Cousin 0.0 100.00 Niece 2 0.1 50.00 50.00 NeEhew 0.0 100.00 Cassava production technology According to the recorded data, 18.9% of the cassava growers in Northeast Brazi l use to hoe the area as part ofthe soil preparation activities; for most ofthe growers, 63 .7%, soi l preparation is performed only with hoe. Seventy three percent ofthe cassava growers in Northeast Brazil use stakes oftheir own to start a new planting, and 42% ofthem said they carry out any kind of stake selection prior to harvesting. The apical and basal portions of the stem are discarded and only the medium portien is used as planting material as stated by 82.9% ofthe Northeastern cassava growers; the stake must be 1 Oto 20 cm long and show a diameter higher than 2 cm according to 63 .8%. Most ofthe growers (82.6%) transpon the cassava stem as a tied up bundle, 61.2% plant the stake in a horizontal position, and 63 .2% of them plant in a shallow hole. If necessary, the cassava planting material may be stored for up to 60 days according to 50.5% the cassava growers, and 38.8% ofthem said that they store the planting material under a tree shade. Fertilizer and correction of soil pH by liming is nota usual practice among cassava growers in Northeast Brazil, since 65 .5% ofthem said they do 142 not carry out these practices. For 63 .3% ofthe growers cassava planting is performed by the beginning of the raining sea son. Regarding to the topography of the area to be planted, 70% said they use flat areas, 27.5% ofthem plant in the slope area, and only 2.5% install their cassava fields in lowland areas. Pest and disease incidences As shown in Figure 5:5.2, leaf-cutting ants (29.4%) is the most important cassava pest in Northeast Brazil, followed by root rot disease (20. 7%), horworm (16.5%), cassava green mite (13 .0%), whiteflies (8.2%), witches ' broom disease (CWBD; 3.4%), stake weevil (3 .1%), and other pests and diseases (5 .8%). Except for the cassava root rot disease that causes root infection, all the other cassava pests and diseases damage to the leaves and branches as shown in Table 5.5.4. Leafcutter ants have been considered a very important cassava pest in the last years, followed by CGM, homworm, whiteflies, and root rot. In general, cassava pests are responsible for around 22% decrease of root yield, except for witches' broom and root rot diseases that cause, respective! y, 71 .9% and 51.8% root yield losses. The stage of plant development at the time of pest/disease detection varíes from 1-3 month after planting for ants, and up to 12 months for root rot. Root rot 20.7% ~hte 13.0% 8.2'/o Homwonn 16.5% 3.2% Others 5.8'/. Figure 5.5.2. Main pests and diseases of the cassava crop in Northeast Brazil, as identified by cassava growers. 143 Table 5.5.4 Cassava pests and diseases in Nortbeast Brazil and their importance to tbe crop, according to growers. Infection Infected organ (%) Period of Yield Pest/disease Homwonn Mi te Leaf-cutting ants Root rot Whiteflies Witches ' broom Stake weevil Others (%) Root Aerial part observation (year) loss (%) 16.5 98.3 13.0 81.5 29.4 99.0 20.7 8.2 3.4 3.1 5.8 99.3 96.6 100.0 95 .5 15 16 19 11 13 5 7 22.7 21.6 22.3 51 .8 22.9 71.9 33 .5 The cassava growers' action to solve these problems are presented in Table 5.5.5. The data show that cassava growers in Northeast