|
|
| English - Français |
|
 |
| IDRC.CA >
Publicaciones >
Libros >
Todo nuestros libros >
USING DIVERSITY >
|
Identificación: 85277Creado: 2005-07-21 10:05Modificado: 2005-07-21 10:06Refreshed: 2012-02-10 21:35 
|
|
Documento(s) 27 de 38
Abstract: High altitude rice is constrained by chilling injury and Sheath Brown Rot (ShBR) diseases. The modified-bulk method of breeding supplemented by a farmer participatory breeding (FPB) is adapted to efficiently utilize limited resources, and farmers' expertise to provide selection choice. This study has been conducted by the Lumle Agricultural Research Centre (LARC) at Chhomrong (2000 m) and Ghandruk (2000 m) villages in the Western Hills of Nepal since 1993 with the objective of developing productive varieties adapted to local needs and preferences, and also enhancing genetic diversity on-farm. Selection of segregating lines in farmers' fields has been encouraged and their selected lines are multiplied and entered into the formal system to ease official release. The reasons for adopting an innovative approach as compared to a conventional method are described. Promising F5 bulk seed of selected lines was distributed to expert rice farmers for decentralized selection. Selection procedures, evaluation criteria of farmers and factors affecting their decisions were monitored. Selection methods of farmers which successfully resulted in the choice of different rice entries are discussed in this paper. Results were assessed by joint field visits, preference ranking, post-harvest evaluations, crop cut yield data and on-station trial data. Final adoption decisions were based upon multi-stage evaluations, including post-harvest by women farmers. Farmers selected Machhapuchhre-2, 3 and 4 from the Fuji 102 x Chhomrong Dhan and Himchuli-2 from K332 x NR10157-2B-2 cross. Two populations of Machhapuchhre-3, selected independently by farmers of Chhomrong and Ghandruk, have promoted genetic diversity in farmers' fields. Both populations are the most preferred by farmers and are spreading quickly. Farmer-selected varieties also performed better in researcher-managed yield trials, with better resistance to ShBR disease and chilling. Preference ranking by farmers indicated that there is good agreement between farmers and breeders suggesting that expert farmers' selection is as reliable as breeders'. The success of this kind of program depends upon how well expert farmers can be identified and whether the problem is relevant to the farming community. IntroductionChilling injury in rice is common in Nepal in high altitude areas (i.e. >1000 masl). Of 1.4 million hectares of rice in Nepal, 26% are grown in temperate areas (1000-2000 m) (Shahi and Heu, 1979), and spikelet sterility caused by chilling injury is a major constraint above 1500 m, limiting both the area of production and the length of the growing season (Sthapit and Shrestha, 1991). In Nepal, rice improvement programs have been less successful in high altitude areas to which limited research resources have been allocated by national and international programs. Of the 39 rice cultivars recommended so far by the National Rice Research Program (NRRP) only two, Chhomrong Dhan (CD) and Palung-2, have been released as chilling tolerant cultivars suitable for the high hills (>1500 m). Screening of internationally known cold tolerant materials at Lumle (1450 m) and Chhomrong (2000 m) has identified lines with good chilling tolerance during vegetative growth, but which failed to produce grain because of incomplete panicle exsertion or spikelet sterility.Nepal is rich in genetic diversity of chilling tolerant rice landraces (Nagamine, 1992) but the identification and utilization of local germplasm within the national breeding program has been limited (Sthapit, 1992). Since 1985, Lumle Agricultural Research Centre (LARC) has conducted a high altitude rice improvement program by evaluating local and exotic germplasm in farmers' fields at Chhomrong (2000 m). As a result, in 1991, Chhomrong Dhan (CD) was released by National Variety Release and Registration Committee (VRRC) for high hills of Nepal. Only CD shows reliable adaptation and good performance above 1500 m. About twenty-five years ago, this variety was introduced from India by a curious farmer at Ghandruk and Chhomrong (>1900 m) where rice was never grown before. Pure line selection of the local population for chilling tolerance and ShBR resistance was done by LARC. The chilling tolerant rice participatory breeding program commenced with farmers in Chhomrong and Ghandruk in the rainy season of 1993. The idea to start such program sprang from discussions during a field visit to Chhomrong by Dr. J.R. Witcombe and LARC rice breeders in October 1992. The reasons for adopting such an approach are many:
Varietal selection is difficult for very heterogenous environments where farmers have a range of preferences and circumstances. A combination of traits is required for adaptation and various socioeconomic criteria need to be considered to achieve adoption of a variety. This requires a research system that will deliver a wide range of rice varieties suitable for the diverse agro-ecological and socio-economics circumstances of farmers. Use of conventional selection approaches, concentrating on grain yield and wide adaptability, is not always possible in developing countries where research resources and trained manpower are limited. This necessitated the re-orientation of LARC's high altitude rice improvement methodology to address two issues; resource availability and research appropriateness. Maurya et al (1988), Galt (1989), Joshi and Sthapit, (1990), Sperling et al (1993) and Sthapit et al (1994) considered alternative approaches to address these problems. Traditionally, the development of rice varieties has been a task of breeders, and farmers are supposed to wait for finished products. However, at LARC, we thought that a joint venture between expert farmers and crop breeders for developing new varieties could be fruitful in terms of incorporating farmers' wisdom in identifying appropriate technology for their environment. The objective of the program has been to examine the practicability and success of adopting a farmer participatory approach to high altitude rice breeding as a means of minimizing resource use, utilizing farmers' knowledge, developing suitable varieties and enhancing the diversity of rice gene pools. Methods of the participatory varietal selectionIn 1985, LARC started a high altitude breeding program with the objective of developing cold tolerant white grain rice with resistance to blast and ShBR diseases. An example of the scheme of this farmer participatory rice breeding program is given in Table 1. The approach emphasized farmers' selection of segregating materials from carefully chosen crosses from the on-going high altitude breeding program. An evolutionary approach was taken for the development of practical methods by using materials generated by the conventional modified bulk population method. There were no established FPB methods known to LARC breeders at the time which could have helped in the design of procedures that were practical as well as scientific enough to produce reliable results. There was an important element of risk.Identification of siteHigh selection pressure was employed at Chhomrong (2000 m) for spikelet fertility and ShBR resistance during F2-F4 generations, while retaining variability in which farmers were interested. Chhomrong (2000 m) village had been a high altitude rice testing site since 1985 and was considered a good site for natural screening of both cold air and water-induced spikelet sterility and ShBR disease caused by Pseudomonas fuscovagainae (Sthapit, 1992). This site was also selected because farmers were interested in improving the quality of their local rice and had been impressed by LARC's past farmer participatory research.
