"Hybrid technology offers a tremendous potential for the much needed second Green Revolution", was one of the conclusions of a last year's expert consultation on hybrid seed production at the FAO's Regional Office in Bangkok. Undoubtedly, the characteristics of hybrid plants enable commercial seed production. But all that glitters is not gold. What may be a blessing for the industry, may be a curse for farmers, especially in less favourable regions. It is therefore the task of public research institutes to develop seed types other than hybrids. Apomixis may also be an option.

Hybrid plants have become increasingly important in various commercial food crops around the world. In crops such as maize, sunflower, sorghum, sugar beet, cotton, and many vegetables, hybrids account for a large share of the market. Not only the USA and Europe, but also many developing countries rely in their food production to a large extent on hybrids. Sales of hybrids in various crops account for nearly 40 per cent of the global commercial seed business of about US$ 15 billion. This share is likely to increase as hybrids are considered to be an important means to feed a growing population, especially for developing countries.
The emphasis on hybrids technology in food production results from the merits which are usually attributed to hybrids. First and foremost is their yield potential. Hybrids generally must have 15­20 per cent higher yields than open­pollinated varieties (OPVs) in order to be commercially feasible. Another advantage over OPVs is that new desirable characteristics, such as disease resistance, can be more easily bred into the hybrid, because breeders work with two inbred lines instead of populations.
Hybrids are also extremely uniform. This characteristic makes hybrids rewarding objects in plant breeding where uniformity is an explicit goal. Uniform maturity and size is necessary for farmers who harvest mechanically, while uniformity in fruit size and colour is required by the processing industry and preferred by direct consumers. Uniformity is also a necessity if one wants to identify crop varieties and their breeders.
Finally, hybrids have a built­in protection against multiplication. Unlike OPVs whose seed gives yields similar to its parents, the yield of the seed borne by the hybrid plants is significantly lower than that of the first generation. This feature has two consequences. First, farmers must normally buy seed every year in order to obtain the high yields of the hybrid. Second, the lower yield of the second generation will eliminate the trade in seed that is propagated by seed producers without authorization by the breeder, while it also prevents the redistribution of the commercial crop as seed by middlemen (common in grains). If the seed industry can keep the parental lines in­house, no competitor or farmer is able to produce the hybrid.
The lower yields of the second generation hybrids has been a crucial factor in the establishment of private seed companies in various parts in the world, not least in the USA. Hybrids are therefore also considered to be one of the most important incentives for the development of a private seed industry in developing countries.