Brazil do not apply any control measure for CGM, termites, whiteflies, stake weevil nor witches' broom disease. On the other hand, attempts to control cassava pests by growers has been based on insecticide application. For cassava root rot disease, growers have been using lime and ash as soil amendment aiming at controlling the disease. Commercialization Looking at the source of information from where the surveyed growers know about prices of cassava products, it was found that, for cassava flour, starch and fresh root, the information comes from the local market, while for cassava chips the information comes from the extension personnel (Table 5.5.6). The majority ofthe cassava production in the Northeast, 76.8%, is soldas cassava flour, 19 .1% is soldas fresh root, 1.6% is used as animal feeding, and only 0.7% is transformed to chips (Table 5.5.7). Regarding the manner of commercialization, 21.17% ofthe cassava flour, 0.05% of "tapioca", and 1.03% ofthe starch products are sold to wholesalers; 0.37% ofthe "beiju" production is sold directly to the consumer; 0 .37% ofthe chips is commercialized through growers associations; and 5.85% ofthe root production is sold to truck drivers (Table 5.5.8). 144 Table 5.5.5. Control measures for pests and diseases used by cassava growers in Northeast Brazil. Pest Gro"'en Klnd of Month "'hen control measures are a(!J!lied Disuse {%l Control 2 3 4 S 6 7 8 9 10 11 12 Mites 100 None Homwonn 50 Non e so Chemical + + ++ ++ ++ ++ + + + Ants u non e 98.5 chemical + ++ ++ ++ ++ ++ + + + Tennites lOO none Whiteflies lOO non e Stake weevil lOO non e Anthracnose so non e so chemical + + ++ ...... ++ ...... + + + Root rot so non e 33.3 lime ++ +- ++ + 16.7 Ash + +- ++ -+ ++ + + + CWBD 100 non e Table 5.5.6. So urce of information regarding price of cassava products . ........................................................ f..~~~.!-:.~~-~~ .. PE~.~-~-~~---······························· ······················ Source of information Chi~s {%) Flour {%) Starch {%) Fresh root {%) Growers association 2.5 1.4 1.1 2.5 Extension worker 6.8 8.2 4.3 7.8 Other grower 3.6 31.3 9.6 17.4 Wholesaler l.l 31.3 10.7 24 .9 Radio 1.4 6.0 1.1 1.8 Newspaper/TV 1.1 1.4 0.0 0.7 Local market 2.5 64.8 12.1 17.4 Non e 0.4 1.4 1. 1 0.4 Table 5.5. 7. Destination of the cassava production in Northeast Brazil. Destination Percentage Sell as fresh root 19. 1 Cassava flour Chips Animal feeding Others 76.8 0.7 1.6 1.8 145 Table 5.5.8. Destination of cassava roots and cassava l!roducts in Northeast Brazil. Flour "Beiju" "Tapioca" Starch Chips Roots Destination (%} (%} (%} (%) {%} {%} Consumer 5.64 0.37 0.02 0.05 0.00 1.87 Truck drivers 6.28 0.24 0.00 0.02 0.09 5.85 Wholesalers 21.17 0.03 0.05 1.03 0.00 5.50 Sellers 15.90 0.36 0.00 0.76 0.00 1.01 Associations 0 .00 0.00 0.00 0.00 0.37 0.78 For the family 26.46 1.73 0.96 2.42 0.04 0.98 Total 75.45 2.73 1.03 4.28 0.50 15.99 Regarding to "casa de farinha", the place where cassava roots are processed to flour and other cassava products, such as "tapioca" and "beiju", the results of the survey showed that 39.5% ofthe cassava growers use a communal "casa de farinha" to process their harvest; 33 .1% ofthem rent the "casa de farinha" from somebody else; 14.9% have their own "casa de farinha"; 6.0% does not pay for using somebody else ' s "casa