The idea is to open up new options for farmers

An interview with Donald L. Winkelmann

Ceres: CIMMYT recently marked its 20th anniversary and much attention was drawn to the achievements of the last two decades. Do you think that the Centre will be able to maintain this tempo of change? What are the prospects for new developments over the next 15 or 20 years?
Winkelmann: Our best-known achievements of the past 20 years are the dwarf wheats. We think it is unlikely that we will have that kind of breakthrough over the space of the next two decades, but CIMMYT staff is pursuing work that could give us another such boost. Progress in maize has been less dramatic but notable even so and represents the kind of continuing improvement, useful for national programmes and farmers, that we count on for the future for both maize and wheat.

Q: Is it fair to say that the very success achieved in widespread adaptation of improved varieties of wheat and maize has also generated other problems of an economic and social nature?

A: Well, let's start with the recognition that all change has implications that run well beyond those initially envisioned, whether it involves dwarf wheats, the wheel, or the fermentation of yeast. If it is a significant change, the implications are almost inevitably wide ranging. There was substantial consternation early in the so-called green revolution that its implications for income distribution were detrimental, that it made the rich richer and the poor poorer. There was criticism of CIMMYT, and of the whole initiative in agricultural research, because of a sense that the gains had not fallen in a way initially expected. But it is my perception that these attitudes have changed dramatically over the past decade or so and that people now see that the poor of the world - poor customers and poor producers - were the major beneficiaries of those discoveries. The arguments for this are well known and well documented. This is not to say that one can be completely sanguine about what transpired on the income front. For example, we know that there are substantial areas of the world that have not yet benefited from improved agricultural technology. Indeed, that recognition now has a heavy influence on the orientation of our research. In an earlier day our concentration was on wheat for the well-watered regions of developing countries. And justifiably so, as it turns out that three-fourths of the spring bread wheats in developing countries are grown under well-watered circumstances.

Now, however, a substantial portion of the total effort of our spring bread wheat programme is focused on those more difficult environments characterized by drought, soil problems, and high heat. The work on disease resistance in well watered areas, of course, goes on.

Q: Will your efforts to help those harsher environments require major changes in your research system?

A: One thing we see is a need for sharper characterization of those regions. Through our associations with colleagues in national programmes we are developing a better sense of the high-priority problems that affect those environments. One should understand that the tailoring that is required for any specific environment is, with high probability, beyond our reach as an institution that makes its products available to many national programmes on a global basis. It would not be efficient for us to try to pursue such tailoring. Instead we develop materials that are generally useful for a major environment like lowland tropical maize, and make those available to national programme areas in that environment.

Q: That's where the fine tuning is done?

A: Yes. One of our functions is to ensure that national programmes have easy access to the materials from these international networks of plant breeders. This apparatus is really phenomenal. It brings together the energies of more plant breeders than have ever focused on a single problem in human history. And we're getting better at this. What we're coming on to now is an ever more precise delineation of major environments and ever closer association of individual experimental stations and plant breeders with such environments and hence an ever more efficient exchange between people in like environments in different countries or even on different continents. This means that an individual plant breeder associated with the network in, let's say, the coast of Guatemala, has easy access to material put together by a person in a similar environment, let's say, in Cd'Ivoire, or by another breeder in a like environment in the coastal areas of Ecuador. In this way the energies of a large group of people coalesce around a particular set of problems. This is not to suggest that the material that comes from Cd'Ivoire is going to fit neatly into the niche in Guatemala, but it puts the breeder there far ahead of where he might have been by himself. He is then able to make the refinements that his local circumstances require.

The second thing that we do is to offer training of various kinds which helps the national programme personnel on methods and techniques, on sensitivities to new issues and new themes in plant breeding. They acquire new inputs for their programmes and new ideas on how to push these ahead. This makes for sharper, more effective activity than 20, or even 10, years ago.

Q: CYMMIT is working on wheats for more tropical environments. The basic genetic material for such wheats must be much more limited than those suited to temperate or semi-tropical zones. Where are these genes coming from?

A: Well, it turns out that wheat has much greater plasticity than we initially suspected. A decade ago when we first started putting wheat down on the coast of Mexico, Helminthosporium sativum overwhelmed everything. From as many as 500 lines, perhaps only one or two would survive. Now the differences are really notable. There are materials that not only survive but produce. These are the result of the introduction of new genetic material over a number of years. We also see an increasing tolerance to late-season heat. These obviously are going to be useful in environments where, heretofore, wheat had not been an option. Having access to that late-season heat tolerance might change the production patterns of farmers in environments where wheat has long been a traditional crop, such as the southern part of Pakistan through the Punjab and over toward the Sind. The idea is to open up new options for farmers. We' ve seen maize move well north into Canada and well north into Europe as plant breeders have pursued materials more suited to those environments. I have no doubt that we are going to see wheats move into environments we thought were simply not destined to grow wheats. How prominent they will be in those areas, how large an extension we are talking about, I don't really know. Other people are providing better rice varieties, and these are also options. As I said, the idea is to open up more options to farmers, and wheat which accommodates the stresses of more tropical environments is one way to do that.