The other side of the coin
During the abovementioned FAO meeting, a representative of Pioneer Hi­Bred, USA, in the Philippines called for a massive hybrid seed education programme to remove the suspicion that hybrid seeds are expensive, big consumers of agricultural inputs, and a tool of developed countries for controlling the agricultural economies of developing countries." There is some truth in this argument. The need for additional inputs is not typical for hybrids but also applies to modern OPVs. And beside multinationals, the national seed industry in developing countries is also active in hybrid breeding and trade. Nevertheless, suspicion against hybrids still seems justifiable for a number of reasons.
Firstly, genetic uniformity of hybrids may be a desired trait for many but it also makes the plants vulnerable. This became painfully clear in the USA in 1970. About 15 per cent of the maize crop in that year was lost to an epidemic of Southern Corn Leaf Blight. The epidemic was caused by the susceptibility of the cytoplasm incorporated into the maize lines to achieve male sterility (see box). Because nearly every maize hybrid in the USA carried this cytoplasm, the epidemic swept maize fields in large parts of the country and would have been worse if the weather had been less unfavourable for the disease organism. The risk that uniformity in cytoplasm in hybrids causes an epidemic again is not just theoretical, as in some crops the number of different cytoplasm types incorporated in the hybrids is extremely limited.
Secondly, higher yields may not always be the right justification for emphasizing research on hybrids. Comparing yields of hybrids and OPVs is only relevant when research and development efforts in both plant types have been similar. It has been pointed out that in the case of US maize breeding such a similarity did not exist. According to Lewontin and Berlan, and Kloppenburg, the focus of the US public maize breeding after 1920 shifted entirely to hybrids at the expense of research on OPVs. The reason was not the proven yield superiority of hybrids, but rather a strong political offensive by the private seed industry. Led by Henry A. Wallace, the founder of what now is Pioneer Hi­bred and Secretary of Agriculture in the 1930s, the industry succeeded in making hybrids the central thrust of the public research institutions. Government officials who opposed hybrid maize were replaced, stubborn experimental stations isolated.
Thirdly, hybrids make farmers more dependent on formal external seed deliveries. This is probably the most important disadvantage of hybrids. On­farm seed saving is not useful, because yields of seed of a hybrid plant are significantly lower than the hybrid itself. For the same reason farmers can no longer rely on local informal channels of seed supply, be it exchange among farmers in the region or the inofficial seed supply by their dealers. The annual purchase of fresh hybrid seed is a necessity.
The higher and annually returning costs of hybrid seed may be less of a problem for commercial farmers who work in areas with beneficial agronomic conditions and with good access to loans and maize markets. But for farmers in developing countries who work in less favourable conditions, higher seed costs may constitute too high a burden. Transport to remote areas makes the seeds even more expensive. These farmers will probably avoid the use of hybrids and miss all research involved in this type of seed.
Farmers who do shift to hybrids reduce ­ and in due course eliminate­ their own ability to produce seed. Instead, they become reliant on a seed industry whose operations are beyond their control. In case of an incompetent seed industry, farmers run the risk that seed is not supplied before the optimal planting time or is not delivered at all. As is pointed out in the article on Tanzania (see artikel by Esbern Friis­Hansen), this situation forces farmers to rely on the second generation hybrids.
Fourthly, farmers will become more dependent on the market for selling their crops. Farmers can only afford to buy seeds when they can earn by selling their products or labour. But in several (parts of) developing countries the markets for surplus production or rural labour are very poorly developed. Transport can become problematic too, as larger volumes are involved.
 

Hybrids: The techniques  With the development of hybrids, breeders take advantage of a natural phenomenon, called heterosis or hybrid vigour, which is the tendency for offspring of genetically diverse plants to perform better than their parents. By crossing two cultivated different parental lines which contain specific characteristics, the first filial (F1) of this cross can combine all these characteristics with the hybrid vigour. Yield potential is very important. In some cases the hybrids yield significantly more than the available open­pollinated varieties (OPVs). It is generally reckoned that hybrids must at least yield 15­20 per cent better than the best OPV to be commercially competitive.  In order to produce hybrids, a group of inbred lines must be formed. This is done by repeated self­pollination of individual plants. Each inbred line is now (almost) genetically uniform and different from each of the other inbred lines.  Male sterility  The key to the successful commercial production of hybrid seed is sufficient control of the pollination process. Self­pollination of the (female) seed line plants must be prevented. This can be achieved through various ways. Maize, for example, has distinctly separate male and female flowers which makes the plant well suited to manual or mechanical emasculation. The tassels are removed from the seed plants before they are able to shed pollen. Another approach, employed in rice, sorghum, sunflower and millet, is the induction of male sterility in the seed line on the basis of sterilizing cytoplasm. This type of male sterility is conditioned by hereditary particles in the cytoplasm. Cytoplasmic male sterility (CMS) of the seed line can be achieved through crossing with naturally occurring CMS plants. Nuclear male sterility (NMS) has been developed too. This genetically engineered pollination system inhibits pollen development in the seed line during hybrid seed production. NMS is being applied to make oilseed rape hybrids. In most crops the male parent must contain the specific feature to restore fertility in the hybrid seed, in case CMS or NMS lines are being used.  Male sterility can also be induced by day length or temperature, called environment­sensitive genic male sterility (EGMS). In rice, for example, EGMS lines may have complete pollen sterility either when the day length is more than 14 hours (which make them useful in temperate zones) or at a temperature of 30­320C (useful in tropical zones). In research on hybrid wheat, chemicals are being used to achieve male sterility. A gametocide is sprayed on the seed line to make the pollen sterile.  Single and double crosses  Hybrids of single crosses (with two inbred parental lines) are costly to produce. In some crops, such as in maize, the parental lines suffer from an inbreeding depression resulting in lower yields of the seed line. Consequently, large areas must be utilized in order to generate the necessary quantity of hybrid seed, making the seed production expensive. That is why the early expansion of hybrid maize seed was based on the double cross technique that combines four inbred parents in two single crosses. As single cross hybrids yield about three times more seed per plant compared to the inbred parent, the crossing of two hybrids (the double cross) makes hybrid seed cheaper. The plant resulting from a double cross has lower yields compared to the single cross, but in the early years of hybrid maize in the USA, these yields still exceeded those of available OPVs. Moreover, a second generation of a double­crossed hybrid shows, at least for some grains, a smaller reduction than the second generation of a single cross. Three­way crosses also have been developed. These crosses tend to fall between single and double crosses in terms of costs, variability and yield.  The trend in hybrid maize has been towards a greater use of single cross hybrids. This seed type covered nearly 90 per cent of US maize acreage by 1980. In developing countries, double and three­way crosses are also still in use.