de farinha"; and 6.4% did not give a specific answer. Results related to utilization of the aerial part of the cassava plant indicate that, although 38.3% ofthe aerial part is used for animal feeding, about the same amount, 37.5%, is not utilized by the growers, and only 23 .3% is used as planting material . Selling the aerial part of the cassava plant is nota common practice among cassava growers in Northeast Brazil, since only 1.0% ofthem does it. 146 6 PARTICIPANTS AND INSTITUTIONS Alba Rejane Nunes Farias EMBRAP A/CNPMF Alrnir Dias da Silva IPA Aloyséia Cristina da Silva Noronha EMBRAP A/CNPMF Ana Maria Mascarenhas Eloy Canto IICA/CNPMF Anthony Bellotti CIAT Antonio Mela EMEPA Antonio Rairnundo Cavalcante EMATERCE Apoliano A. dos Santos EPACE Aristoteles Pires de Matos EMBRAP A/CNPMF Arlinda Pereira Eloy CNPq/CNPMF Bernardo Ospina Patiño CIAT Célia Carnara do Vale IICA/CNPMF /UFC Chigeru Fukuda EMBRAP A/CNPMF Cícero T. C. Pereira EMATERCE Cláudio Luiz Messias UNICAMP Cláudio Jonsson EMBRAP A/CNPMA Edna Castilho Leal EMBRAPA/CPATC Edson Mathias EMEPA Elizabeth Alvarez CIAT El ton de O. dos Santos IPA Fernando Gallindo IPA Francisco de Assis Paiva Ca,pos IPA Genário Marcolino Queiroz EPACE Gilbert J. de Moraes EMBRAP A/CNPMA Henrique A. Lima EMATERCE !talo Delalibera Juniorr IICA/CNPMF 147 Jandira F. Warumby IPA José Albércio A. Lima UFC José da Silva Souza EMBRAPNCNPMF José de Arimatéia Gon~alves EPACE José Eduardo Borges de carvalho EMBRAPNCNPMF José Fortunato Silva EBDA José Humberto A. Cerqueira EBDA José Maurício Sim6es Bento IICNCNPMF Joselito da Silva Motta EMBRAP NCNPMF Joseph Tohmé CIAT Josias Cavalcante EMBRAP NCPTSA Laércio Duarte Souza EMBRAPNCNPMF Lee Calvert CIAT Lincoln Smith CIAT Luciano da Silva Souza EMBRAPNCNPMF Mabrouk El-Sharkawy CIAT Marcelo Brito EMDAGRO Maria das Gra9as C. de Sena EMBRAPNCNPMF Maria de F á tima Muniz UFAL Maria de Fátima Barros Gon9alves UFC Maria Luzia Cavalcante EPACE Mauto de Souza Diniz EMBRAPNCNPMF Nicolau Schaun EMBRAPNCNPMF Pedro Alves de Almeida EMBRAPNCNPMF Pedro Luiz Pires de Mattos EMBRAPNCNPMF Sandra Lúcia de Carvalho Fontes EBDA Stephen Lapointe CIAT Wagner Pereira Felix UFCFJCNPq EMBRAP NCNPMF Wania Maria Gon~alves Fukuda 7 ACRONYMS CGM CIAT CM CNPMA CNPMF CNPq COPAL CPATC CPATSA CVMV CWBD EBDA EMATERCE EMATERPB EMATERPE EMBRAPA EMDAGRO EMEPA EPACE ESCaPP FPR IFAD Cassava g¡:een mite Centro Intenacional de Agricultura Tropical Cassava mealybug Centro Nacional de Pesquisa de Monitoramento e A valia~ao de Impacto Ambiental Centro Nacional de Pesquisa de Mandioca e Fruticultura Tropical Conselho Nacional de Desenvolvimento Científico e Tecnológico Comite de Pesquisa Agrícola Local Centro de Pesquisa Agropecuária dos Tabuleiros Costeiros Centro de Pesquisa Agropecuária do Trópico Semi-Árido Cassava Vien Mosaic Virus Cassava Wítches' Broom Disease Empresa Baiana de Desenvolvimento Agrícola Empresa de Assistencia Técnica e Extensao Rural do Ceará Empresa de Assistencia Técnica e Extensao Rural da Paraiba Empresa de Assistencia Técnica e Extensao Rural de Pernambuco Empresa Brasileira de Pesquisa Agropecuária Empresa de Desenvolvimento Agropecuário de Sergipe Empresa Estadual de Pesquisa Agropecuária da Paraíba S/ A Empresa de Pesquisa Agropecuária do Ceará Ecologically Sustainable Cassava Plant Protection Farmer Participatory