Q: At any stage of developing varieties likely to be widely adapted is there any input into the planning of centres such as CIMMYT on what the possible nutritional implications of a major shift in cropping patterns would be?

A: IFPRI (International Food Policy Research Institute) has made a major effort on issues that relate to nutrition. What happens to calories and proteins? Ten years or so ago there was substantial anxiety because farmers were shifting away from pulses and into cereals. Our colleagues at ICRISAT (International Crops Research Institute for the Semi-Arid Tropics) showed that the total production of proteins - and of calories - had increased dramatically in that particular case as a consequence of going into the higher-yielding cereals and going out of the pulses. There will be other cases in which they won't be increased. I think that the whole system is going to have to give more attention to such themes.

One of the major changes over the past 20 years has been a shift in emphasis from a production or output orientation in our work in agriculture and agricultural development to a productivity orientation - input oriented. So instead of asking questions about what we can do to increase production of, say, maize or wheat, the questions we are asking now are what we can do to increase the productivity of farmerheld resources that are committed to maize and to wheat. This is a more interesting, more profound conceptual framework for organizing our work.

Q: You have referred in some of your speeches to the importance - and the difficulty - of identifying comparative advantage. Are you talking about the comparative advantage between crops or about the comparative advantage among different regions for producing a particular crop? Does this refer to the problem of surplus production in some areas?

A: Comparative advantage is one principle for assigning resources. The idea is to use the resource endowment of a country to maximize, through trade, the total stock of goods and services available to its residents. But what has happened to us in the contemporary world is that real flows of goods and services tied to what resources can do - land, labour, and capital - are something on the order of one-twentieth of the financial flows in today's world. The financial flows often overwhelm whatever markets are trying to tell-us about the value in production of labour, land, and capital. If you are in a place like Mexico, wondering about which direction to go, where to put your own investments for a longrun payoff, do you put more research on wheat, or on cotton, or more on maize?

Q: Are you getting a distorted picture?

A: Let's call it kaleidoscopic. You see one thing today, another thing tomorrow and something else later.

Q: Would you cite, for example, the case of Saudi Arabia's venture in wheat production to demonstrate the impact of an excessive amount of capital?

A: No, the Saudi case relates to a different question. They were paying five times the world price for wheat. Look what happened. An incredible increase in production, which, again, just says that there are prices at which you could produce bananas in the Arctic. The financial flows and comparative advantage issue is a different one. In the past, when people had more stable prices, they could more easily make judgements about their comparative advantage. But today, the relative prices of currencies vary enormously because of financial flows, and these influence judgements about comparative advantage. What this means is that one is much less certain about where to orient domestic research resources than when prices were more stable.

Q: Are you saying, in effect, that both the scale and tempo of international financial transactions are having an impact on the direction and planning of agricultural research?

A: It clouds, it obscures, it makes it more difficult to divine the direction in which one should be moving. Imagine managing the national research programme of, say, Mexico. You have a series of options which are biologically feasible and you have limited resources to invest across that array of options. What you try to do is to choose the sorts of things that you think Mexico will want to place in the market ten years from now. Such decisions are now more difficult to make because of those enormous ebbs and flows on the financial side with their consequences for exchange rates.

Q: Do you see biotechnology markedly affecting developments in maize and wheat breeding?

A: Wheat and maize varieties in the next decade will probably emanate from conventional breeding practices. Bear in mind that even in Mexico, where we get two crops a
year, it takes five years to develop a variety. So the crosses that are being made now will become the varieties five years on. But we are trying to monitor biotechnology closely, for two reasons. First, because developments in the new sciences have implications for us and our work. But beyond that they have potential implications for national programmes. We see it as one of our responsibilities to help keep national programmes apprised of what is available through biotechnology. So we are adding staff in that area. As well, we are seeking to develop working relationships with the most skilled practitioners of these arcane arts.

Q: Most of whom are in the private sector still?

A: It is really a mix of private and public sector. The Plant Breeding Institute in England, for example, has a gifted staff. The land grant system in the US, the public systems in France and Germany, also. But let's recognize that a very substantial proportion of the resources in biotechnology are accounted for by the private sector. Clearly, these people are optimistic about the future of biotechnology.

