Our friends at Bioversity have meta-done it again. After a milestone contribution a few years ago on the patterns of landrace diversity in farmers’ fields, now arrives a monumental review of the kinds of things that can be done to keep it there. ((Jarvis, D., Hodgkin, T., Sthapit, B., Fadda, C., & Lopez-Noriega, I. (2011). An Heuristic Framework for Identifying Multiple Ways of Supporting the Conservation and Use of Traditional Crop Varieties within the Agricultural Production System Critical Reviews in Plant Sciences, 30 (1), 125-176 DOI: 10.1080/07352689.2011.554358)) It comes as part of a special issue of Critical Reviews in Plant Sciences entitled “Towards a More Sustainable Agriculture.”
It’s a long and complex article and impossible to do full justice to here, but the motivation behind it is simple enough: “to understand better the nature and contribution of traditional varieties to the production strategies of rural communities … and the ways in which they are maintained and managed.” That landraces still make such a contribution, despite expectations to the contrary in many quarters, ((Not least much of the rest of the CG system, one is tempted to say.)) is undeniable. As the authors admit, there is a lively ongoing debate about whether this can, or indeed should, continue. The superior adaptation of landraces to marginal conditions, the stability of their performance over time, the socio-economic conditions of small-scale farmers, and growing concerns about reducing agriculture’s dependence on inputs would all seem to point to a continuing, significant role in livelihoods strategies, at least in specific situations. But if so, why do they need help, as the authors also concede?
The problem is that lots of external factors work against landraces being all they can be, from dysfunctional seed systems to short-termist government policies. The diagram in Fig. 1 makes this clear (clearer still if you click on it):
The paper is really about what can be done to make it a more even playing field for landraces, by overcoming these multitudinous constraints. To that end its centrepiece, Table 1, is a huge list of the sorts of specific actions that have been undertaken around the world over the years in support of landrace management on farm.
So, for example, re-introducing landraces from a genebank (be it international or community), based on local adaptation and farmer preferences, would address constraints 1a and 2a, which have to do with the lack of sufficient planting material. Getting breeding programmes to use more landraces and farmer selections would be great for 2d, 3a, 3b, 3c, 3d, 4a, 4b and 4c (which incidentally seems pretty good value for money). Enabling farmer groups to develop a marketing strategy for landrace products would take care of 3a, 4a and 4c. And so on, and on, for 16 pages. The list of potential interventions is nothing if not exhaustive, and each is in turn exhaustively documented with references.
This is, like the previous review, a tour-de-force. If anything is missing from the discussion, it is perhaps a sense of what the best bets might be. That is, having identified a particular constraint or set of constraints, what are the things that are likely to be most effective in different contexts, for different crops? For that, a rearrangement of Table 1 summarizing and prioritizing interventions for each constraint would have been a useful start. Thus, if you find specific constraint X, you can do A, B and C, and B is what has worked best in the past in similar situations. Another area that could have been explored in more detail is how interventions work — or fail to work — when implemented together. But all this would be no more that tinkering with the masses of data that have been put together, which is perhaps already going on in Bioversity’s Maccarese eyrie.
The next major analytical step, surely, is to work out, or document if the information is already available, what these interventions actually cost. That would allow some calculation of possible returns on investment. Whether, and to what extent, any of the things that are so comprehensively described in this paper are ever done on a large scale in support of landrace conservation and use on farm will probably in the end be down to that. After all, those pushing the “alternatives” have already done their sums.
Luigi: It’s the old problem of a pay-wall. I am not paying $41 to read it. And it would be useful if institutes with international public funding did not protect their results behind pay-walls. But I am intrigued. You mention the “the stability of their performance over time” for land races. But this may be from a low base, as local varieties become constrained by local co-adapted pests and disease. The Green Revolution increased yields a lot and, marginally, increased yield stability. Also, there can be a considerable turn-over of land races (even given the farmer’s constraint of obtaining replacement varieties). This would preclude any measure of long-term stability.
You also mention the: “superior adaptation of landraces to marginal conditions”. I think a generic problem of current crops is that they are all (almost) adapted to marginal conditions, in the sense that trees are often the ecological climax vegetation and crop relatives and the crops we obtained from them grew in marginal (stressed?) conditions that prevented the growth of trees (fire, flood and possibly lots more). As a unfortunate and expensive consequence, farmers have to spend lots of time ploughing, weeding, and growing annual crops to keep the trees at bay.
It is possible that more varieties of more crops can be maintained and used as garden/orchard plants, rather than broad-scale field crops. The success of heritage vegetable varieties gives a model of what can be done. Fruits also capture interest. When working out of the vast fruit collection in CATIE I was amazed at the money and effort hobby growers in Florida gave to obtaining and growing all kinds of obscure fruits: they certainly knew what they wanted.
The major species diversity for crops in most countries is a result of past crop introduction. This can be a bottleneck that can in part be overcome by yet more introduction or breeding with introduced wild relatives – as with African oil palm in Malaysia being crossed with American oil palm.
“So, for example, re-introducing landraces from a genebank (be it international or community), based on local adaptation and farmer preferences”
I’m doing this task in Argentina, our group is re introducing materials from genebanks in many groups of family farmers in five provinces.
It is possible and more than that, necesary
Ana
Dear Ana,
That’s really interesting. Is there somewhere we can find out more about your work? Maybe you would like to tell people here more yourself. We welcome guest posts.