Research Intemational Fund for Agricultural Development 148 IICA liTA IPA PROFISMA 149 Instituto Interamericano de Cooperación para la Agricultura Intemational Institute of Tropical Agriculture Empresa Pemambucana de Pesquisa Agropecuária Proteyao Fitossanitária Sustentável da Mandioca na América Latina e África PROSERTAO Projeto de Apoio ás Famílias de Baixa Renda da Regiao Semi-Árida UFAL UFC UNICAMP de Sergipe Universidade Federal de Alagoas Universidade Federal do Ceará Universidade de Campinas 8 PUBLICA TIONS ACAROLOGY Refereed Science Journals: Bohórquez Chaux, A. , L. Smith, G. Gallego & J. Tohmé. 1996. Estudios de Cepas del Hongo Entomopatógeno Neozygites sp. (Zygomycetes: Entomophthorales) Sobre Acaros de la Familia Tetranychidae. Resumenes XXIII Congreso de la Sociedad Colombiana de Entomología, 17-19 July, 1996, Cartagena, Colombia. p. 97. Bóhorquez-Chaux, A. , A. R. Braun, G. Gallego, L. Smith and J. Tohmé. submitted. Characterizing Populations ofCassava Green Mite by Molecular Techniques. Molecular Ecology. Brower, J. H., L. Smith, P. V. Vail and P. W. Flinn. 1995. Biological Control, pp. 223-286. In B. Subramanyam and D. W. Hagstrum (eds.), Integrated Management oflnsects in Stored Products. Marcel-Dekker, New York. Burbano M., E. G., P. Chacón de Ulloa & L. Smith. 1996. Efecto de Variedades de Yuca Sobre el Acaro Verde Mononychellus tanajoa (Bondar) (Acari: Tetranychidae). Resumenes XXIII Congreso de la Sociedad Colombiana de Entomología, 17-19 July, 1996, Cartagena, Colombia. p. 43 . Cuéllar, M. E., R. Zúñiga, T. Agrono & L. Smith. 1996. Efecto de Diferentes Dietas Sobre el Desarrollo y la Fecundidad de Cinco Especies de Acaros Predadores de la Familia Phytoseiidae. Resumenes XXIII Congreso de la Sociedad Colombiana de Entomología, 17-19 July, 1996, Cartagena, Colombia. p. 41. 150 De Moraes, G. J., E. L. Melo & M. G. C. Gondim Jr. 1997. Description ofa new species of phytoseiid mite from northeastern Brazil and redescription ofNeoseiulus gracilis (Acari: Phytoseiidae). Florida Entomologist 80(3): 319-324. Edwards, E., E. L. Melo, L. Smith & M. A. Hoy. 1997. Discrimination ofThree Typhlodromalus species (Acari: Phytoseíidae) Using Random Amplified Polymorphic DNA (RAPD) Markers. Exp. Appl. Acarol. 21 : _-_. Guerrero, J. M., A. C. Bellotti and L. Smith (presenter). 1996. Survey ofPresence of Neozygites (Zygomycetes: Entomophthorales) in Phytophagous Mites Associated with Cassava in the Neotropics. Annual Meeting ofthe Society for Invertebrate Pathology, 1-7 September, 1996, Cordoba, Spain. p. 33 . Leite, L. G., L. Smith, G. J. de Moraes & D. W. Roberts. 1996. Production ofHyphal Bodies ofthe Fungus Neozygites nr.floridana "in vitre". V SICONBIOL, Simpósio de Controle Biológico, 9-14 June, 1996, Foz do Igua~u, Brazil. Anais: Sessao de Posters. p. 448. Noronha, A. C. S. , G. J. de Moraes and L. Smith. 1995. Introdu~ao de ácaros predadores da família Phytoseiidae para o controle biológico do ácaro Brasil. Pesquisa em Andamento, CNPMA, EMBRAPA 34 (nov 95): 1-3 . Osorio, L. , M. E. Cuéllar & L. Smith. 1996. Estudios Basicos Sobre la Biología y Produccion en Colonias de Typhlodromalus tenuiscutus McMurtry & Moraes (Acarina: Phytoseiidae) Tendientes a la Optimización de un Metodo de Cria. Resumenes XXIIT Congreso de la Sociedad Colombiana de Entomología, 17-19 July, 1996, Cartagena, Colombia. p. 40. Smith L., J. S. Yaninek, G. J. de Moraes, A. C. S. Noronha. 