Identification of segregating materialsTwo F4 lines from the formal breeding program nursery were selected by the breeder for this study in 1992, and F5 bulk rows were harvested. The selected F5 bulk lines, from three crosses, were given local names for easy identification in discussions with farmers (Table 2). F5 materials have significant genetic variation left in the bulk from which farmers can choose genotypes to meet their own needs and circumstances.In 1993, a limited quantity (20-25 g) of seeds of F5 bulk were distributed to expert farmers2 (Table 2). Additional F5 bulk of Nilgiri-1, Himchuli-2 and Machhapuchhre-6 and M-7 was included in 1994. A total of ten farmers from Chhomrong and four from Ghandruk of Ghandruk Village Development Committee (VDC) were involved in the program. The number of farmers was limited due to the limitation of seed, but whenever possible, extra seed was distributed to farmers of adjoining or similar recommendation domains to test the extent of diffusion (Table 2). Table 3 gives the origins, species, sources and specific traits of the parents of the three crosses under study.
The identification of farmer-breedersA total of 14 expert farmers from Chomrong and Ghandruk villages, who had considerable knowledge and skill in rice farming, and who were willing to participate in this study, were identified with the help of the local community. Four additional cooperating farmers from Lumle and Sabet villages were also selected, but FPB procedures were not followed.Both male and female members of the same household were involved in this study to accommodate differences in gender knowledge and perceptions. Wives of male cooperating farmers were automatically selected for post-harvest evaluations. They are usually expert in evaluating harvesting, seed selection, milling, storage and cooking characteristics, whereas male farmers are more astute evaluators of threshing and yield potential. Farmer participationIn the beginning (i.e. 1992), farmers were not very enthusiastic about the program and they were hesitant to participate in the conventional labor-intensive, researcher-managed trials to which they were so accustomed. However, farmers' participation increased as they perceived that the research was relevant to them. In October 1992, a small gathering of about 20 rice growers was organized to discuss the objectives of the program. The purpose was to develop rice varieties with white grain. White rice is highly preferred by the local community, and therefore, was used as a catching phrase to encourage them to get involved in rice breeding. At the meeting, the role of participating farmers and their knowledge of plant types and heritability of traits was also discussed.Farmers' managementThe farmers were asked to grow and manage the lines along their normal practices except that test entries were maintained separately in the field and store to avoid mixing. Farmers were allowed to assess and select as per their own criteria and objectives and the procedures were monitored and recorded by field staff. Seeding for nurseries and transplanting for all test and local varieties were done within a week. Sign-boards with the name of the variety and expert farmer were provided by the program to catch the attention of non-participating farmers.Training in plant selectionIn 1993, farmers were asked to make selections in the field. At the beginning of the selection, farmers were informed that progenies would segregate for grain color, plant height, maturity, etc., when two divergent varieties are crossed, and therefore, selection for desired traits should be done for two to three years from F5 bulk seed until the trait is fixed. Breeders' knowledge of genetics and heritability was offered to farmers in order to complement farmers' indigenous knowledge of diverse environments. Farmers were encouraged to carry out their own selection, which enhances gene diversity at the farm-level. Farmers were asked to keep half of seed to grow the next year and to return half of the selected seed to LARC for on-station varietal testing.Farmers' preference rankingIn 1993, farmers plots were visited by a group of participating farmers and breeders but no preference ranking was done. In 1994, the on-farm and experimental plots at Chhomrong and Ghandruk villages were jointly monitored by participating farmers and other farmers from the villages with breeders, a socio-economist and affiliated researchers. Each farmer's plot was labelled with the farmer's name. At the outset, the breeder explained the purpose of the farmers' field day.A total of 14 farmers from the village of 20 households visited individual plots, examining crop performance, inherent soil fertility and water sources. Representatives from six household could not participate in the field day, which took about three hours. After the field visit, farmers were individually asked to rank varieties from 1 (for excellent) to 7 (for worst) on the basis of their own criteria and knowledge. With the help of researchers, farmers listed the salient positive and negative characteristics of each variety using an open-format questionnaire. The extent of agreement among farmers, and also between male and female farmers based upon preference ranking, was also assessed by Kendall's W (Siegel, 1956), a measure of the concordance among multiple judges. Rank correlation was also done to measure agreement between breeders and farmers using Spearman's coefficient of rank correlation (Steel and Torrie, 1960). On-farm yield measurementCrop cuts from 1 m plots were taken from fields of all participating farmers to measure grain yield. Measures were adjusted to account for 12% moisture content. Plant height (cm) and fresh straw yield at harvest were also recorded. At harvest time, the farmer's method of plant selection was observed.On-station yield trialsIn 1994, the farmer-selected variety Machhapuchhre-3 was included in the National Rice Cold Tolerance Nursery (NRCTN)3 which had been selected by Rudra Bahadur Gurung from Ghandruk-6 in 1993, and