SCIENCE / Needs musth when love calls: New research is beginning to shed light on the crucial part chemicals play in the elephants' elaborate mating game. Colin Tudge reports

AJAY DESAI, wearing flip-flops and shorts, approaches a wild bull elephant in the forest of India's Mudumalai wildlife sanctuary, and tosses a stick at it with a rag tied to the end. Normally, this would be suicidal: elephants are erratic, often bad-tempered, have no natural love for human beings and are absurdly strong. One casual brush with the trunk can prove fatal.

But when Desai tosses the rag at him, this young bull stops in his tracks like a cartoon cat, his front legs thrust out in terror. He flattens his ears to his head, turns sideways in submission, then turns and runs, all dignity gone.

The rag is soaked in urine from a big old bull in musth; the elephant's equivalent of rut in a stag. The urine contains a pheromone, a chemical signal which says to elephants everywhere: 'I am in musth, I am in peak condition, I am seeking mates, I will attack anyone who gets in my way, and if I get into a fight I will see it through whatever the cost.'

Faced with this olfactory message, bulls that are not in musth retreat. And, as Desai has found, the pheromone works even when delivered by a puny human: 'With elephants,' he says, 'seeing is not believing. Smelling is believing.'

Musth is connected with mating: bulls that are not in musth can mate, but not as often or successfully as those that are. Musth is brought about by a huge surge of the male hormone testosterone; an elephant in musth is like an athlete on drugs - strength and aggression enhanced beyond what is normally possible. They come into musth approximately once a year; younger ones may be in musth only for a few weeks, but in old bulls it lasts up to five months.

The state of musth is visible: between the eye and the ear, elephants have a temporal gland which swells during musth and secretes a sticky fluid that leaves a dark stain. Even more conspicuously, bulls in musth dribble urine continuously from a half-extended penis. Joyce Poole, studying African elephants in Kenya's Amboseli reserve, calculated that in a long musth, a bull dribbles no less than 400 litres.

Being in musth is a huge physiological burden. Such bulls are restless, and expend a lot of energy without feeding as well as they should. But what is musth really for? Why, in particular, has natural selection favoured this massive outflow of urine and accompanying loss of peak physical condition?

Most biologists are happy with the explanation provided by Desai's demonstration: musth pheromone repels rivals, leaving the mating bull with a clear field. But Desai, carrying out research for the Bombay Natural History Society, argues that this cannot be the purpose of the signal. 'The real competition between the bulls would take place wherever the oestrous cow happens to be,' he says.

'But once the bulls are gathered around the cow, the secretion of the temporal gland would suffice to show which one is in musth, and is therefore not to be tangled with. Any bulls that are smaller than the musth bull would simply give way. A bull of the same size, but who was not in musth himself, would also give way. But a very big bull would not give two hoots about the signal from the musth bull, because he would win anyway.'

So the secretion of the temporal gland would help the musth bull to some extent in the mating game. This secretion is cheap, physiologically speaking. It is the urine signal that is so costly. And why give so much advance warning to rival bulls? What benefit would there be to the signaller? How could natural selection favour such a costly strategy if it brought no benefit?

Desai concludes that the pheromonal signal in the urine must be aimed not at other bulls, but at the cows. It produces the same effect that sheep farmers achieve by introducing 'teaser' rams: it brings the females into oestrus. 'Either the pheromone induces ovulation,' says Desai, 'or, more likely, it simply shortens the oestrus cycle.'

Joyce Poole at Amboseli reached the same conclusion - but Desai has extended the idea. Female elephants come into oestrus only once every four months, and then they are receptive only for the four days or so around the time of ovulation. However, the birth interval in the wild is about four-and-a-half years; and simple arithmetic shows therefore that only 22 per cent of females in any one area at any one time will even be able to become pregnant. Cows live in herds with their own and their sisters' offspring while adult bulls wander, often alone, in search of females in oestrus. When in musth, the bulls increase their territory several fold, so as to encounter more herds of females.

Each big musth bull stays in the vicinity of any one herd for about

a month. If he simply wandered about, without 'seeking' to stimulate the females into oestrus, then

his chances of encountering a receptive female would be low. But if he can bring oestrus forwards by just a few weeks, his chances greatly increase; enough, at least, to ensure that natural selection would favour the bulls that can do this, even at great cost to themselves.

Desai's explanation makes more and more sense the longer you examine it. Among bull elephants the prize - the cows - goes to the biggest; and the biggest are the oldest, for they continue to grow in height after puberty, and put on weight until they are geriatric. Furthermore, their tusks grow faster than their bodies. A 50-year-old tusker is formidable indeed: very big, and often with proportionally huge tusks. The bigger they are, the greater their physiological reserves; the longer they can stay in musth; and the more females they will mate. This suits the females, who seek genes from the males who have lived longer and are therefore most obviously successful.

Younger, smaller bulls of course miss out. Their periods of musth are shorter, and in head-to-head tussles with the older males they are sure to lose. With Asian elephants, young versus old is three tons versus six. With Africans, it could be four versus 10; and weight is what counts.

However, if young bulls delay their own musth until after the old bull has been and gone, then they are able to cash in on the old one's stimulation of the cows. Old bulls do not tolerate young rivals; but they do tolerate juveniles. Very young bulls, in Desai's experience at least, tend not to dribble urine during their first attempts at musth; instead, they pass themselves off as juveniles and so avoid the wrath of the patriarch. Then, while he isn't looking, they occasionally manage a sneak mating.

Desai seeks now to test this hypothesis, by exposing cows in zoos - far from the scent of wild male elephants - to pheromones, to see if this indeed shortens the periods between oestrus. His work shows once again how wild animal behaviour and ecology are moulded by natural selection. It will also be of great practical value. Pheromones might in principle be used to control elephant movements in the wild, which becomes more and more important both in Asia and Kenya as clashes between elephants and farmers become more frequent. His work would also enhance captive breeding programmes, in camps, reserves, and zoos - again, increasingly necessary as the wild herds are robbed of habitat, and are poached.

Meanwhile, I again find myself in admiration of modern field biologists. Pheromone or no pheromone, I would not approach a wild bull elephant with a rag on a stick.

Colin Tudge's book on genetics, 'The Engineer on the Garden', is published in October by Jonathan Cape.