1966. Classical Biological Control ofCassava Green Mite Mononychellus tanajoa (Bondar) in Africa and South America by Phytoseiid Mites (Acari : Tetranychidae, Phytoseiidae). Intemational Congress ofEntomology, Florence Italy. 26 Aug., 1996. p. 661. Smith, L. 1966. Modelling Population Dynamics of the Maize Weevil Sitophilus zeamais Motschulsky (Coleoptera: Curculionidae) and Its Hymenopteran Parasitoid Anisopteromalus Calandrae (Howard) (Hymenoptera: Pteromalidae). Intemational Congress ofEntomology, Florence Italy, 26 Aug., 1996. p. 561 . Smith, L. 1996. Classical Biological Control Activities at the Intemational Center of Tropical Agriculture, Cali, Colombia. V SICONBIOL, Simpósio de Controle Biológico, 9-14 June, 1996, Foz do Igua~u. Brazil. Anais: Conferencias e Palestras. p. 414-418. 151 Smith, L. and A C. Bellotti. 1996. Successful Biocontrol Projects with Emphasis on the Neotropics. Proceedings of Comell Community Conference on Biological Control, 11- 13 April, 1996, Comell University, lthaca, New York. at World Wide Web site: http ://www .nysaes. comen. edu/entlbcconf7talkslbellotti. html Smith, L., D . K. Weaver and R. T. Arbogast. 1995. Suitability ofthe Maize Weevil and Angoumois Grain Moth as Hosts for the Parasitoids Anisopteromalus ca/andrae and Pteromalus cereale/lae. Entomol. Exp. Appl. 76: 171-177. Smith, L., E. L. Melo, J. R. Escobar, and G. J. de Moraes. 1996. Biodiversity ofPhytoseiid Mites Associated with Biological Control ofCassava Spider MitesMononychellus Spp. in Northem South America [Acari: Phytoseiidae, Tetranychidae]. IOBC Intemational Conference, "Technology Transfer in Biological Control: from Research to Practice", 9-11 September, 1996, Montpellier, France. p. 268. Smith, L. , J. R. Escobar & E. L. Melo. 1996. Distribucion Geografical y Asociaciones Climaticas de Acaros Fitoseidos Asociados con Yuca en Colombia, Venezuela y Ecuador. Resumen es XXIII Congreso de la Sociedad Colombiana de Entomología, 17- 19 July, 1996, Cartagena, Colombia. p. 42. Smith, L. , M. E. Cuellar andE. L. Melo. 1996. Evaluating Prey Preference by Several Phytoseiid Predators for Mononychellus tanajoa (Bondar) and M . caribbeanae McGregor (Acari: Tetranychidae) in Cassava. Biological Control in press. Smith, L., M. E. Cuellar, & E. L. Melo. 1996. Comparative Preference of Several Phytoseiid Predators for Mononychellus tanajoa and M . caribbeanae (Acari: Tetranychidae) in Cassava. V SICONBIOL, Simpósio de Controle Biológico, 9- 14 June, 1996, Foz do Igua~u, Brazil. Anais: Sessao de Posters. p. 239. Papers at Conferences: Augmentative Biological Control in Colombia. L. Smith. Workshop on Augmentative Biological Control in the Tropical and Subtropical Americas, 14-18 August 1995, CA TIE, Turrialba, Costa Rica. Classical Biological Control Activities at the Intemational Center ofTropical Agriculture, Cali, Colombia. L. Smith. Simpósio de Controle Biológico, 9-14 June, 1996, Foz do Igua~u, Brazil. Classical Biological Control ofCassava Green Mite in Africa and Northeast Brazil. L. Smith. Simpósio de Controle Biológico, 9-14 June, 1996, Foz do Igua~u, Brazil. 