was considered best from the point of view of yield potential and resistance to cold and diseases. The objective of this was to assess the performance of the farmer-selected variety in comparison to varieties developed through the centralized breeding program and to verify the unconventional on-farm results with on-station data in order to satisfy the variety release regulations. A broader objective was to influence research policy itself.The NRCTN was conducted at four sites during the summer of 1994: Chhomrong (2000 m) and Shera (1250 m), testing sites of LARC; Khumaltar Agricultural Farm (1350 m), and Kavre Agricultural Farm (1700 m). At LARC, twenty varieties supplied by NARC along with the farmer-selected M-3G, eight progenies of Fuji 102/CD and three local checks were assessed for variation in chilling tolerance and yield potential, using the standard evaluation system for rice (IRRI, 1988). Trials were designed in randomized complete block design (RCBD) with two replicates at Shera and Chhomrong, whereas in Kavre and Khumaltar nurseries were non-replicated with twenty entries. The trials were researcher-managed with standard recommended agronomic practices. The nurseries were transplanted during mid-July 1994 with the application of 60-30-20 Kg NPK/ha except at Khumaltar where chemical fertilizer was applied 100-40-30 kg NPK/ha. Plot size varied from 1.2 to 4m. Data on agronomic traits and disease incidence related to chilling tolerance were measured. Yield data were not recorded in Khumaltar. Analyses of variance were used in replicated trials to evaluate the significance of differences between varieties for various traits using a randomized block design. Post-harvest evaluations by women farmersPost-harvest evaluation was done three months after the harvest. In 1994 season, this study was carried out only with those varieties for which farmers had retained seed for the next season. The questionnaire was designed after discussions with women farmers known for their seed selection skills, cooking and milling practices. Farmers milled and cooked the rice along local methods and their assessments were recorded. A total of seven questionnaires on post-harvest evaluations were completed.Monitoring of varietal spread and rejectionVarietal spread was monitored with all cooperating farmers during 1993, 1994 and 1995. The reasons for adoption or rejection were also recorded.ResultsFarmers' selection of siteThe majority of participating farmers grew their crop in medium fertility conditions at altitudes ranging from 1400 m to 2000 m. Four cases illustrate how farmers' decisions influenced site selection for variety evaluation. Him Bahadur Gurung from Chhomrong (2000 m) transplanted Himchuli-2 rice to a small plot where cold water first gets in from the channel and which used to cause severe tip sterility in his local rice. Him Bahadur aimed for risk aversion by experimenting in the worst plot for cold water tolerance. He has no plan to continue with Himchuli-2 as the performance of the variety was not good. Similarly, Rudra Bahadur Gurung from Ghandruk (2000 m) also tried M-3 first in the worst plot of his land which was situated at the corner of a high terrace which received little sunlight and was exposed to cold water. He selected a few good plants from the plot and has expanded the area sown to the variety from 6 to 1250 m over two years of testing. As an extreme case, Jhanan Nath Devkota of Lumle (1400 m) transplanted Nilgiri-1 to a plot where he never had seen a rice plant produce a single grain. The plot has an inlet of cold water from the mountain stream, which also brings animal manure to the terrace. It has been well-established that cold water in presence of high fertility induces high spikelet sterility (Sthapit, 1992). The variety yielded 3.3 t/ha at 12% moisture content estimated from a 2 m crop cut. Mr. Devkota was excited at this unexpected result and immediately decided to expand use of the variety to better lands and over larger areas. However, at planting time, he changed his mind because of the high shattering, poor straw quality and taste. Another farmer, Dhananjay Devkota of Lumle, evaluated the variety under high fertility conditions to see whether Himchuli-2 (4.0 t/ha) could stand the conditions. He has 0.4 ha of highly fertile rice fields and local varieties such as Raksali (2.9 t/ha) and Kathe (2.0 t/ha), which are adapted at Lumle, are partially sterile under his conditions. Mankaji Gurung of Ghandruk (2000 m), experimented with M-1 under rainfed upland conditions as he has larger areas falling into this category.The farmers' strategies illustrate that varietal selection under uniform on-station conditions cannot represent the very heterogeneous environments used by farmers. Decentralized selections of site, management, and variety have helped to generate useful information on the specific requirements of farmers--information which formal might find costly to access. Crop PerformanceThe results are based on the field performance of entries given to participating farmers in 1993 and 1994 in Chhomrong, Ghandruk and neighboring villages. In 1993, most entries included under the participatory program were cold tolerant; however, the performance of Himchuli-1 was not very satisfactory due to high panicle sterility and degeneration. Among Machhapuchhre lines, M-1 was very dwarf and was more sterile than Chhomrong Dhan. Farmers dropped M-1 and M-5 in the first year of testing. M-5 performed poorly and was unattractive to farmers. Performance of M-2 was average and farmers gave it another chance for evaluation. Over one year, testing of M-3 and M-4 appeared promising and farmers retained seed, including that of M-2, to verify further in the 1994 season.In the 1994 season, the comparative performance of rice varieties was judged by preference ranking of groups of male and female farmers, and breeders. The overall performance of M-3 was best, followed by M-2 and M-4. Himchuli-2 was ranked worst (Table 4) because of high sterility and small panicle size.
