A study released this week tracing some instances of male homosexuality to the "older brother effect" is more than another entrant in the race to find the biological roots of sexual orientation. The mechanism it points to fits with emerging research on the powerful effects of conditions in the womb.

The theory dates to 1996, when scientists reported an odd correlation. For each additional older brother that a boy has, his chance of growing up to be gay increases by one-third. The correlation doesn't explain all homosexuality -- many of the estimated 7 million gay men in the U.S. have no older brothers, and most younger brothers are straight. But if the conclusion is right, a rough calculation shows that "about 1 million American men are gay or will grow up to be gay because their mother had sons before them," says psychologist S. Marc Breedlove of Michigan State University, East Lansing, who studies sexual orientation.

How having older brothers increased the chance of being gay wasn't clear. One suspect was sibling dynamics. Perhaps later-borns are so desperate to differentiate themselves from their brothers that they elect, consciously or unconsciously, to be gay. (There is no known sibling effect on female sexual orientation.)

To see if such psychological factors are at work, a scientist who discovered the older-brother effect studied 944 men in a melange of families -- biological brothers growing up together, biological brothers reared apart (due to adoption, usually), stepbrothers and half-brothers, and unrelated boys reared together. The idea was to separate the psychological effects of growing up with older brothers from the biological effect of gestating in a womb that bore boys before.

Growing up with older stepbrothers has no effect on whether younger brothers will be gay, Anthony Bogaert of Brock University, St. Catharines, Canada, concludes this week in Proceedings of the National Academy of Sciences. But having an older biological brother, even one who grows up separately, does.

That means the mechanism by which having older brothers affects sexual orientation isn't social or psychological, let alone one that involves conscious or unconscious choice. Instead, the biological fact of being the second (or later) occupant of the same womb make it more likely that a boy will grow up to be gay. Somehow, the prenatal environment fosters homosexuality in later sons.

The older-brothers study adds to the growing research concluding that some people are born to be gay, although none are definitive.

A 1993 study, for instance, linked a region at the tip of the X chromosome, which boys inherit from mom, to a higher likelihood of homosexuality. But the finding hasn't been replicated by other labs, and no "gay gene" has been found.

Another suspect is sex hormones. Lesbians tend to have masculinized fingers (the ratio of the index finger to the ring finger on the right hand is smaller than in straight women) and other traits caused by higher exposure to fetal testosterone than straight women got.

In men, the sex-hormone picture is murkier. Some anatomical legacies (finger-length ratios) of fetal testosterone suggest gay men were exposed prenatally to more of this hormone. Others (brain organization) suggest they got less. Most indicate no difference in exposure between gays and straights.

Even if fetal sex hormones can't explain homosexuality, something else in utero might. Research has found a strong effect of the uterine environment on a fetus's risk of developing adult-onset diseases. Thanks to "fetal programming," a low birthweight raises the chance that a man will die of heart disease; it also raises the risk of Type 2 diabetes.

What seems to happen is that undernutrition in utero adversely affects how the liver, heart and kidneys function forever after. It reduces the number of kidney cells, for instance, forcing each to work overtime to regulate blood pressure. The cells wear out, causing adult hypertension and heart disease.

Something similar may affect sexual orientation. When a woman is pregnant with her first son, the placenta keeps most proteins in his blood from reaching her circulation. But giving birth mixes fetal and maternal blood. Her immune system, exposed to foreign (male) proteins made by Junior's Y chromosome, may produce antibodies to them.

Because antibodies cross the placenta, fetuses in subsequent pregnancies would be exposed to them. Prof. Breedlove speculates that "these maternal antibodies might somehow affect brain development in the younger son, increasing his chance of growing up to be gay."

Looking in the brain for the roots of sexual orientation makes sense because it is the organ of behavior. Indeed, a 1991 study of 41 brains found that in gay males a tiny region behind the eyes, in an area crucial for sexual function, is smaller than in straight men. The genesis of the smaller size was a mystery. Mom's exposure to the male proteins of her first son might explain it, but only in part: A man would need a dozen older brothers to have a 50-50 chance of being gay. As Prof. Breedlove says, "There are probably many ways biology can produce a gay man."