Genetics can feed the starving millions

There are excellent models for studying animal genetics - such as the fruit fly, drosophila - and we humans rely heavily on the mouse. But for plants, what do we have? Wheat, rice and maize can turn for basic understanding of plant genetics and development to an inconspicuous weed, Arabidopsis thaliana ; it's been used in classical plant genetics for more than 45 years, and is now a key model in laboratories the world over. Its advantages include its small size - several can be grown in a square centimetre; its short generation time - six weeks; just five chromosomes; and a relatively small number of genes - about half that of the fly. Its DNA is being sequenced, and will be completely known next year.

One hero in the story of arabidopsis is Friedrich Laibach, who in 1907 published a paper on its chromosomes. Why he chose this plant is not known, but when he became professor of botany at the University of Frankfurt he wrote, in 1943 - amazingly, in the middle of the Second World War - how valuable arabidopsis was as a research tool, and listed its advantages. Its introduction into wider use was stimulated by a Hungarian, GP Rede'i, who came to the US in 1957 and had to struggle to get the plant's usefulness recognised. At an international meeting on arabidopsis in Australia this year, 500 scientists attended.

It is quite easy to induce new mutations in arabidopsis , giving insights into many basic mechanisms governing plant life. Some are rather surprising; for example, a gene was found that is active for only a few hours around dawn; it is part a circadian clock that governs flowering-time in relation to day length. Other mutations have given insights into how plants develop. Unlike in animals there is no cell movement, but mutations can change one structure into another - homeotic mutations - common in cultivated plants such as ornamental flowers, including roses and camellias. Homeotic mutations convert stamens to petals, and so give larger and more beautiful flowers. Study of mutations has also shown why some important features of the green revolution worked.

Since the Sixties the world output of food has probably doubled, and about half of the increase is due to the introduction of new varieties. One aim of the plant-breeders was to produce dwarf plants whose height was reduced, so they didn't grow tall and fall over, but whose pods stayed just as big. Selecting for such plants enabled farmers to add more fertiliser and get greater yields without the undesirable increase in height. The basis of this reduction in height is the reduced ability of the plant to respond to a growth hormone, gibberellin. Recent work on arabidopsis has shown that this is the result of mutations in genes that code for proteins, which activate the growth of the plant when the gibberellin binds to its special receptor, and so results in a dwarf form.

The green revolution was based on conventional plant breeding, but it is hard to see why there should be worries if the gene controlling height of the plant had been knocked out by genetic modification (GM) using molecular techniques.

There are those in green and anti-GM movements who think that there is no real food shortage in the world, and that famine and hunger are due to grossly unfair distribution of food, and to war. But the best estimates are that the world's population will increase from 6 billion to 9 billion in the next 25 years, which will require food production almost to double.

One estimate from China is that by the year 2020 China will want to import the whole of the United States grain production for its own use. Arabidopsis could be a very helpful model for modifying plants to increase output, if such techniques are in fact permitted to be used.

The writer is professor of biology as applied to medicine at University College London