The evaluation scores among men and women farmers in Chhomrong village showed significant agreement (Table 4). Agreement between breeders was higher than amongst farmers but this could be due to the small number of breeder judges. Agreement between breeders and farmers was high (0.82) which suggests that farmers participation in selection is as reliable as breeders, if expert farmers are carefully chosen. Out of 14 farmers who participated in the program, two farmers at Ghandruk and three at Chhomrong retained seeds from the test lines (Table 2) for further testing and evaluation. Farmer-managed trial yield dataTable 5 shows grain yields and other agronomic traits measured from the 1 m on-farm plots where farmers carried out their preference ranking. These figures cannot be statistically analyzed and interpreted as the data were based upon a single crop cut of small plots. However, there was good agreement between farmers perception of the variety and crop cut results. The most preferred farmer-selected variety, M-3, had a higher yield than the Chhomrong local. The local variety produced 4.1-6.4 t/ha in 1994 from crop cuts of 10 adjacent fields. Yield levels of the other test varieties were lower than the local. Aside from grain yield, the plant height and straw yield of M-3 was also found superior to the locals.
Farmers' methods of plant selectionFarmers used various criteria to select the best plants and gave some of their harvested seed to breeders for comparison with formally-screened varieties. There was variation in the methods of plant selection amongst farmers. At maturity, some farmers used grain color without consideration of plant height and maturity, a process which results in an unimpressive crop when working with segregating plants. Some farmers only bulked seeds of those plants which were similar in plant height, maturity time and grain color. Performance of these varieties was remarkably different in the field. Apart from grain color, farmers' criteria for selecting plants were long, compact and drooping panicles, good grain setting, and freedom from ShBR disease.For example, Mim Bahadur Gurung of Chhomrong village selected intermediate plants from M-3 which had compact panicles and filled grains. He selected plants which still had their green flag leaves after maturity. He thought the straw quality of those plants would be better. He also compared the maturity of the plant with the local, grown alongside. Plants with medium height were selected by one farmer, Purna Bahadur Gurung from Ghandruk, who did negative selection of undesirable plants first: discarding either dwarf or too tall plants and diseased plants. He then selected panicles from white colored hills with high grain density and long panicles. He is a retired livestock field staff of LARC who has some knowledge of selection and heritability. Another farmer, Rudra Bahadur Gurung from Ghandruk, tested the M-3 in his worst plots and harvested the best tall plants; he then bulked seed to be grown in 1994 in his best field. In summary, the major criteria for selecting the plants in the field were mainly grain color, yield potential, plant height, and maturity. Besides grain yield, farmers considered how densely the grain had set in a panicle, panicle length and tillering ability. The overall phenotype of any plant hill was considered for selecting or rejecting the entries. Farmers showed a willingness to spare time and care in selection, though the relative degree of care and time given by farmers were significantly higher in the second year of participation. Monitoring of spread and biodiversityGood varieties selected by farmers have spread within two years of introduction (Table 6). For example, Mim Bahadur Gurung of Chhomrong village got half of the seed from his friend Jai Bahadur Gurung, to whom LARC scientists originally gave the seed in 1993. Jai Bahadur Gurung could not select superior types from the same seed lot, whereas Mim Bahadur Gurung has selected M-3 (identified as M-3C) and has increased the area under it from 3 m to 150 m within a year. His plot was excellent as he put more care into it. This indicates that farmers' expertise in selection varies. Similarly, Rudra Bahadur Gurung from Ghandruk also selected M-3 (identified as M-3G), increased the area from 6 m to 50 m in 1994, and has sown a nursery for 1995 of about 1250 m. The variety occupied 2.5% of his total rice fields after first year of selection and 62.5% in the second--a significant rate of spread. Farmers of both villages come from the same ethnic group and agro-ecological background yet selected two different promising varieties, M-3C and M-3G, from the same F5 bulk seed. This indicates that this F5 bulk seed of Fuji 102 x CD cross has important genetic variations which farmers thought important to meet their particular needs. Therefore, they selected two populations independently. This type of decentralized selection in breeding lines has allowed farmers, who have grown only Chhomrong Dhan for the last 25-30 years, to maintain biodiversity in their own fields .
Farmers' perceptionsTable 7 summarizes some important comments made during field visits and farmers gatherings. These comments typically reflect the various levels of satisfaction found amongst participating farmers. Pitamber Shrestha, field-based staff at Chhomrong, shared his experience with his fellow research team members as follows:"In the beginning it was very difficult to find a participatory breeder-farmer, but now, the whole community is willing to offer their help. It is a nice feeling."
