The discreet charm of human scent

ONE ASPECT of the biological clock is that it helps to control the regularity of a woman's menstrual cycle. Research into factors that can influence this cycle has revealed the possible existence of another component of body chemistry - the human pheromone. Many animals routinely use pheromones, which are hormone-like substances released by one individual in order to affect the physiology or behaviour of another. One example is the insect pheromone released by a queen bee to prevent her hive sisters from becoming reproductively active.

Many perfume manufacturers accept without question that men and women use pheromones to communicate subconsciously, but scientists are less convinced. No one has yet isolated a body chemical that can unequivocally act as a human pheromone. Nevertheless, several research teams have produced evidence that there are pheromones at work when people interact.

The power of the human sense of smell is undeniable. A mother can identify her newborn baby or older child by smelling T-shirts worn by her offspring. But that is not the same as proving the existence of pheromones.

Nearly 30 years ago, Martha McClintock, a psychologist at the University of Chicago, produced the first tantalising evidence that there may well be body chemicals transmitted from one to another and capable of affecting human physiology. She found that the menstrual cycles of room-mates tended to converge over a period of time. Doubters suggested this could come about through other cues, subconsciously sensed by the young women.

McClintock's latest research, published last year in Nature, dispels this notion. She attempted to collect pheromones by asking a group of women to wear cotton pads in their armpits (the highest concentration of skin glands) for eight hours. She subsequently disinfected the pads and then wiped them over the upper lips of a second group of 20 recipient women. Half were exposed to pads collected from women who had already ovulated, and half from women who had not.

McClintock's team found that the biological clocks of the recipients were affected. Fourteen of the women had shorter menstrual cycles when exposed to secretions collected before ovulation, and experienced a delayed menstruation when exposed to pads from women who had ovulated. Hormone measurements showed the shift was due to a change in the timing of ovulation, with the average being two days shorter or longer, but some changed by two weeks.

Another research team, led by Karl Grammer at the Ludwig-Boltzmann Institute for Urban Ethology in Vienna, have recently demonstrated that pheromones may be at work in helping us choose sexual partners. They asked a group of non-smoking students - 16 men and 19 women - to sleep for three nights in the same T-shirts, without using deodorants. The students were then asked to rate the smell of each (empty) T-shirt on a scale of "sexiness". They were also asked to judge the attractiveness of the members of the opposite sex involved in the study. The "sexiest" T-shirt tended to coincide with the attractiveness of the wearer.

Their conclusion was that there may be pheromones that help to underpin our choice of sexual partner. Although the research, along with McClintock's, provides intriguing support for the human pheromone hypothesis, sceptics will want scientists to isolate the key chemicals involved. For the moment at least, the body chemistry of human relationships remains a mystery.