Hybrids and public institutes
In the hybrid seed market multinational seed companies are typically active. Companies like Pioneer Hi­bred Inc., DeKalb, or Limagrain have expanded on the basis of hybrids. All seed multinationals breed, produce and sell hybrids, rather than OPVs. But hybrids are not exclusively the domain of seed multinationals, not even in maize. National and international public institutes in many developing countries also develop hybrids.
The International Rice Research Institute (IRRI, the Philippines), for example, spends around 10 per cent of its R&D budget for the irrigated rice programme on hybrid rice, compared to 30 per cent on the development of OPVs. The International Maize and Wheat Improvement Center (CIMMYT) allocates 18 per cent of its maize programme funds to hybrids. Around 13 per cent goes to development of OPVs and 30 per cent to population improvement. The remaining part is spent on, among other things, training and crop management. The Asian Vegetable Research and Development Center (AVRDC) devotes 20 per cent of its breeding budget for various crops to hybrids. AVRDC has developed hybrids in tomato and Chinese cabbage.

The involvement of the international agricultural research centres (IARCs) in the development of hybrids may be somewhat surprising. The IARCs generally perceive themselves as "stewards" of the world's genetic resources. They have no intention to keep OPVs, inbred lines, or raw plant materials they have developed proprietary. Why then hybrids?
According to S. Shanmugasundaram, Director International Cooperation Program of AVRDC, and P. Roger Rowe, Deputy Director General of Research of CIMMYT, the research on hybrids has indeed long been disputed at their centres. They point out that their centres develop both OPVs and hybrids. "We feel this broad effort is required because of the different situation in our partner countries. Some countries now have near 100 per cent hybrids and a strong private sector, others have only OPVs and a poor seed sector", says Rowe. Shanmugasundaram recognizes that it is a disadvantage of hybrids that this seed type favours rich and large­scale farmers. But, "whether we like it or not, the private companies are producing hybrids in vegetable crops such as tomato, pepper, cabbage, and onion. The farmers are willing to pay a higher price since it is definitely superior to OPVs". The IARCs consider research on hybrids at their centres to be a way of supporting the seed industry in developing countries. It is this industry (public and private) that uses the control entailed by the hybrid to get remuneration for its investments, not the IARCs.
The position towards hybrids seems to vary among IARCs. For example, the International Center for Tropical Agriculture (CIAT, Colombia) does not conduct research on hybrids in rice and beans. William R. Scowcroft, Deputy Director General­Research of CIAT explains: "The reasons for this are the additional cost of research, the likely inability of small­scale farmers to afford the more expensive hybrid rice, and the current lack of demand by national agricultural research systems for hybrid rice". CIAT is only willing to consider hybrids technology when hybrid vigour can be 'fixed' through apomixis.
 