Post-harvest assessments by women farmersWomen farmers reported that they would like to make their decision on variety selection after the post- harvest evaluation. Table 8 shows that farmers used several qualitative criteria to assess the quality of rice. Varieties M-6 and M-7 had more negative attributes than M-2, M-3 and M-4. As a result, farmers showed willingness to expand the area under M-3, M-2 and M-4. In Chhomrong, consumers preferred white-grained rice over red-pericarped rice as it saves women time in milling ( See Table 7, comment 2). A dry-flaky cooked rice goes well with curry and pulse soup as compared to a soggy rice. These traits associated with aroma, softness and ability to expand after cooking are the most desirable traits of good quality rice in Nepal. The post-harvest evaluation shows the ultimate criteria upon which farmers either reject or adopt the varieties. For example, Jhanan Nath Devkota from Lumle village was so impressed with Nilgiri-1 rice at first but refused to expand the area beyond his problem plots because of poor taste (Table 6).
In general, the milling percentage for all rice varieties was relatively poor, ranging from 40 to 56% by volume, with 1.3 to 5% broken rice. Farmers considered 50% milling recovery acceptable. M-4 had the highest milling recovery (56%) with the minimum percentage of broken rice. Purna Bahadur Gurung of Ghandruk was so impressed with the milling recovery of M-4 that he planted all seed which was not selected before. The experience of Chij Bahadur Gurung suggested that the variety M-4 was easier and quicker to mill by Dhiki (the local method of dehusking which requires two persons, normally women) as it has a very thin husk cover. The milling recovery of other cultivars was within the range of the local variety CD. It is interesting to note that the best panicles selected by farmers were kept for seed whereas the grains from the late tillers and small panicles were used for post-harvest evaluations. As a result, the milling percentage amongst the tested varieties was relatively low with a high percentage of broken rice. Therefore, farmers believed that the variety with a similar milling percentage to the check variety should have a better recovery if average or only good panicles were selected. Farmers showed reluctance to continue with the M-7 variety as its milling recovery was quite poor (40%). M-2, which performed at average in the field, was rejected by farmers because of the peculiar smell of the cooked rice. On-station yield trialThe results of the on-station yield trial are presented in Table 9. At the Chhomrong site (2000 m), though M-3 performed at par with the check variety Chhomrong Dhan (p0.05), M-3 was superior in terms of white grain quality. Both varieties showed good resistance to ShBR disease and chilling at anthesis. At Kavre site (1700 m), IR 52423-B-3-3-1-3 produced the highest grain yield (7.1 t/ha with poor straw yield) followed by progenies of Fuji 102/Chhomrong Dhan i.e. LR 88001-7L-0L-0L and LR 88001-21C-0L (6.8 t/ha) and M-3 (selected by farmers)(6.7 t/ha). The farmer-selected M-3 variety outyielded the check Chhomrong Dhan (5.4 t/ha)(p>0.05). It also outyielded other superior local varieties known for better cold tolerance in Nepal. M-3 also had comparable incidence of ShBR and neck blast under natural inoculum pressure as compared to most breeder-selected varieties.
At Khumaltar site (1350 m), study of agronomic traits for the farmer-selected variety M-3 scored 1 for chilling tolerance and phenotypic acceptability (data not shown). According to the standard evaluation system, rice varieties receiving a rating between 1 and 3 are considered desirable for parental sources and for commercial varieties (IRRI, 1988). At Shera site (1250 m), the yield performance of the farmer- selected variety was at par with the Chhomrong check and was found superior to other white grained local cultivars. DiscussionResults indicate that there are significant differences between farmers' and researchers' strategies for site selection for testing. It is a common practice in formal research system to select better and uniform pieces of fields for trials rather than to select representative target environments-- for which varieties are bred and screened. In contrast, farmers test the materials first under their worst land (stressed environments) for which they need a solution and then spread the variety to better fields. Ceccarelli (1989) has suggested that when the stress environment has a much lower yield potential, direct selection in the environment is the most efficient breeding strategy. The results of this decentralized selection of segregating breeding lines in farmers' heterogeneous environments not only identifies location-specific varieties but also eliminates the risk of releasing variety which may later be unacceptable to farmers.Many workers have advocated decentralization of research policy (Maurya et al, 1988; Farrington, 1988; Galt, 1989; Tripp, 1989; Joshi and Sthapit, 1990; Sperling et al, 1993; Sthapit et al, 1994). It is well documented that the formal research system in developing countries is highly centralized and does not reflect the problems of resource-poor farmers. Poor adoption of officially released rice varieties in India (Maurya and Bottrall, 1987) and Nepal (LARC, 1995; Chemjong et al., 1995) provides evidence of this. However, it has not yet been shown that farmer-selected varieties can be equally good in terms of yield and other desirable traits if breeder and farmer work jointly, using the formal varietal testing system. One of the farmer-selected rice varieties, M-3G, which had been selected initially by breeders for chilling tolerance and ShBR resistance by pedigree-bulk method, is now being multiplied and included in the 1994 national varietal testing system of the NARC. It is being promoted to IET as a promising entry. M-3 matches the characteristics of the local varieties but has better grain quality and chilling tolerance, and has been preferred by the farmers of Chhomrong and Ghandruk. The area covered with the variety selected by expert farmers has been quite high; there have been promising results within a short period of three years. Morris et al., (1992) has