Girma Abebe, Teshale Assefa, Hussen Harrun, Tewodrose Mesfine and Abdel-Rahman M. Al-Tawaha 1

Transcription

1 World Journal of Agricultural Sciences (): 22-27, 2005 ISSN IDOSI Publications, 2005 Participatory Selection of Drought Tolerant Maize Varieties using Mother and Baby Methodology: A Case Study in the Semi Arid Zones of the Central Rift Valley of Ethiopia Girma Abebe, Teshale Assefa, Hussen Harrun, 2 Tewodrose Mesfine and Abdel-Rahman M. Al-Tawaha Melkassa Agricultural Research Center, P.O. Box 436, Nazreth, Ethiopia 2 Department of Plant Science, McGill University Macdonald Campus, 2, Lakeshore Road, Ste. Anne de Bellevue, QC H9X 3V9, Montreal, Canada Abstract: In Ethiopia, so far, maize variety selection was without much consideration of farmers' interest. However, farmers have the ability for selecting crop varieties to suit their environments. Mother and baby methodology was adopted by International Maize and Wheat Improvement Center (CIMMYT) and used as a means for obtaining farmers input and feedback on the selection of new maize varieties that are in advanced stages of development. Elite Open Pollinated Varieties (OPVs), which were under a national performance test, were used. These varieties belong to the early and medium maturity group and are suited for the dry agroecologies. The result clearly showed a significant difference among both early and medium set varieties for most agronomic traits at both Wolenchiti and Melkassa in 2003 cropping season. The results also revealed that farmers preferences in some cases coincide with the breeders selection. However, in general farmers have shown their own way of selecting a variety for their localities. Hence, it is paramount important to include farmers in a variety selection process. The methodology used in the study was found useful to collect data and identifying new varieties for farmers use. Key words: Drought Mother and Baby Open Pollinated Varieties (OPVs) Participatory INTRODUCTION decisions were only by researchers, however, this did not lead to the expected speed of variety release, or their Maize is one of the worlds three primary cereal dissemination afterwards. In addition, in developing new crops. It occupies an important position in world economy materials and extending them to farmers, classical plant and trade as a food, feed and industrial grain crop. Maize breeding faces two major obstacles. First, new varieties consumption is projected to increase by 50% globally and can be disappointing to farmers where undesirable traits by 93% in sub Saharan Africa from 995 to 2020 (IFPRI). go undetected during the breeding process. Secondly, Though much of the global increase in use of maize is for breeders necessarily discard many crosses and varieties animal feed, human consumption is increasing and during the selection process because of traits considered accounts for about 70% of all maize consumption in sub undesirable; however, these traits may actually be of Saharan Africa []. It is also one of the major crops grown interest to farmers. These illustrate the communication by small farmers in the semi-arid low rainfall areas of gap between breeders and farmers, which is also reported Ethiopia. Some reports of diagnostic surveys indicated elsewhere [3]. that 93% of the farmers in the lowlands of Ethiopia are Participatory plant breeding/selection has shown maize growers. Maize grain is used for food, for sale and success in identifying more number of preferred varieties for marketing local brewery and the stover is used for by farmers in shorter time (than the conventional system), construction, animal feed and domestic fuel [2]. in accelerating their dissemination and increasing cultivar Varietal selections in maize in Ethiopian have usually diversity [4]. Advanced lines in yield trials or at prebeen dominantly based on grain yield. Large numbers release stage [5-] or even segregating populations of of breeding lines have been developed at various later generations have been used in PPB approaches [2]. research stations and their performance evaluated across The crops included rice, beans, sorghum, pearl millet. multi-location tests over several years and only a few The level of farmer participation in PPB approaches varieties are so far identified. Varietal evaluation and is consultative/collaborative [3]. Therefore, adding Corresponding Author: Abdel-Rahman M. Al-Tawaha, Department of Plant Science, McGill University Macdonald Campus, 2, Lakeshore Road, Ste. Anne de Bellevue, QC H9X 3V9, Montreal, Canada 22

