Science: Flower extract that promises lingering death: Researchers have a new weapon against malaria-carrying mosquitoes. Now they are developing pesticides from the same plant sources. David Spark reports

RESEARCHERS at Rothamsted Experimental Station, Hertfordshire, are scanning the plant world for a successor to one of their greatest finds, the pyrethroid insecticides. One new compound has already been patented.

A pyrethroid called lambda-cyhalothrin, produced by ICI under the trade name Icon, is a new weapon for curbing the growing threat of malaria in Africa and other areas where malaria parasites are resistant to drugs. It shares with other pyrethroids the advantage that, unlike DDT, it degrades swiftly in contact with the soil and so leaves no harmful residues. But it will linger to kill mosquitoes for six months on an inert surface such as a house wall.

Research in Tanzania and South Africa has shown it to be 10 times more effective than DDT for killing malaria-carrying mosquitoes in houses, according to Dr Graham White of ICI Public Health. Unlike DDT, it does not repel mosquitoes and allow them to escape. In one Tanzanian village, it greatly reduced malaria among newborn babies. It is the first insecticide to win World Health Organisation approval for malaria control since the WHO revised its evaluation scheme in 1982.

Countries in impoverished Africa have little money for spraying. But 16 countries worldwide have bought the new insecticide. One is Zimbabwe, and Dr White calculates that taking into account the low transport and handling costs for the small quantities needed, Icon is as cheap to use in that country as any mosquito killer.

The pyrethroid success was based on compounds extracted from the pyrethrum flower which kill insects but break down rapidly in sunlight. Sixteen years ago, Dr Michael Elliott and colleagues at Rothamsted discovered more stable and active pyrethroids suitable for more uses, especially in farming. These compounds have little impact on the environment. A new generation of non-ester pyrethroids - simpler molecules than the old - is being developed.

After the pyrethroids came the highly active avermectins, microbial products discovered in the United States and Japan. Their toxicity towards mammals makes them unsuitable for arable use, but they have a role in veterinary medicine.

Now research is returning to plants, which synthesise a wide range of compounds to protect themselves against their enemies. The normal approach to screening compounds is to check hundreds a week by a rapid technique such as spraying leaves and putting insects on them. Anything highly active produces a clear-cut answer, but a compound present in only minute quantities may be missed.

The Rothamsted team prepares extracts with three solvents, each dissolving a different range of compounds. This makes the effectiveness of any particular compound easier to spot. 'We consider plants a valuable resource,' Dr Bhupinder Khambay, the leader of the research group, says. 'We look at each species carefully.'

They rear insects methodically, including insecticide-resistant strains. The extracts are applied to individual insects: 'This is labour-intensive,' Dr Khambay says, 'but the test is more sensitive and the results are more accurate.'

In discovering the pyrethroids, Rothamsted used only two test species: susceptible houseflies and mustard beetles, but now it also uses mites, whitefly, the cornroot worm and the diamond-backed moth (scourge of cabbages in the tropics). 'We are on the lookout for selective compounds as well as broad-spectrum ones,' Dr Khambay says.

Dr Khambay began his current work with plants supplied by Biotics Ltd, based at Sussex University. Professor Bob Thomas set up Biotics as a bridge between plant-supplying tropical countries and the organisations that can screen plant compounds for useful properties as drugs or pesticides.

This has the advantage that royalty payments for the use of a successful compound can go back via Biotics to the country of origin. The source country can also benefit by cultivating the successful plant to produce a cheaper, local source of pesticides. Dr Khambay says: 'We have direct links in several countries, with similar arrangements.'

He points out that, apart from income from commercial successes, developing countries can gain valuable practical experience of work in Rothamsted's well-equipped laboratories.

This is the first of three articles on modern advances in pesticides.