Fragrance companies have invested hundreds of millions of dollars on the assumption that scent is a major factor in attraction. But perfumes mask one's natural odor, and that could be a bad thing when it comes to choosing good genes. Time and again, so-called t-shirt experiments have shown a link between body odor and differences between the major histocompatibility complexes (MHCs) of the sniffer and the sniffed.
In a typical test, a subject will rank the pleasantness (or otherwise) of t-shirts worn overnight by members of the opposite sex. The t-shirt picks up the scent of its wearer, the result of a complex, poorly understood interaction between the MHC-mediated skin compounds and the microbial fauna that feed upon them.
Despite strong and surprising evidence that the differences people perceive have a genetic basis, it's unclear what they actually do...
"We don't know whether this is due to differing methodologies or differences in the study populations," says Craig Roberts of Liverpool University who performs such studies. Offspring might have better immunity if their parents have different MHCs, but similar alleles at the MHC might signify better overall genetic compatibility. There are experimental results in favor of both hypotheses, with other studies even suggesting that something between the two extremes, perhaps representing a good compromise, is best.
The MHC seems to make its presence felt through sight as well as scent. While studying the odors and responses of identical twins (which have the same MHC but probably a different microfauna), Roberts discovered that the effect also held for measures of facial attractiveness: People prefer the faces of individuals with similar MHC alleles. "The mystery now is how we marry a preference for MHC similarity in faces with a disassortative preference based on odor," he says. "Perhaps the facial preference puts you into the ballpark of a genetically complementary mate."