Hybrid wheat  Wheat is one of the most important food crops in the world. But the relative ease for farmers to save its seed, and the extent of redistribution of commercial wheat as seed are a headache for private breeders around the world. Hybrids would make wheat breeding far more profitable for them.  The first attempts to make hybrid wheat in the USA date back to the 1940s. Especially achieving male sterility was problematic, because of the physical and genetic features of wheat. Around 1970, most public institutes and smaller companies in the northern hemisphere dropped their hybrid wheat programmes, for two reasons. First, the hybrids produced could not compete with newly developed semi­dwarf varieties which were high yielding. Second, it had become possible, in the USA (as was already the case in Europe) to protect varieties against unauthorized propagation under the Plant Breeders' Rights system. Legal protection apparently weakened the incentive to build in biological protection.  Many seed multinationals which nevertheless continued their efforts in hybrid wheat during the 1970s, terminated their programmes later too, for various reasons. Yields of hybrids only little exceeded those of wheat OPVs, the results with disease resistance and stability of the hybrids were not satisfactory, and some companies faced problems with the toxicity of the gametocide which was used to sterilize the pollen.  At present, only little hybrid wheat research is taking place (e.g. in China and India). Research focusing on commercial hybrids seems to be limited to two France based organizations: Orsan and Hybritech (the latter is a company owned by Coop de Pau and Monsanto). Whether these organizations can make hybrid wheat commercially feasible remains to be seen. The area currently cultivated with hybrid wheat in Western Europe is negligible. With ever declining prices for commercial wheat and still better performing new wheat varieties, yields of hybrids must significantly increase to justify their higher costs. And this is unlikely to happen.

Apomixis
A new technology that could reduce farmers' dependency resulting from hybrids is apomixis: the production of seeds without fertilization (see article by Richard A. Jefferson). In some crops, apomixis could be used to obtain seed from a hybrid plant while retaining the hybrid vigour of the first generation.
In various IARCs, research on apomixis is currently underway in crops such as pearl millet, rice, maize, and forage grass (see box in other article). The reported results are exiting. The coordinator of the International Network on Apomixis Research (APONET) and associate scientist at CIMMYT, Yves Savidan, stresses that apomixis should primarily be considered as a tool for small farmers, because it gives them the opportunity to be more efficient in selecting their seed for the next cycle. When the farmer selects the best looking ears to make seed for the next cycle, he will be sure to get a percentage of good looking ears which is equal to the percentage of apomixis that will have been introduced in his variety.
Nevertheless, one gets curious about the impact apomixis may have on the big commercial crops. It has the potential to remove the hybrid's biological protection against propagation, and would permit more of the world's farmers to use hybrids. And it is not just theory. Savidan believes that an apomixis transfer from Tripsacum to maize (!) will be completed within the next three years. "It will be a surprise for the seed industry. My guess is that nobody there really knows and/or believes we are so close to producing the first apomictic grain crop, and that it will be maize".
If these expectations come true, apomixis may undermine the basic incentive of the private seed industry. That would make legal protection of plant material ever more essential for the industry. The politics involved in apomixis research promises to be interesting.
Jeroen van Wijk

Sources
Richard C. Lewontin and Jean­Pierre Berlan (1990), "The Political Economy of Agricultural Research: The case of hybrid corn". In: C. Ronald Caroll, John H. Vandermeer, and Peter Rosset, Agroecology. Biological Resource Management Series. McGraw­Hill Publishing Company, pp.613­628.

Mary K. Knudson and Vernon W. Ruttan (1988), "Research and Development of a Biological Innovation: Commercial hybrid wheat". Food Research Institute Studies, Vol.XXI, No.1.

Jack Kloppenburg Jr. (1990), First The Seed: The political economy of plant biotechnology.

Suri Sehgal and Jan Van Rompaey (1993), Prospects for the Hybrid Seed Industry in Developing Countries. Gent, Belgium: Plant Genetic Systems.

Asian Seed and Plant Material. A bimonthly newsletter published by the FAO/DANIDA Trustfund Project, Vol.1, No.2, April 1994.