argued that the rate of adoption has long lag period, usually 5-6 years after the release of a variety. It is yet to be seen whether this variety will spread from farmer-to-farmer amongst a group of roughly homogenous farmers with similar circumstances. Results indicate that farmers wish to fulfil their requirements. However, some farmers have already started distributing seed to other farmers. Interestingly, the performance of the farmer-selected variety matched on-farm crop cut data and results of researcher-managed on-station trials in diverse altitudes. With our experience in Chhomrong and Ghandruk, we anticipate that the variety will be widely spread among similar areas by the time the variety is released through the formal system due to the early farmer-to-farmer spread of the preferred variety. The slowness of the formal procedures and the system's capacity to generate only a few new varieties every year-- juxtaposed to the need of a basket of varieties to address farmers' needs in diverse and complex farming systems-- has encouraged the LARC to involve farmers at early stage of breeding (with a few carefully selected crosses). With the present FPB system, farmers are exposed to new material six to seven years earlier than in a conventional system, and the proposal for official release can be submitted three years earlier, even taking into account time needed to select for uniformity in farmers' cultivars. This approach in Nepal, therefore, will allow farmers to have increased access to breeding materials several years before normal varietal release, when the variety still has its full genetic potential (Galt, 1989), and its spill over effect will be greater than those released late by the formal system. In the past, a large number of exotic cold tolerant rice varieties supplied through IRCTN and NRCTN were evaluated at Chhomrong and the majority of them failed to set grain in high altitude villages (Sthapit, 1992). In that context, these varieties developed jointly with farmers are far superior and are already spreading through farmer-to-farmer seed exchange. This has been possible not only due to the FPB system but also due to the identification and utilization of locally-selected Chhomrong Dhan as a cold tolerant donor, which was selected from a range of local germplasm. Hence, use of indigenous germplasm and farmers' wisdom in crop cultivation, supplemented by breeders' knowledge of heritability and genetics is turning out to be fruitful in selecting the right plant type to produce a population which is superior to the local cultivar. Utilization of genetic resources to meet farmers' production needs is a realistic way to encourage in situ conservation as indigenous genes are recombined with other useful gene pools within local environments. Though the concept is new, it is thought to be resource-efficient and problem-oriented as the role of farmers is not merely limited to providing land and supplying labor but is aimed at directly decentralizing the variety evaluation process. Breeding is not often the favored option for research managers of short-term research organization like LARC, but now they have shown interest in farmer participatory breeding due its low cost and its rapid generation of an appropriate variety. Farmers' methods of plant selection varied with the farmers' own expert knowledge and circumstances. Not all farmers were good at this exercise. Segregating lines of Machhapuchhre-3, which all had desired genetic recombinations, were given to three farmers, but only two farmers succeeded in identifying superior types, according to their own needs. Interviewing farmers who failed to identify superior types from the same F5 seed revealed that they lacked knowledge of segregation and just concentrated on grain color for mass selection. Therefore, as Loevinsohn and Sperling (1995) suggest, a basis of collaboration can be found once the respective strengths of the two parties are recognized. This study found that expert farmers could be identified within a year of preliminary work and a program should be build on that. Women farmers are particularly skillful in assessing post-harvest traits such as milling recovery, cooking and eating quality of rice. Expert women farmers for this purpose can be identified from a village workshop. Men farmers possess more skill in assessing the standing crops for yield potential, management requirements and threshing criteria. An anthropologic tool is required to help amass all this information. Breeders and field staff should also interact with farmers frequently and should stay in the village for considerable time so as to start to understand the indigenous knowledge system. If the success of these initial efforts is to be sustained, research management should ensure a congenial environment for field staff who work in difficult areas. This is often forgotten by policy makers or research managers who have tried to replicate successful and innovative approaches from elsewhere. This study also reveals that farmers evaluate new varieties at different stages of crop growth, particularly at near maturity and also at threshing, milling and eating sessions. Monitoring indicated that farmers keep on changing their decisions depending upon the availability of new information. At present, farmers' involvement in formal variety testing is limited to preference ranking at maturity, however, the majority of farmers make their final decision (re:retaining or rejecting the variety) during milling, cooking and eating. For example, the area under M-4 was expanded due to its good milling recovery whereas M-2 was rejected due to its peculiar smell when cooked. M-3 was preferred at all stages but mainly because of its yield potential and straw height. Nilgiri-1 was first selected by a farmer in Lumle village who later changed his decision of expanding the variety to his remaining good fields because of the high shattering and poor taste. He wished to continue to grow it only in one plot where an inlet of cold stream water causes high sterility. Himchuli-2 was rejected by farmers at Chhomrong, Ghandruk and Lumle where rainfall is very high at the time of maturity and causes pre-germination before harvest. In contrast, Himchuli-2 