2 World J. Agric. Sci., (): 22-27, 2005 information on farmers' perspectives of plant and grain 2 = good, 3 = average, 4 = poor and 5 = very poor) for trait preferences to these criteria will be helpful to the each variety. Farmers were also asked to give an overall variety selection process. Research costs can be reduced score to each variety. Breeders have also evaluated both and adoption rates increased if the farmers are allowed to early and medium set maize mother trials for yield and participate in variety testing and selection [0]. other agronomic traits. To bridge the gap between breeders and farmers and to ensure that new varieties satisfy farmers preferences Satellite or baby trials: Baby trials were also conducted and suit their socioeconomic situations, the International on three farmers' fields at Melkassa and Wolenchiti Maize and Wheat Improvement Center (CIMMYT) is each using early and medium sets of maize varieties. developing and adapting participatory methods for Twelve varieties of maize from both early and medium identifying farmers maize variety preferences in East sets were tested and compared, i.e., ten new maize Africa. Several methods are being used: breeding on- varieties and two standards check (Katumani and ACV-6) station under stress conditions (simulating farmers (Table ). conditions); participatory rural appraisals; farmers Farmers field near to mother sites was selected for evaluation of new varieties on-station; and mother and each satellite or baby trial. Farmers were asked to select baby trials. The latter is a new approach consisting of a 5 participants among themselves to test a subset of central researcher-managed mother trial comprising all varieties from the mother trials. Each of the 5 randomly tested varieties and satellites or baby trials, which are selected farmers was given 200 g of each of ten new, farmers managed and test a subset of varieties from the varieties and the two standard checks. Farmers were mother trial [4]. requested to manage all plots equally. Overall, the Therefore, this experiment was conducted to obtain varieties tested were the same as those grown in the farmers' input and feedback on the selection of new maize mother trial and the same randomization process was varieties that are in an advance stage of development or used. ready for release using mother and baby methodology. Farmers evaluated the varieties at tasseling and at harvest. At tasseling, farmers scored each variety for MATERIALS AND METHODS earliness; at harvest varieties were scored for cob size, grain color, grain Texture, vigorousity, disease resistance Mother or central trial: Early and medium set maize and yield. Again, scores were given on a scale from mother trials were planted on three farmers field at (very good) to 5 (very poor). At both tasseling and Melkassa Research Center and Wolenchiti sub-station, in harvest, farmers were asked to give an overall assessment 2003, both locations representing semiarid area. The trial of each variety, using the same scale. consisted of 0 new maize varieties and two standard checks (Katumani and ACV-6) were compared. The RESULTS AND DISCUSSION statistical design used was RCBD with three replications. Each variety was grown with a plot size of 4 rows (with 75 Early set mother and baby trial breeders evaluation: cm between rows and 25 cm between plants) and 6 m long Table and 2 shows breeders evaluation of for each set. Two seeds were sown per hill. Recommended multilocational trial for early set mother and baby. fertilizer rate and cultural practices were done. Breeders evaluated the varieties base on yield and other Farmers were invited to the stations to evaluate the agronomic traits. The varieties revealed a distinct new varieties one week before harvest and at harvest, statistical variation in most agronomic traits and also assisted by researchers from Melkassa Research Center there was significant difference between the standard and Extension staffs from the Adama Wereda Bureau of checks and the new varieties. As it is indicated in Table Agriculture. Out of the 25 participants 6 were