Q: There has been much controversy - and perhaps some confusion as well - over the question of plant genetic resources. How do you see CIMMYT's role in the conservation of germ-plasm as far as the crops for which you have a mandate are concerned?

A: With regard to conservation, this has been a sticky and contentious issue. We've spent a great deal of time talking past one another. Happily, over the past several months we have been able to get a far more precise sense of what the international community expects from us and of what it is that we can effectively offer. Wheat is easiest to talk about. The genetic resources of wheat are held in a variety of places. There is a major holding of the wild relatives of wheat in Japan; there are major holdings of wheats in the USSR, the German Democratic Republic, Italy, Ethiopia, Turkey, and the US. What we have agreed to is that, for spring bread wheats and spring durums, we will hold in our base collection the materials that have emerged from the dwarfing genes, those materials that are most closely associated with CIMMYT's own history. These are essentially first working collections, and second for intermediate storage - 25-year span. We have agreed to hold, in intermediate storage and working collections, all the spring triticales that have emerged. For maize, the custodial responsibilities are notably different. We have agreed to hold, in long term, over 50-year storage, a collection that represents all the land races of the Western Hemisphere, which make up something like 90 per cent of the world's land races of maize. We have also agreed to keep records on the land race materials that are being held in other parts of the world - significant numbers of land races are held in northern India, Thailand, and Japan. While we won't necessarily hold copies of these, we will know where all that material is and how it is characterized. We have entered into no formal agreements on this; we simply, in collaboration with IBPGR, have agreed that this be our current role within the CGIAR system on this particular issue. As well, we are seeking to facilitate the development of national germ-plasm facilities, especially in maize. The idea is to have the materials widely enough distributed so that simultaneous loss is unlikely. Even the fact that we are on the flight path coming into Mexico City airport may constitute the kind of threat that might induce us to think about having duplicate collections. Indeed, it has been recommended to us that copies of everything at El Batan be placed elsewhere in Mexico.

Q: Is there a case to be made for in situ preservation in the case of wheat or maize?

A: This is pretty much the strategy that we have come to with trypsicum and teozintle. But in situ in maize? Think of it: 12 000 accessions. That is a lot of material to tend. The other side of it is that you would get a constant genetic drift over time if you were simply to continue planting them out year after year after year. Another interesting side of conservation is the physiology of seed at very low temperatures, something that we don't completely understand. So we are initiating, as well, a modest research programme in conjunction with our germ-plasm storage effort, to get at some of these issues.

Q: Is there no base in past experience for knowing how long seed can be stored under different conditions?

A: The conventional wisdom is that seed stored at -4 and about 10 per cent moisture will be viable for 50 years. But it turns out that materials taken out of long-term storage and planted need special attention. They need a little bit of tender loving care when they come out of the box, before they go into the ground. We know of materials that were lost because they were planted in the same way fresh seed might have been planted. There was insufficient seedling vigour, and plants were lost. This raises questions about the viability of the seed per se. I don't think that this is scientific black magic. You just have to be more careful because vigour has diminished as a consequence of that long stay at low temperatures.

Q: What has happened to hybrid wheats? Is that an exercise that never got off the ground? Is there a future for hybrid wheats?

A: There are selected national programmes that are interested in hybrid wheats. There is a continuing effort in hybrid wheats among commercial companies in the US and in Europe. And there are surely environments where hybrid wheats will be profitable and will be employed by farmers. It's hard to believe, though, that hybrid wheats have much of a place in developing countries in the near future.

Q: Because of the cost of seed replacement?

A: Exactly. Take, for example, single cross hybrids. It turns out that the cost of production for such hybrids might well run to the equivalent of 12 to 15 kilos of commercial grain. So, if I were to buy one kilo of seed, single cross hybrid, I'd have to trade in 12 to 15 kilos of commercial grain. If I need 20 kilos per hectare for seed, my extra seeds costs for going to the hybrid would be in the order of 250 kilos per hectare; that's the extra grain that I have to get back just to pay for the seed. And for a farmer who's getting 1 500 to 2 000 kilograms with his current practices, it's not very likely that just changing seed will give him that kind of boost in yields.

Q: You mentioned earlier that production was no longer the main theme. Are you developing strains with the idea that these fit into lowyield agriculture?