was liked by farmers of Patlekhet village (1500-1700 m), where rainfall and humidity at maturity is less. In Lumle, the mother of Dhananjay Devkota selected non-sprouted panicles and planted them in a 100 m plot to see whether the problem of sprouting would continue. There are several such examples which support the decentralization of selection and such findings can be fed to outreach or extension staff. This study also found that high altitude farmers have their own way to assess eating and cooking qualities of rice. White grain rice with softness, content feeling and ability to expand after cooking are preferred traits. Whether these criteria can be related to standard grain quality testing procedures needs to be investigated in order find a method which can be used in breeding programs. The present varietal testing systems do not have mechanisms to consider farmers' relevant traits, such as post-harvest variety evaluation. Laboratory measurement of such grain quality traits is important as these are important criteria for selection or rejection of varieties by farmers. Grain quality is often not assessed in the formal varietal testing system of Nepal until it reaches the final stage of release (Sthapit, 1995). Farmers participatory breeding is a controversial issue for plant breeders who feel their own role threatened. Major changes in researchers' attitudes appear likely when more productive results become available. A main concern in this approach is to screen for disease resistance and to resolve the seed certification problem. If small quantities of seed are given to farmers, susceptible material will be rejected-- if breeders and pathologists can give farmers some training in identifying the symptoms and selection procedures. Bacterial ShBR disease, which is prevalent in cool temperate rice growing environments of Nepal, can be screened naturally as resistance to ShBR is heritable and local landraces are good sources of resistance (Sthapit et al., 1995). Fortunately, Chhomrong Dhan has durable resistance to ShBR disease. This may not be true with other diseases and, therefore, disease screening of F3 and F4 populations is essential to safe-guard farmers from disease epidemics and also to avoid possible criticisms of such an innovative approach. The cost-effectiveness of FPB could be greatly increased by distributing farmer-selected varieties outside of the project villages by making seed of identified cultivars widely available. Seed certification will be a problem because FPB method does not produce a pure line variety. However, this is possible to overcome in two ways: (a) introduction of the farmer-selected variety into the formal testing system, and (b) further selection for uniformity, using few extra resources. Both of these possibilities are being studied at present. Decentralized selection is, of course, how farmers have traditionally developed landraces. Farmers are also curious researchers who are constantly in search of productive varieties for their specific niches. Publicly-funded research institutions cannot afford to conduct varietal testing for such heterogeneous environments, and farmers should be encouraged and trained to do informal research in order to solve their specific on-farm problems. LARC, which is working in a very difficult and diverse farming systems, has realized that the varietal identification process can be made more rapid and efficient by increasing the range of genetic variation submitted to selection. Our experience suggests that farming communities have a clear understanding of the local production constraints and a range of preference and needs for a traditional crop, whereas formal research systems have a comparative advantage in terms of access to wide range of germplasm of different origins and of contrasting characteristics. Breeders working together with farmers have more chance of developing appropriate varieties, and still maintaining and enhancing biodiversity. Result suggests that biodiversity has been already promoted in participatory villages by allowing farmers to select populations of their choice, for example, two types of Machhapuchhre-3 (M-3G and M-3C) were generated from the same seed lot given to three different farmers. They are tall and intermediate Machhapuchhre-3 populations. Both M-3G and M-3C have good yield potential, cold tolerance and white grain color. The dilemma for breeders is then which variety to promote? How many similar varieties to multiply for seed? Can similar phenotypically-related varieties be certified? Berg et al (1991) has suggested that genetic diversity does not need to be expressed in real plant differences, and that it can persist as variation hidden in the genetic structure of the plant population. These materials, therefore, can be included in an informal research and development (IRD) program which will allow farmers to select varieties according to their own needs and circumstances-- without researchers' biases (Joshi and Sthapit, 1990). Such approaches in combination will also provide researchers insights into a wide range of rural problems, which may assist them to design more realistic and output-oriented research in the future. The decentralized selection of segregating material from a few carefully chosen crosses, drawing on the active participation of expert farmers, presents attractive prospects for fostering a more sustainable and productive agriculture, better adapted to local needs. This would consequently lead to identifying farmers' need-based technologies in a cost effective manner. A similar argument was also put forward by Loevinsohn and Sperling (1995). This new approach tried by LARC can be of considerable significance in developing other crop varieties suitable for diverse and complex situations. The prerequisites of the method are that objectives be clearly identified and that breeders be sufficiently flexible to learn and reciprocate with the farmers. Conclusions and recommendations