152 Comparative Preference of Several Phytoseiid Predators for Mononychel/us tanajoa and M . caribbeanae (Acari: Tetranychidae) in Cassava. L. Smith, M. E. Cuellar, & E. L. Meto. SICONBIOL Simpósio de Controle Biológico, 9-14 June, 1996, Foz do lgua~, Brazil. Efecto de Diferentes Dietas Sobre el Desarrollo y la Fecundidad de Cinco Especies de Acaros Predadores de la Familia Phytoseiidae. María Elena Cuéllar, Rodrigo Zúñiga, Tomás Agrono, Lincoln Smith. SOCOLEN, Sociedad Colombiana de Entomología, 17-19 July, 1996, Cartagená, Colombia. p . 41. Efecto de Variedades de Yuca Sobre el Acaro Verde Mononychellus tanajoa (Bondar) (Acari: Tetranychidae). Elsie Gladys Burbano M., Patricia Chacón de Ulloa, Lincoln Smith. SOCOLEN, Sociedad Colombiana de Entomología, 17-19 July, 1996, Cartagena, Colombia. p. 43 . Estudios Basicos Sobre la Biología y Produccion en Colonias de Typhlodroma/us tenuiscutus McMurtry & Moraes (Acarina: Phytoseiidae) Tendientes a la Optimización de un Metodo de Cría. Liliana Osorio, Maria Elena Cuéllar, Lincoln Smith. SOCOLEN, Sociedad Colombiana de Entomología, 17-19 July, 1996, Cartagena, Colombia. p. 40. Estudios de Cepas del Hongo Entomopatógeno Neozygites sp. (Zygomycetes: Entomophthorales) Sobre Acaros de la Familia Tetranychidae. Adriana Bohórquez Chaux, Lincoln Smith, Gerardo Gallego, Joe Tohme. SOCOLEN, Sociedad Colombiana de Entomología, 17-19 July, 1996, Cartagena, Colombia. p. 97. Population Dynamics During Establishment ofPhytoseiid Mites in Biological Control of Tetranychid Mites in Cassava. L. Smith. Third Intemational Symposium on Population Dynamics ofPlant-inhabiting Mites, 26-29 June, 1995, Gilleleje, Denmark. Preferencia de especies de ácaros predadores de la familia Phytoseiidae por las presas Mononychellus tanajoa (Bondar) y Mononychellus caríbbeanae McGregor (Acari: Tetranychidae) plagas del cultivo de la yuca. L. Smith, M. E . Cuénar & E. L. Mela. SOCOLEN, Sociedad Colombiana de Entomología, Bogotá, Colombia, 27 July, 1995. Successful Biocontrol Projects with Emphasis on the Neotropics. L. Smith & A. C. Benotti (presenter). Comen Community Conference on Biological Control, 11-13 April, 1996, Comen University, Ithaca, New York. Papers at NARS Meetings: Activities at CIA T in support of classical biological control of cassava green mite in northeast Brazil. L. Smith. CNP.MF, Cruz das Almas, Brazil, 7 Nov, 1995. Presentations to Farmers and Other Agricultural Specialists: Biological control of the cassava green mi te. L. Smith. Intemational Center for Tropical Agriculture, Cali, Colombia, 1 June, 1995. Control Biológico de Insectos Plaga de Cultivos en Colombia. L. Smith. Asocaña, Cali, Colombia, 23 Oct., 1996. Articles in the Popular Press: A Research Culture is Bom. Affinities, Oct. 1996. Intemational Center for Tropical Agriculture, Cali, Colombia. Article for Cassava Newsletter, 1996. Intemational Center for Tropical Agriculture, Cali, Colombia. 153 Global collaboration and farmer empowerment for sustainable Cassava Plant Protection. In R. Markham (ed.), brochure for System-wide Project on Integrated Pest Management. 1996. CGIAR. Also at World Wide Web site: http://www.cgiar.org:80/spipm/escapp.htm. ENTOMOLOGY Caicedo, V. , A.M. and A.C. Bellotti. 1996. Reconocimiento de Nematodos Entomopatógenos Nativos Asociados con Cyrtomenus bergi Froeschner (Hemiptera: Cydnidae) en Ocho Localidades de Colombia. Rev. Col. Entomol. Vol. 22. No. l. Pp. 19-24. PATHOLOGY Alvarez, E., Loke, J. B. and R. Mayer. 