women. ECA-EE-36 has the highest grain yield(2245 kg ha ) and Farmers used five different colored ribbons to evaluate harvest index, but ECA-EE-9 was a variety with more varieties. The white, black, green, blue and red were used biomass but low grain yield. In other words, the combined and farmers were asked to give a general evaluation. At analysis result for early set mother trial showed that there the beginning of the evaluation farmers gathered to was significant difference among the varieties for all the discuss what they thought were the important criteria for parameters measured except grain yield at Melkassa and selecting a given variety at a particular development Wolenchiti in 2003 (Table ). However, there was no stage. These criteria were ranked and top ones were used. significant difference between the two locations Each criterion was scored from to 5 ( = very good, indicating similarity in agro-ecologies. 23

3 World J. Agric. Sci., (): 22-27, 2005 Table : Breeders evaluation of 2 new early-set maize varieties compared with standard checks (Katumani and ACV-6) in mother trials on two research stations in 2003 yield biomass height seed Harvest ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE Katumani ACV Location (A) NS NS * * * * NS NS Varieties (B) NS * * * * * * * AB NS NS * * * * NS NS CV (%) * = Significant at 0.05%, NS = Non-significant at 0.05% Table 2: Breeders evaluation of 2 new early-set maize varieties compared with standard checks (Katumani and ACV-6) in Baby trials on two research stations in 2003 yield biomass height seed Harvest ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE Katumani ACV Varieties (A) NS NS NS NS * NS NS NS CV (%) * = Significant at 0.05%, NS = Non-significant at 0.05% In early set baby trial, the same material evaluated on and other agronomic traits except 000 seed weight farmers field showed similar response as the mother trial. and harvest index at Melkassa and Wolenchiti in year The analysis result of early set baby trial showed that, 2003 (Table 3). The highest grain yield was 2489 kg ha there was no significant difference for most agronomic from variety SADVIB# with harvest index traits measured except days to 50% silking. The highest (Table 3). grain yield was obtained from ECA-EE-2 (2868 kg ha ) The medium set baby experiment result showed that, and ECA-EE-36 (2823 kg ha ) (Table 2). there was no significant difference among varieties for grain yield, total biomass, plant height and days to 50% Medium set mother and baby trial breeders evaluation: maturity. The highest yield (2409 kg ha ) was obtained Table 3 and 4 shows breeders evaluation of from variety SADVIB#. However, there is significant multilocational trial for medium set mother and baby. difference for days to anthesis, days to silking, 000 seed The medium set maize varieties also performed well in weight and harvest index (Table 4). The same variety both locations. The result of combined analysis for (SAVIB#) has shown a superior yield in the medium set medium set mother trial indicated that there was baby trial. These reveal a stable performance of varieties significant difference among varieties for grain yield across location. 24

4 World J. Agric. Sci., (): 22-27, 2005 Table 3: Breeders evaluation of 2 new medium-set maize varieties compared with standard checks (Katumani and ACV-6) in mother trials on two research stations in 2003 yield biomass height seed Harvest ECA-VL ECA-VL ECA-VL ECA-VL ECA-VL-6-TR SADVIB-# ECA-VL SADVEB-# ECA-VL A Katumani ACV Location (A) NS NS * * Varieties (B) * * * * * * NS NS AB * NS * * * * NS NS CV (%) * = Significant at 0.05%, NS = Non-significant at 0.05% Table 4: Breeders evaluation of 2 new medium-set maize varieties compared with standard checks (Katumani and ACV-6) in baby trials on two research stations in 2003 yield biomass height seed Harvest ECA-VL ECA-VL ECA-VL ECA-VL ECA-VL6-STR SADVIB-# SADVEB-# ECA-VL ECA-VL A Katumani ACV Varieties (A) NS NS NS * * NS * * CV (%) * = Significant at 0.05%, NS= Non-significant at 0.05% Farmers evaluation the varieties: The farmers who participated and evaluated in the trial were representative to the area. Most of them having long experience in farming. The evaluated farmers were most interested in some of the parameters like earliness, yield and overall appearance (vigrousity of the varieties). Ten varieties were considered to be better or comparable to the standard check. At the end of each cropping season, the farmers evaluated both the early and medium sets mother and baby trials at Wolenchiti and Melkassa, generally farmers responded positively to the new varieties they have evaluated (Table 5-7). In the mother early set materials farmers' assessment indicated that there was a matching with researchers need (based on grain yield and other agronomic parameters). However, in most cases farmers have expressed their will different from the breeder. In other words, the results also show that farmers are eager to participate in selecting new varieties but their selection is very different from that of breeders. Moreover, farmers evaluations and testing on-farm both show variance. Table 7 Farmers overall evaluation indicate that variety ECA-EE-36 and ECA-VL-7 were the best from early and medium set, respectively. The standard checks have showed poor performance in most cases. In agreement with De Groote et al. [4], there were a growing interest among farmers in the use of early maize varieties in short rainfall season, farmers have shown 25

5 World J. Agric. Sci., (): 22-27, 2005 Table 5: Early sets of maize varieties selection criteria used by farmers and their assessment in two locations in central rift valley of Ethiopia, ranked in order of importance Ranking Melkassa farmers Wolenchiti farmers of farmers (average Grain Grain Cob Grain Grain Cob score) color texture size Yield Earliness Vigorousity Disease color texture size Yield Earliness Vigorousity Disease ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE ECA-EE Katumani ACV Table 6: Medium sets of maize varieties selection criteria used by farmers and their assessment in two locations in central rift valley of Ethiopia, ranked in order of importance. Ranking Melkassa farmers Wolenchiti farmers of farmers (average Grain Grain Cob Grain Grain Cob score) color texture size Yield Earliness Vigorousity Disease color texture size Yield Earliness Vigorousity Disease ECA-VL ECA-VL ECA-VL ECA-VL ECA-VL-6-STR SADVIB-# ECA-VL SADVEB-# ECA-VL A Katumani ACV Overall Score: (-5) Scale; = Very Good, 2 = Good, 3 = Average, 4 = Poor and 5 = Very Poor Table 7: Ranking of early and medium set varieties according to farmers and breeders Early set Breeders Farmers Medium set Breeders Farmers varieties rank rank varieties rank rank ECA-EE-36 ECA-VL-7 7 ECA-EE ECA-VL ECA-EE ECA-VL ECA-EE ECA-VL- 3 4 ECA-EE-49 5 ECA-VL-6-STR 9 9 ECA-EE SADVIB-# 5 ECA-EE-45 0 ECA-VL-2 4 ECA-EE SADVEB-# 6 2 ECA-EE ECA-VL ECA-EE A-5 2 Katumani 2 9 Katumani 2 7 ACV ACV strong interest. Some farmers were willing in adopting the varieties since they were confident of their high yield and earliness. The methodology used in this study to identify farmers criteria and to facilitate farmer evaluation of the varieties was very convenient for data collection but not for data analysis [4]. For future trials, sufficient resources need to be made available to assure enough high quality data for statistical analysis. These data will make it possible to adjust the breeders index in order to make it more responsive to farmers preferences. In some cases farmers preference coincide with breeders selection. However, in general farmers have their own way of selecting a variety for their localities. Hence, it is paramount important to include formers in a variety section process. As it has been demonstrated by farmers a new and promising varieties were identified using mother and baby methodology and this method contributed a lot to facilitate variety selection based on farmers preference. However, some of the data s were inconsistent and difficult to interpret. Most of the varieties consistently outscored the Standard checks. Poor correlation between farmers and breeders evaluations in most cases was also observed. Mother and baby methodology clearly showed how classical breeding has problems responding to farmers variety preferences. Similarly to this finding, 26