ReferencesBerg, T., A.Bjornstad, C., Fowler, and T. Skroppa, T., 1991. Towards an integrated plant breeding. In Technology options and the gene struggle, pp. 128-141. A report to the Norwegian Research Council for Science and Humanities (NAVF). Oslo:Agricultural University of Norway.Ceccarelli, S., 1989. Wide adaptation: how wide? Euphytica, 40:197-205. Chemjong, P.B., B.H. Baral, K.C. Thakuri, P.R. Neupane, R.K.Neupane, and M.P. Upadhaya, 1995. The impact of Pakhribas Agricultural Centre research in the Eastern Hills of Nepal: farmer adoption of nine agricultural technology. Dhankuta, Nepal:PAC. Farrington, J., 1988. Farmer participatory research: editorial introduction. Experimental Agriculture, 24:269-279. Galt, D., 1989. Joining FSR to commodity program breeding efforts earlier: increasing plant breeding efficiency in Nepal. Network Paper No. 8, London: Overseas Development Institute. International Rice Research Institute, 1988. Standard evaluation system for rice. 3rd Edition, June 1988. International Rice Testing Program. Manila, Philippines:IRRI. Joshi, K.D., R.B. Rana, M. Subedi, K.B. Kadayat, and B.R. Sthapit, 1995. Addressing diversity through farmer participatory variety testing and dissemination approach: a case study of Chaite rice in the Western Hills of Nepal. ROGRAM Seminar Paper (forthcoming). Joshi, K.D. B.R. Sthapit, 1990. In-formal research and development (IRD): a new approach to research and extension. Discussion Paper No. 4/1990. ROGRAM, Pokhara, Nepal. Lumle Agricultural Research Centre (LARC), 1995. The adoption and diffusion and incremental benefits of fifteen technologies for crops, horticulture, livestock and forestry in the Western Hills of Nepal. LARC Occasional Paper 95/1. Loevinsohn, M.E. and L. Sperling, 1995. Using diversity: conserving crop genetic resources by meeting farmers' needs. Paper prepared for the "National Seminar on Biodiversity Conservation", Asian Development Research Institute, Patna, Bihar, Feb 18-19, 1995. Lohar, D.H., B.R. Sthapit, K.D. Joshi, 1995. The convention on biological diversity and appropriate policy on plant genetic resources in Nepal. ROGRAM Seminar Paper 95/.... Paper presented at the National Seminar on " Appropriate Policy in Nepal on Plant Genetic Resources with regard to the Convention on Biological Diversity", Kathmandu, Nepal, 6-7 April 1995. Maurya, D.M., A. Bottrall, and J. Farrington, 1988. Improved livelihoods, genetic diversity and farmer participation: a strategy for rice breeding in rainfed areas of India. Experimental Agriculture, 24:311-320. Morris, M.L., J.H. Dubin, and T. Pokhrel, 1992. Returns to wheat research in Nepal. CIMMYT Economics Working Paper 92-04. Mexico. Nagamine, T., 1992. Altitudinal cline in chilling responses in indigenous rice varieties collected from Nepal. Japanese Journal of Breeding, 42:309-317. Siegel, S., 1956. Non-parametric statistics for the behavioral sciences. Tokyo: McGraw-Hill Kogakusha. Sperling, L., M.E. Loevinsohn, and B. Ntabomvura, 1993. Rethinking the farmer's role in plant breeding: local bean experts and on-station selection in Rwanda. Experimental Agriculture, 29:509-519. Steel, R.G.D. and J.H. Torrie, 1960. Principles and procedures of statistics with special reference to the biological sciences. London:McGraw-Hill Book Company, Inc.. Sthapit, B.R., P.M. Pradhanang, and J.R. Witcombe, 1995. Inheritance and selection of field resistance to sheath brown rot disease in rice. Plant Disease (in press) Sthapit, B.R., B.K. Dhital, K.D. Joshi, and J.Gurung, 1989. Achievements of farmer participation in agricultural research: a case study of an early rice variety NR 10158-2B-2. Prabidhi Sangalo (Nep),4(2):1. Sthapit, B.R. and K.P. Shrestha, 1991. Breeding for cold tolerance at reproductive phase in the high hills of Nepal. International Rice Research Newsletter, 16(5):14 (Oct, 1991). Sthapit, B.R., 1992. Chilling injury of rice crop in Nepal: a review. In J. Inst. Agric. Anim. Sci., 13:1-32. Sthapit, B.R., Joshi, K.D. and K.D. Subedi, 1994. Consolidating farmers' role in plant breeding: a proposal for developing cold tolerant rice varieties for the hills of Nepal. ORRID Discussion Paper No. 94/1, ORRID, Pokhara, Nepal. Sthapit, B.R., 1995). Variety testing, selection, and release system for rice and wheat crops in Nepal. Seed Regulatory Frame Works: Nepal. Lumle Agricultural Research Centre, Pokhara, Nepal. Tripp, R., 1989. Farmer participation in agricultural research: new directions or old problems? Discussion Paper 256, Brighton, England:Institute of Development Studies at the University of Sussex. Footnotes:1 The authors are thankful to all the participating farmers in different parts of LARC Research Command Area who contributed to make the approach successful. We are also grateful to Dr. C.N. Floyd, Research Advisor, Messrs. P.K. Shrestha and A. Vaidya, socio-economists for their valuable comments and advice on the manuscript and to Pitamber Shrestha and Mahendra Chaudhary of Crop Science Section for their technical assistance. Our thanks to Drs. M. Loevinsohn and L. Sperling, IDRC, New Delhi and Dr. P.E. Harding, Director of LARC for supporting us in the seminar. Messrs R. Shrestha and K. Chhetri are gratefully acknowledge for word processing. Lumle Agricultural Research Centre is funded by the Overseas Development Administration (ODA) of the British Government and works in close collaboration with His Majesty's Government of Nepal (HMGN). The support of both governments is gratefully acknowledged. (BACK)2 Expert farmers are knowledgable rice farmer as identified by the community. (BACK) 3 The NRCTN is the entry point of all introduced and breeding lines for temperate rice from which the elite entries are advanced to CVT and FFT. Varieties entries have to go through multilocational tests for five to six years before they are considered for release (Sthapit, 1995). The results of the CVT and FFTs grown across distinct agro-ecological zones are pooled, and varietal releases are based on the three-year overall performance of variety included in the FFT set. An overall good performer stands a good chance of being released, however, a cultivar which performs very well in one specific location, but overall is not near the top of the list, will never be released under the current system of cultivar evaluation and release. (BACK)
Documento(s) 27 de 38
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| guest (Leer)(Ottawa) Login | Inicio|Empleos|Derechos de autor y uso|Información general|Contáctenos|Ancho de banda bajo |