1996. Progress towards biological control of Cassava Stem and Root Rots. Poster presented at the Workshop Sobre Podrid6es Radioculares no Cultivo da Mandioca. Cruz das Almas, Babia, Brasil, 1996. 154 Alvarez, E., Valencia, M., Loke, J.B. andE. Rasero. 1996. A new approach to characterize cassava root rot pathogens. Poster presented at the Workshop Sobre Pod.ridoes Radioculares no Cultivo da Mandioca. Cruz das Almas, Bahía, Brasil, 1996. VIROLOGY Calvert, L.A., Cuervo, M., Ospina, M.D., Fauquet, C., & Ramirez, B.C. 1996. Characterization of cassava common mosaic virus and a defective RN A species. Joumal of General Virology 77: 525-530. Calvert, L. A., Ospina, M. D., & Shepherd, R.J. 1995. Characterization of cassava vein mosaic virus: a distinct plant pararetrovirus. Journal ofGeneral Virology 76: 1271- 1278. Harrison, B.D., Lui, Y., Zhou, X., Robinson, D.J., Calvert, L. & Otim-Nape, G.W. 1996. Properties, differentiation and geographical distribution of geminiviruses that cause cassava mosaic. In Abstracts ofthe Third Intemational Meeting, Cassava Biotechnology Network (CBN ill) 26-31 August 1996, Kampala, Uganda, p.16. CNPMF Bento, J. M. S.,;A. C. Bellotti; J. A. Castillo; G. J. de Moraes & S. L. Lapointe. 1996. Introduction of parasitoids for control of cassava mealybug in northeastem Brazil. Fla. Entomol. (in press). Bento, J. M . S.; G. J. de Moraes; A. C. Bellotti; S. L. Lapointe & J. F. Warumby. 1996. Sucesso no Controle Biológico de Phenacoccus herreni (Hem.: Pseudococcidae) pela Introdw¡:ao de Parasitóides Exóticos (Hym.: Encyrtidae). SICONBIOL (Resumas). Foz do Iguac¡:ú-PR. Bento, J. M .. S.; A. C. Bellotti; G. J. de Moraes & J. F. Warumby. 1995 . Introdu~ao de parasitóides exóticos para o controle biológico da cochonilha da mandioca Phenacoccus herreni Cox & Williams (Homoptera: Pseudococcidae) no Nordeste do Brasil. Congresso Brasileiro de Entomología, 15, Resumas, Caxambú-MG, p. 412. Bento, J.M.S.; l. Delalibera Jr.; G. J. De moraes; A. C. Bellotti; & S. L. Lapointe. 1994. Distribuic¡:ao da cochonilha da mandioca Phenacoccus herreni Cox & Willians (Homoptera: Pseudococcidae) no Nordeste do Brasil. Congresso Brasileiro de Mandioca, 8, Resumas, Salvador-BA, p. 91. Canto, A. M. M. E. Urna viabiliza~ao do MIP: pesquisa participativa. In: V Simpósio de Controle Biológico. Anais. Foz de Igua~u-PR. p 17. 1996. 155 Canto, A. M. M. E. ; A. P . de Matos & H. P. Santos Filho. Impacto do agente biocontrolador Cladosporium sp. em organismo nao alvo na cultura da mandioca. In: V Simpósio de Controle Biológico. Anais. Foz de Igua~-PR. p. 16. 1996. Cequeira, J. H. A.; C. Fukuda; W. M-G. Fukuda & C. da S. Mendes. Valida~ao de cultivares de mandioca: urna experiencia participativa. 1 Congresso Latino Americano de Raízes Tropicais IX Congresso Brasileiro de Mandioca. Anais. Sao Pedro-SP. p. 139. 1996. Delalibera Jr., I. & G. J. de Moraes. Produ~ao " in vitro" de Neozygites sp. (Zygomycetes: Entomophthorales) patógeno de Mononychellus tanajoa. In: V Simpósio de Controle Biológico. Anais. Foz do Igua~ - PR. p.171 1996. Delalibera JR., I Produ~ao, distribui~ao geográfica e epizootiologia de Neozygites sp., patógeno de Mononychellus tanajoa. Piracicaba, 1996. 