The grass pea is a remarkable plant. It thrives in countries such as Bangladesh, China, Ethiopia and India when other crops fail owing to drought or flooding. If there is famine, it becomes a survival food. Except that there is a catch. Eaten in large amounts, the grass pea's natural neurotoxins cause irreversible paralysis below the waist.

That is why international scientists based in Syria are seeking to reduce or eliminate the toxin, employing the techniques of biotechnology, such as genetic modification. They have made good progress towards a grass pea that can help people to survive without wrecking the lives it saves.

Those who see biotechnology as an important tool for feeding the world believe the grass pea may be just the first of many unconventional and subsistence crops that could benefit from scientific research. But they may never discover whether their hunch is right. For one thing, many people object to genetic modification on environmental and ethical grounds. For another, the research depends on public sector research and the public sector is desperately short of funds.

The poor and hungry are always with us. But the question of unconventional crops has taken on a new urgency in the past few years.

This is partly because biotechnology has created new tools and opportunities, which some scientists say must be seized. Behind these calls are issues highlighted by the international wrangling over Aids drugs - how to deliver to poorer countries the technological wealth of the rich world, which is often protected by strong patents.

The urgency is also partly because the population is growing, while the scope to put more land under the plough is limited by such things as urban sprawl and desertification.

"With no new land available for cultivation, the only option lies with increasing yields," says Philip Pardey of the International Food Policy Research Institute, a non-profit organisation in Washington, DC. "I don't think that has hit home with policymakers."

Furthermore, researchers take 10 years or more to create new crop varieties. Some scientists argue that if biotechnology is to have an impact within the next 30 years, when the world population is predicted to increase from 6bn to 8bn people, the funding must flow soon.

"It is very urgent," says Paul Christou, a trainer of developing world scientists at the John Innes centre, Europe's leading public sector plant research establishment.

"It should have happened before. But now it is the 11th hour."

In spite of such arguments, it is by no means certain that biotechnology will be used extensively to develop subsistence crops in the poorer regions of the developing world.

The heated arguments over its benefits and risks continue. Industry and pro-biotech development agencies claim that while the technology is no "silver bullet", its promise of increasing crops yields makes it vital to future global food security.

Environmental and development pressure groups do not dispute the worth of projects such as the grass pea but oppose the general use of biotechnology on the broader grounds of corporate monopoly control and possible environmental damage.

"More public research is needed but to boost it simply in response to biotech is perverse," says Patrick Mulvany of the Intermediate Technology Development Group, one of many aid agencies pressing for policy to centre on the needs of small farmers.

Even if the biotech lobby wins this argument, it will still be left with the challenge of living up to promises that the technology can be applied to important, but non-commercial, subsistence crops such as sweet potato, yam, millet and cassava.

This challenge is one in which the public and private sectors are inextricably linked. Industry has used the food security argument as part of its public relations strategy to secure acceptance of genetically modified crops in international markets.

However, the private sector's efforts are on bulk crops such as maize and rice that are grown in the largest and potentially most profitable markets.

Philanthropic arms, such as the Syngenta Foundation for Sustainable Agriculture, are involved in projects such as "golden rice", an attempt to increase the crop's vitamin A content. However, they accept the limitations on what is possible.

Klaus Leisinger, interim director of the foundation, said in a recent paper: "For private industry, a focus on profitable markets is necessary for survival.

"Some people can regret this reality - but then they must move on to look for alternatives. The alternative to private sector research is public research."

However, the public sector is unable to contribute much. In the past 20 years public agricultural research has fallen from fashion and has been scaled back, particularly in developing countries.

"Most of the 20th century saw substantial increases in agricultural research around the globe," says IFPRI's Mr Pardey. "Now there is an unprecedented slowing down."

IFPRI itself is a case in point. It is one of 16 international research centres run by the Consultative Group of International Agricultural Research, veterans of the Green revolution of the 1960s and 1970s and funded by national governments and international bodies such as the World Bank.

Many of its centres, which are highly respected, have embraced biotechnology within "public good" principles of equity, with regard to long-term impact and appropriate trusteeship of the ownership of genetic resources. Their research projects include the grass pea, genetically modified bananas and golden rice.

However, the CGIAR's annual budget of about Dollars 330m (Pounds 232m) - of which Dollars 25m now goes on biotech - is stagnant and cannot compare with the priv-ate sector research spending - more than Pounds 10bn in 1995, according to the United Nations Development Programme's annual report, published this month.

Outside the CGIAR centres, the situation is often worse. Many developing countries, notably in sub-Saharan Africa, lack the capacity to carry out their own research, particularly in applying the work of others to suit local conditions. In countries where there is local expertise, such as India, funding frequently remains a problem.

The shortage of funds is in spite of strong evidence that money invested in research pays high dividends. According to the UNDP, research produces an average rate of return of 44 per cent in terms of increased crop production worldwide.

The UNDP report called for extra spending on public research, to be raised by increases in overseas aid from donor governments and contributions from private industry. Sub-Saharan countries could finance more work by cutting arms spending, it said.

To fill the gaps over subsistence crops, much of the future lies with agreeing partnerships with organisations such as the CGIAR and philanthropic bodies such as the Rockefeller Foundation brokering deals between public and private sectors, industry and national governments. Much of this will require companies to give away technology rights in exchange for public relations benefits.

Other models to encourage private sector participation - such as direct subsidies - may have to be found. But the need for more public spending is unavoidable, Mr Pardey argues.

Nevertheless, most people feel that the prospects of more money for public research in the foreseeable future are bleak.

"It seems things will have to get worse before they get better," Mr Leisinger says. He believes the world will not pay attention until it witnesses increasing poverty-driven migration, political upheavals, human disasters and environmental destruction on such a scale that it can avert its gaze no longer.

** NOTICE: In accordance with Title 17 U.S.C. Section 107, this material is distributed for research and educational purposes only. **