Scientists in Philadelphia have established for the first time that the human body produces pheromones, special aromatic chemical compounds discharged by one individual that affect the sexual physiology of another.

Although animals have long been known to secrete pheromones, which typically function as sex attractants, and although the existence of such chemicals in humans has long been speculated, the new research is the first to establish their existence in humans.

The human pheromones are not sex attractants, nor do they act almost immediately as animal pheromones do. Instead, the human pheromones act over a period of weeks or months to alter the timing of women's menstrual cycles.

The discovery, by scientists at the Monell Chemical Senses Center and the University of Pennsylvania medical school, is believed to help explain a long series of puzzling findings -- linking sexual behavior and the health of a woman's reproductive system -- that the researchers have obtained in a landmark series of studies over 13 years.

In those studies it was found that women who have sex with men at least once a week are more likely to have normal-length menstrual cycles, fewer infertility problems and a milder menopause than women who are celibate or who have sex in a sporadic "feast-or-famine" pattern.

The pheromone findings indicate that an essential factor, aside from sexual intercourse itself, is exposure to specific aromatic chemicals exuded in a man's normal body odors. When a woman receives these chemicals, by smell or skin absorption, even though she may not consciously notice them, they automatically improve her physiological functioning.

The evidence suggests that the male chemicals, which are secreted in special sweat glands in the armpits, and possibly also around the nipples and in the genital region, are not effectively transmitted except in the intimate contact ordinarily associated with sex.

Although the chemical nature of the substances is not yet fully understood, the scientists hope to create synthetic versions for various practical applications. Among the possibilities are nasal sprays to correct certain forms of infertility, to regularize the menstrual cycle, to make the rhythm method of birth control more reliable and to delay or ameliorate menopause.

The Philadelphia researchers have been able to duplicate some of the male pheromone's effects by exposing women, who had no current sexual relationship, to male pheromones in the form of what they dubbed "male essence." This contained substances extracted with alcohol from absorbent pads that male volunteers wore under their arms.

The study was done with female volunteers whose menstrual cycles were longer than 33 days or shorter than 26, deviations from the normal average of 29.5 days. Three times a week the women came to the clinic to have male essence rubbed under their noses. The women said they could smell only the alcohol.

After about 12 to 14 weeks, their menstrual cycles changed, slowing in some cases and speeding up in others, to approximate 29.5 days. A control group of similar women who were dosed only with plain alcohol showed no significant change in cycle length.

In a related experiment, the scientists found that women produce another pheromone that can cause other women's menstrual cycles to shift into synchrony. The researchers used a similarly obtained "female essence," which was collected at measured intervals during the donor's menstrual cycle and administered in the same sequence. Within a few menstrual cycles, the recipients shifted their cycles to synchronize with those of the donor.

This finding confirms widely reported anecdotes that women who live or work together eventually develop synchronized menstrual cycles.

Although claims of discovering a human pheromone are not new, the older claims have not been based on controlled experiments and most scientists have not found the arguments persuasive. The new findings are to be published next month in Hormones and Behavior, a prestigious, peer-reviewed scientific journal.

"I think we've finally answered the question. Pheromone effects are real in human beings and the anecdotal evidence suggests they even occur here in the U.S. where we're all deodorized and perfumized," said George Preti, who collaborated on the research at the Monell center with Winnifred B. Cutler. Monell is an independent research facility devoted to the chemically stimulated senses of taste and smell. Cutler is an authority on the relationship between human sexual behavior and hormones and Preti specializes in the chemistry of hormones and external secretions of the human body.

One distinction between the male and female pheromones is the distance over which they act. While the female pheromone can diffuse through a large room and still be strong enough to synchronize menstrual cycles, the male pheromone requires intimate contact.

"Social distances don't seem to work for the male pheromone," Preti said. He said the volunteers all had ordinary social contact with men but not necessarily intimate relations. "Women don't get the effect working in offices or classrooms with men. It has to be intimate contact."

Although little is known about how human pheromones work, the scientists say molecules of the substance must reach receptors of the nervous or endocrine systems in or around the nose and somehow act upon them to trigger signals to the brain. The brain, in turn, relays signals to influence the endocrine system of glands and hormones.

