|
|
| English - Español |
|
 |
| Archives Web >
Publications du CRDI >
Livres en ligne >
Tous nos livres >
USING DIVERSITY >
|
ID : 85272Ajouté le : 2005-07-21 9:59Mis à jour le : 2005-07-21 10:00Refreshed: 2012-02-11 23:32 
|
|
Document(s) 22 de 38
LOEVINSOHN: A question to John Witcombe. I wonder why you have used only released varieties in your participatory selection scheme, rather than including a range, possibly also including local checks or near local (i.e. varieties found locally or nearby) so as to make a wider pool available from which farmers might select? WITCOMBE: When we formulated the project, we had a process which is called a ’search process’ in which we defined all possible sources of germplasm to try with the farmers and that included failed cultivars, cultivars which had been rejected from the coordinated projects, even though we felt they had farmer desirable attributes, advanced lines from plant breeders, prereleased cultivars and released cultivars. But what actually happened is that as we went into the search process, we found that there was a huge diversity of choice amongst the released material that was unexploited. We saw the advantages of going with released material because, if you start there first and get something which farmers prefer which is released, then your uptake part, the ettectiveness of your research is so much greater. If that procedure fails, or if we want to build on the research, our second route has been to go to participatory plant breeding as the route of choice. For example, in the maize where we have had very little success with released cultivars, (although ’Shweta’ which is released in Uttar Pradesh has been adopted by our farmers), we have gone to creating specifically a composite which includes local landraces and farmer preferred cultivars — very broadly based. GHILDYAL: Dr. Witcombe, you are working in a rainfed environment and this environment is not uniform. Over even a 100 m distance, the environment changes, the water regime and soils change. Therefore, the lack of adoption you mention may be due to the very varying environments. WITCOMBE: The recommended cultivars in the area are all dwarf, high yielding cultivars. The rainfed cultivars which are recommended are of late maturity. All of the material which has succeeded are recommendations from outside of the states in which we are working, and the reason why there is no adoption has been because the farmer have not been exposed to those cultivars. It is a lack of popularization — which is a failure of the release system. When a cultivar is released in one state, for example in Orissa, which is the case of Kalinga 3, there is no formal mechanism for ensuring that it is tested in other states. We were the first people to test Kalinga 3 in Gujarat and Madhya Pradesh more than one decade after it was popular in Orissa. RILEY: I have a question for Trygve Berg. When he referred to the barley work that was conducted in California, which was conducted, as I understand it, under natural selection conditions, to what extent would the results be applicable to what we are talking about in terms of farmers’ selection of landraces, where farmers are the predominant entity in shaping the nature of landraces? Could extrapolation of that natural selection experiment perhaps lead to some erroneous conclusions? BERG: One of the first scientists to analyze this material was Suneson who published his paper in 1956 and proposed ’evolutionary plant breeding’. What he claimed was that the natural selection which had occurred had not only established material with a high degree of stability and disease resistance, and a reasonably high yield, but also established a base for line selection of improved materials, and therefore he wanted to exploit this. The base for the success and the evolutionary progress of this material was, of course, wide diversity. I am not sure that this wide diversity, which can bring about such progress, is present in all the landraces which farmers have. I don’t think it is, but we don’t know enough about it. I think scientists can contribute a lot to create diverse materials from which selection, both natural and artificial selection, can be made under local conditions. GUPTA: The process of taking advanced lines of segregating populations to farmers can serve two purposes. One is the selection itself, and the second is to use it as a heuristic tool to help untold the criteria which people use .... These criteria may not be revealed in response to questions about the varieties that they have been using for ages. The second point relates to parallel processing — where farmers can be brought to station and can be exposed to far greater choices than you can ever share with the farmers in their fields, because of the constraints of what you can take to farmers’ fields .... As farmers make selections in their fields and on-station, so scientists will make their selections. And the two can be compared to see if consistent characters are being selected for or whether different genetic qualities are being advanced in each. That would be a useful approach. SAHAI: A question to Dr. Sperling. You mentioned this model — ’division of labor’ you called it — and what the inputs are that farmers could provide and what the inputs are that breeders can provide. Is this model actually working anywhere? SPERLING: The ’division of labor model’ is one that has been most fully developed in Rwanda, but also tested, with beans, in Zaire and Tanzania and with agroforestry species in Burundi. I want to mention that the model changes, even with the same crop, when taken to different sites, so that models have to be adapted to specific situations. lf we consider the general principle of saying that, as the breeders’ function we should offer the farmers a wide array, that is diverse, and then set up the mechanisms for farmers themselves to select and target these varieties then we could say that this general division or principle is being experimented upon in many crops and sites: e.g. cassava in Colombia, pearl millet in India, barley in Syria. SINHA: I’ve restrained myself from commenting earlier, but I think this discussion has some important implications for plant breeding in India. I think it is important that we consider some of the historical developments of plant breeding. One of the first phases was when there were shortages and there was a need for wider adaptability, particularly in the areas of assured inputs. What we have heard today are examples drawn largely from rainfed and upland conditions where there has not been such a large impact of selections or release of new varieties. In areas of assured input, however, there have been important gains. It is also now realized that local selection of regional specificity is going to be far more important — not that farmers have not done it in this country. There is a farmer I happened to visit last year, Mr. Rana, who also happens to be a graduate of St. Christopher’s College, who took to farming. He has himself selected a variety of wheat, which is being sold, which gives eight tons — as against the six or seven tons from commercial varieties. He has also selected varieties of cotton out of material being provided — so this kind of effort is going on. A second, important point made by L. Sperling, was about the selection of crops, here beans, for certain conditions, like under bananas, etc, in intercropped situations. Most of our plant breeders were trained in Britain where single crop cultivation was a normal practice. When they came back, they essentially devoted their time to those situations to which the had been exposed. It is true that selection for intercropping and multiple cropping systems has not been effective and that this is an important, required aspect of tropical agriculture .... There is now a lot of consciousness regarding the need for location-specific selection and we have started by giving 10 to 15 pre-released varieties to the Sone Command Area [Bihar]. However, the mechanisms as such need to be changed at the governmental level ....
Document(s) 22 de 38
|
|
|||||||||||
|
||||||||||||||
| guest (Lire)heure de l'Est (É.-U. et Canada) Login | Accueil|Carrières|Droits d'auteurs et usage|Informations générales|Nous rejoindre|Basse vitesse |