A: Well, first, I don't want to leave the impression that production is not important. It's just that production frequently gives you the right answer to the question about varieties and technology, while productivity always gives you the right answer. That slight twist in perception can make a difference. Now, with respect to what kinds of materials we are developing, we're looking for a great deal of stability and robustness, materials that are well buffered so that they can readily accommodate the stresses that nature imposes during the growing season. As well, we want materials that are responsive to various kinds of inputs. But we certainly are not pursuing hothouse varieties that need to be almost manicured if they are to yield well. Robustness, stability, the capacity to accommodate the stresses of nature - these are important considerations for our work. By the same token we want things that are responsive to better management. Over time, one of the things that the farmer does when he finds robustness and stability in the face of nature's adversity is to cultivate more intensively. The evidence in India is clear on this, also in Bangladesh and other countries. Give the farmer the variety that is resistant to diseases, that accommodates the stress that nature imposes, and his confidence in those materials makes him willing to accept the risks associated with heavier investment in the plant - more fertilizer, better water control, more care in his weeding. So we want varieties that respond to that extra attention.

Q: Is there any role for what is known as horizontal breeding in developing lines that are suited to these harsher environments?

A: Well, Norman Borlaug was well on the way toward multilines when the dwarf varieties hit the market. The yield increases from the dwarf varieties were so high the multilines were just overwhelmed. Now, some of our colleagues, Khem Sing Gill in India in particular, have done good work on multilines and have multilines available. In the midwestern United States, the programme at Iowa State University has done phenomenal work with oats multilines. Our principal breeder in spring bread wheats is interested in multilines. And, of course, in maize you always have multilines because it is an outcrossed plant. I had always thought of the major advantage in multilines as being in protection against disease. But perhaps in those harsher environments there are other kinds of buffering that the multilines can offer. Diseases are usually less pressing in those environments.

Q: How much feedback do you get from farmers, either directly or through national programmes?

A: Our real link with farmers is through national programmes, and I want to emphasize that our clients are national programmes. What we do is serve up a set of intermediate goods that national programmes can use in their efforts to develop improved technology for farmers. To the extent that national programme researchers are themselves conscious of farmers' reactions, we get a great deal of feedback. To the extent that national programme researchers are isolated from farmers, then that cuts the thread.

Q: Do you see the national programmes ever becoming strong enough for there to be no need for CIMMYT?

A: Well, two points. The first and easiest is that I think there will always be great advantage in having an institution like CIMMYT to serve as the hub of an international network of breeders of wheat and maize. These know that they can share their best materials through us and receive the best of others in return. To the extent that you can build trust and confidence, this could be done by an individual national programme. The great advantage that CIMMYT and other international centres have - certainly one of our most valuable assets - is the trust of the national programmes, their faith in our evenhandedness at the hub of that network. Now we are pursuing the possibility of devolving certain activities in which we are currently engaged to some of the more advanced national programmes so that we, ourselves, can move toward more strategic research. It's really a process that has been going on for many years. For example, the kind of work that we do in agronomy today is really markedly different from what we were doing 20 years ago.

Q: Could you give some specific examples?

A: Certainly. Twenty years ago CIMMYT staff were working alongside Indian and Pakistani researchers developing technologies for Indian and Pakistani farmers. Those were new varieties; they had new requirements in husbandry. Our associates went out there to help. By now, national breeders and agronomists know how to handle those varieties; they are familiar with their growing habits, characteristics and needs, and they take care of those aspects. Meanwhile, CIMMYT staff are turning their attention to newer problems. One example relates to sustaining yields in agriculture. This is going to be an ever more challenging task over the coming decades because we are using our agricultural resources in a much more intensive way than ever before - for example, the wheat/rice rotations in the Punjab and across into Bangladesh, or the soya/wheat rotations in the pampas of Argentina. These represent an enormous intensification in land use.

How do we sustain those yields?

Q: Apropos of that, there seems to be some concern in the United States that it is going to become increasingly difficult, as well as more costly to maintain the very high level of yields that they have attained. In other words, there has been a deterioration in the basic resource.

A: I don't know whether there has been a deterioration, but the evidence suggests that something is happening that we need to know about. We have some data out of Pakistan, for example, that suggests, yes, yields are being maintained. But if you look at the input side, what you find are ever more experienced farmers using ever higher levels of inputs but not improving yields.

Q: From the farmer's point of view, are you past the point of optimum profitability?

A: Well, first, one point is that the farmer never got to the optimum level of profitability. He is pursuing it. But now what has happened is that the potential profits that are available to him are shrinking. My hypothesis is that will not only require the kind of agronomy that we were practising 20 years ago. It's going to be an agronomy that rests much more closely on underlying scientific disciplines. These are the paths that CIMMYT will be pursuing because of the staff we have, because of the horizons we have for research and the easy connections we have into the basic scientific community. We're locking for new ways to focus talents on problems of an international nature.