The menstrual cycle is known to be under the control of a network of glands, including some in the brain.

Although the pheromone findings are new and have not previously been reported, the evidence of a link between heterosexual behavior and women's reproductive physiology has been published, with little public notice, in a series of reports over the last eight years in various scientific journals.

"It's remarkable. A very clear pattern has been emerging and it all confirms that a woman's optimal reproductive health is a part of a finely tuned system and that a man, on a regular and sustained basis, is an essential part of it," said Cutler, who has led the research effort. "It wasn't clear until our most recent studies how important male essence really is," she said, "but now that we know this, it helps to explain our earlier findings. You might say that exposure to pheromones is the essence of sex."

One of the early hints emerged in 1979 from a study by Cutler and her colleagues at Penn, Celso R. Garcia, a professor of gynecology and a pioneer in the development of oral contraceptives, and Abba M. Krieger, a statistician.

The researchers asked several hundred female college students to keep records of their menstrual cycles and sexual behavior over a period of several months. Privacy was maintained through elaborate double-blind procedures.

When the results were analyzed, they showed that among women who had sex with a man at least once a week, 78 percent had cycles within the 26- to 33-day normal range.

Among women who reported no sexual activity, only 59 percent were in the normal range. Among women whose sexual activity was sporadic, 49 percent were in the normal range.

A more detailed analysis revealed that orgasm alone was not the key factor. Women who achieved orgasm through self-stimulation were not predisposed to more nearly normal cycles unless a man was present. Until the pheromone effect was later discovered, these findings were a puzzle, Cutler said.

In another study, the researchers found that women who had sex less than weekly were more likely to have abnormally short or abnormally long, menstrual cycles. Short cycles are often linked to one type of infertility. Long cycles are sometimes fertile and sometimes not.

A related study examined the link between sexual behavior and a woman's basal body temperature fluctuations. Fertile women usually have a characteristic rhythm of temperature changes during each menstrual cycle. Infertile women have a different pattern.

Cutler and her colleagues found that among women who had sex with men at least weekly, 90 percent had the fertile type of temperature rhythm. Among women with sporadic sexual activity only 55 percent had such a rhythm. The proportion dropped to 44 percent for celibate women.

Yet another finding was that women who had regular weekly sex with men had significantly higher levels of the female hormone estrogen in their blood than women who had either sporadic sex or none at all.

"Although the pattern of sexual activity later in life is important, we also found that it matters when a woman begins her sexual life," Cutler said. Her research shows that women who consulted a fertility clinic because they had never been able to get pregnant did not have their first sexual intercourse until an average of about eight years after they began menstruating. The average time of first intercourse for women seeking routine gynecological exams was just over six years after they began menstruating. The difference is statistically significant.

Cutler suspects there is a critical stage in the maturation of a woman's reproductive system during which intimate exposure to a man helps the system achieve normal fertility.

The end of a woman's reproductive years -- the menopause when the level of sex hormones declines, causing hot flashes, and the reproductive system shuts down -- also seems linked to the regularity of her sexual activity.

Women who reported having regular weekly sex had fewer and milder hot flashes than did women who had less sex or none. In this case, however, Cutler cautions that the cause-effect relationship is not clear. It may be that as a woman's sex hormones decline and the resultant hot flashes grow severe, she loses interest in sex. Or it may be that regular sex keeps a woman's hormone levels up to par, protecting against hot flashes and allowing sexual interests to continue.

"If you look at the data from all our studies, the conclusion is compelling," Cutler said. "A man, or his essence, seems essential for an optimally fertile system."

Cutler's personal views go even further. She holds that the health of a woman's reproductive system is a sensitive measure of her overall health. Countless animal experiments have shown that when females are exposed to stress of various kinds, from noise or crowding to starvation, the first part of the body to shut down is the reproductive system.

"When you see reproductive viability slowed down in a woman," Cutler said, "something is wrong with the individual as a whole. For most women, regular sex with a man would seem to be an important part of good health."