93p. (Mestrado - Escota Superior de Agricultura Luiz de Queiroz). Delalibera JR., I.; R A. Humber; J. M. S. Bento & A. P. de Matos. First record ofthe entomopathogenic fungus Neozygites fumosa on the cassava mealybug Phenacoccus herreni. Joumal oflnvertebrate Pathology (accepted for publication in September, 1996). Delalibera Jr., I; G. J. de Moraes; S. L. Lapointe; C . A. da Silva & M. A. Tamai. Variabilidade na incidencia de Neozygites SP. (Zygomycetes: Entomophthorales) em campos de mandioca. In: V Simpósio de Controle Biológico. Anais. Foz do lgua~u­ PR. p.76 1996. Delalibera Jr., I; R A. Humber; J. M. S. Bento & A. P. de Matos. Infec~ao de Neozygites fumosa (Zygomycetes: Entomophthorales) em Phenacoccus herreni (Hemiptera: Pseudococcidae). In: V Simpósio de Controle Biológico. Anais. Foz do Igua~u- PR. p. l70 1996. Eloy, A. P.; A. P. de Matos & C. Fukuda. Efeito antagónico de Trichoderma spp. sobre Phytophthora drechsleri e Scytalidium sp. "in vitro". In: V Simpósio de Controle Biológico. Anais. Foz de Igua~u-PR, p. 36. 1996. Fukuda, C. ; W. M. G. Fukuda; L. F. Cavalcanti Filho; A. P. de Matos & P. L. P. de Mattos. Avalia~ao de resistencia de cultivares e híbridos de mandioca a podridao radicular no estado da Paraiba. In: I Congresso Latino Americano de Raízes Tropicais IX Congresso Brasileiro de Mandioca. Anais. Sao Pedro-SP. p. 106. 1996. 156 Fukuda, C.; W. M. G. Fukuda; M. L. Cavalcanti & C. Lozano. Hlbridos de mandioca resistentes ao superbrotamento. In: 1 Congresso Latino Americano de Raízes Tropicais IX Congresso Brasileiro de Mandioca. Anais. Sao Pedro-SP. p. 064. 1996. Fukuda, W. M. G.; J. Cavalcanti.; S. L. de Oliveira; l. Delalibera JR. ; C. Iglesias & R C. Caldas. Efeito do estresse hídrico e do ácaro verde (Mononychellus tanajoa) sobre variedades de mandioca sob condi~s semi-áridas. In: 1 Congresso Latino Americano de Raízes Tropicais IX Co!}gresso Brasileiro de Mandioca. Anais. Sao Pedro-SP. p. 063. 1996. Matos, A P. de; S. L. Lapointe; C. Fukuda; Cerqueira, J. H. A & Ospina, B. PROFISMA: Um projeto global para o manejo integrado sustentável da cultura da mandioca. 1 Congresso Latino Americano de Raízes Tropicais IX Congresso Brasileiro de mandioca. Anais. Sao Pedro-SP. p. 116. 1996. Moraes, G. J. de; l. Delalibera Jr.; S. Elliott & C. A. D. da Silva. Natural fluctuations of populations of Neozygites cf floridana and Arnblyseius idaeus, natural enemies of Mononychellus tanajoa. In: V Simpósio de Controle Biológico. Anais. Foz do !gua~­ PR p. l43 1996. Noronha, A C. S. & G. J. Moraes. Observa~oes preliminares sobre Arnbliseius manihoti Moraes (Acari :Phytoseiidae) alimentando-se de mosca branca em mandioca. Anais da Sociedade Entomológica do Brasil (in press). Noronha, A C. S. Criayao massal de Neoseiulus californicus (Acari:Phytoseiidae). In: I Congresso Latino Americano de Raízes Tropicais IX Congresso Brasileiro de Mandioca. Anais. Sao Pedro-SP. p. 254. 1996. Noronha, A. C. S.; G. J. Moraes & A. l. Ciociola. Preferencia de Mononychellus tanajoa (Bondar, 1938) (Acari :Tetranychidae) por diferentes varieades de mandioca. Ciencia e Prática, 19(4):425-428, 1995. Noronha, A. C. S.; G. J. Moraes & L. Smith. Introduyao de ácaros predadores da família Phytoseiidae para controle biológico do ácaro verde da mandioca Mononychellus tanajoa (Bondar) no nordeste do Brasil. In: V Simpósio de Controle Biológico. Anais. Foz de Iguayu. p/ 358. 1966.