Roses are blue, violets are red (2024)

Alamy

BEAUTIFUL flowers—like beautiful women—can separate the most sensible of men from their money. Those men invest in the reproductive organs of plants such as roses to signal, albeit coyly, analogous intentions of their own.

The result is a cut-flower industry in which roses alone are worth $10 billion a year. But that is peanuts compared with what happened in the past. In 17th century Holland, tulips (the fashionable flower du jour) grew so expensive that people exchanged them for houses. One bulb of the most sought-after variety, the flaming red-striped Semper Augustus, sold for twice the yearly income of a rich merchant.

For modern flower growers, the equivalent of the Semper Augustus is the blue rose, which horticulturalists have longed for since the Victorian period. Any blue rose sent on St Valentine's day this year will have been dyed. But if Yoshi Tanaka, a researcher at Suntory, a Japanese drinks company, has his way, that will soon change. Dr Tanaka is currently overseeing the first field trials of a blue rose developed by Suntory's subsidiary, Florigene. If the trials are successful, a dozen blue roses—even if they do look slightly mauve—could, by 2010, be what separates an unsuccessful suitor from Prince Charming.

Flaming tulips. Blue roses. What Dutch growers of old and Dr Tanaka's employers both grasped is that rarity, and hence economic value, can be created by genetic manipulation.

The stripes of the Semper Augustus were caused by the genes of a virus. Not knowing that an infection was involved, the Dutch growers were puzzled why the Semper Augustus would not breed true. The genetics of blue roses, too, have turned out to be more complicated than expected. The relevant genes cannot easily be pasted into rose DNA because the metabolic pathway for creating blue pigment in a rose consists of more chemical steps than it does in other types of flower. (Florigene has sold bluish genetically modified carnations since 1998.) Success, then, has been a matter of pinning down the genes that allow those extra steps to happen, and then transplanting them to their new host.

Buy any other name

Mere colour, however, is for unsophisticated lovers. A truly harmonious Valentine gift should smell beautiful as well. Sadly, commercial varieties of cut rose lack fragrance. This is because there is a trade-off between the energy that plants spend on making the complex, volatile chemicals that attract women and insects alike, and that available for making and maintaining pretty-coloured petals. So, by artificially selecting big, long-lasting flowers, breeders have all but erased another desirable characteristic.

Smell is tougher to implant than colour because it not only matters whether a plant can make odoriferous chemicals, it also matters what it does with them. This was made plain by the first experiment designed to fix the problem. In 2001 Joost Lücker, then a researcher at Plant Research International in Wageningen, in the Netherlands, added genes for a new scent into petunias. Chemical analysis showed that the new scent was, indeed, being made, but unfortunately the flowers did not smell any different. As happens in Florigene's blue carnations and roses, Dr Lücker's petunias dumped the foreign chemical they were being forced to create into cellular waste buckets known as vacuoles. Whereas pigments are able to alter a petal's colour even when they are inside a vacuole, because the cell contents surrounding the vacuole are transparent, smelly molecules must find a route to the sniffer's nose by getting out of the cell and evaporating.

Like Dr Lücker, Natalia Dudareva, of Purdue University, in Indiana, eschews experiments with roses, since these plants have scents composed of 300 to 400 different molecules. She prefers to understand basic odour science using petunias and snap-dragons, which have about ten smelly chemicals apiece. She has made an encouraging discovery. By studying the many different pathways through which flowers make their fragrances, she has found consistent patterns in the way these pathways are regulated.

Such co-ordinated patterns suggest that a type of protein called a transcription factor is involved. Transcription factors switch genes on and off in groups. If Dr Dudareva is right, cut roses have lost their fragrances not because the genes that encode their hundreds of scent molecules have each lost their function, but because the plants no longer make a few transcription factors needed to turn the whole system on.

This suggests that the task of replacing lost fragrance is more manageable than it seemed at first blush. But even when the transcription factors in question have been identified, the problem of the energetic trade-off with pigment production and longevity will remain. So Dr Dudareva is also measuring how quickly the enzymes in scent-production pathways work, in order to identify bottlenecks and thus places where her metabolic-engineering efforts would best be concentrated.

Dr Dudareva's methods may also help to improve the job that flower-scents originally evolved to do—attracting insects that will carry pollen from flower to flower. By modifying the smell of crops such as vanilla, which have specific pollinator species, different insects might be attracted. That could expand the range in which such crops could be grown and thus make some poor farmers richer. A change, then, from making rich but romantic men poorer.