6 World J. Agric. Sci., (): 22-27, 2005 De Groote et al. [4] stated that, scientists like to control 4. Weltzien, E., M.E. Smith, L.S. Meitzner and L. for many factors so that they can accurately state that, Sperling Technical and institutional issues under their very controlled circumstances, a limited in participatory plant breeding-from the perspective number of traits have improved. A problem here is that of formalplant breeding. A global analysis of these highly controlled circumstances are not often issues, results, and current experience. PPB representative of farmers conditions and the limited Monograph No.. Cali: PRGA number of traits might not represent farmers preferences. 5. Maurya, D.M., A. Bottrall and J. Farrington, 988. This is highlighted by the very poor correlation between Improved livelihoods, genetic diversity and farmer farmers and breeders evaluations in the study. participation: A strategy for rice breeding in rainfed The basic system of asking farmers to define their areas of India. Experimental Agriculture, 24: selection criteria and to score new varieties on a scale of 6. Sperling, L., M.E. Loevinsohn and B. Ntabomvura, to 5, was very convenient for data collection but 993. Rethinking the farmers role in plant breeding: cumbersome for analysis. Increasing the number of Local bean experts and on-station selection in farmers can also reduce the variability associated with Rwanda. Experimental Agriculture, 29: farmers evaluation on-station or in mother trials. The 7. Mekbib, F., 997. Farmer participation in common baby trials lots of data were lost due to bad weather and bean genotype evaluation: The case of eastern poor organization resulting high variability. The Ethiopia. Experimental Agriculture, 33: experience indicates that enough resources have to be 8. Baidu-Forson, J., 997. On-station farmer made available to allow for regular site visits to assure the participatory varietal evaluation: A strategy for quality and quantity of data collected and rapid data entry client-oriented breeding. Experimental Agriculture, and analysis. The experience revealed that farmers are 33: eager to participate in selecting new varieties. The 9. Witcombe, J., R. Petre, S. Jones and A. Joshi, 999. methodology still needs work, but it clearly demonstrates Farmer participatory crop improvement. IV. The the potential for bringing together breeders and farmers spread and impact of a rice variety identified by selection indices. participatory varietal selection. Experimental Agriculture, 35: ACKNOWLEDGMENTS 0. Joshi, K.D., M. Subedi, R.B. Rana, K.B. Kadayat and B.R. Sthapit, 997. Enhancing on-farm varietal We are grateful to International Maize and Wheat diversity through participatory varietal selection: Improvement Center (CIMMYT) and Ethiopian A case study for chaite rice in Nepal. Agricultural Research Organization for providing fund for Experimental Agriculture, 33: this work. A special thanks also goes to Wolenchiti and. Mulatu, E. and K. Belete, 200. Participatory variety Melkassa farmers who participated in the field varietal selection in lowland sorghum in eastern Ethiopia: evaluation. We also thank Melkassa Research Center Impact on adoption and genetic diversity. technical staffs namely Ato Alelegne Assefa, Ato Bahiru Experimental Agriculture, 37: Tilahun and Ato Legesse Teshome for the field data 2. Sthapit, B.R., K.D. Joshi and J.R. Witcombe, 996. collection and interviewing of farmers. Farmer participatory crop improvement. III. Participatory plant breeding: A case study for rice in REFERENCES Nepal. Experimental Agriculture, 32: Farrington, J. and A.M. Martin, 988. Farmer. Aquino. P., F. Carrion, R. Calvo and D. Flores, 200. participatory research - A review of concepts and Selected maize statistics. In: Pingali, P.L. (Ed.) practices (2). ODI Agricultural Administration Unit CIMMYT World Maize Facts and Trends. Occasional Paper 9. (#02). Meeting World Maize Needs: Technological 4. De Groote. H., M. Siambi, D. Friesen and A. Diallo, Opportunities and Priorities for the Public Sector Identifying Farmers Preferences for new Mexico. D-F. CIMMYT. Maize Varieties in Eastern Africa. In Bellon, M.R. and 2. Tilahun Mulatu and Teshome Regassa, 987. Nazret J. Reeves (Eds.) Quantitative Analysis of Data mixed farming diagnostic survey report: Shewa from Participatory Methods in Plant Breeding. Region. Research Report No. 2, IAR, Addis Ababa. Mexico, DF: CIMMYT. 3. Kamara, A., T. Defoer and H. de Groote, 996. Selection of new varieties through participatory research, the case of corn in South Mali. Tropicultura (September 996)4: