Each nostril smells the world differently.

Much as each eye sees the world from a slightly different angle, each nostril takes a somewhat different sniff, reports a California-based research team.

Unlike the eyes, however, nostrils switch roles several times a day, seesawing peak sensitivity between two groups of odors, the researchers argue in the Nov. 4 NATURE.

The basic difference between left- and right-nostril sniffs depends on airflow, explains Noam Sobel of Stanford University. A little tissue bulge, called a nasal turbinate, dangles in each nostril. While one turbinate engorges with blood and chokes down airflow, the other shrivels to permit big sniffs.

To feel the difference, block off each nostril in turn and inhale, Sobel says. Checking again several hours later often reveals a switch. To see the turbinates, just look up somebody's nose, Sobel advises.

Researchers have known for more than 100 years that nostril airflows differ, notes Sobel. What's new, he says, is the evidence that this nasal quirk affects sensitivity to odors.

To register as a smell, molecules wafting into a nostril must cross a mucous membrane and hit a receptor. In earlier studies on bullfrogs, other researchers found that some compounds, so-called high-sorption odorants, zing through that membrane quickly, but others only creep.

To see if airflow affects perception of either odorant class, Sobel and his colleagues asked Stanford undergraduates to sniff a mix of two compounds and estimate their ratio. One component, the pepper-minty L-carvone, zips across the nasal membrane, but the other, the aniselike odorant octane, dawdles.

Although they told students that proportions of the compounds would vary, the experimenters kept the mix at 50-50. For each trial, the researchers measured the flow of air as a student sniffed through one nostril.

Airflow did change the sensitivity, the team reports. Seventeen of the 20 students ranked the slow-traveling anise higher when they inhaled through their low-air nostril. When sniffing through the high-air nostril, these 17 ranked the fast peppermint at a higher proportion.

To see if some peculiarity of a nostril caused the difference, the researchers retested eight of the students after their airflow patterns had switched. Seven showed the same link between odor perception and airflow as in the earlier trial.

"It's not that one [nostril] smells oranges and the other smells apples," Sobel emphasizes. "The difference is subtle."

Richard Doty of the University of Pennsylvania in Philadelphia praised the creativity of the experiment but pointed out that Sobel used only a two-component mixture, while most smells are far more complex.

Many people don't show consistent airflow-change cycles, and the cycles dwindle with age, Doty has found. He muses that airflow response "is more akin to visual illusions that, while interesting, play little role in day-to-clay visual processing."

However, study coauthor John D.E. Gabrieli of Stanford proposes that nostril shifts boost nose power by allowing two simultaneous sniffs that have their sensitivities tuned to different kinds of chemicals. As he puts it, "Two heads are better than one." Gabrieli also raises the possibility that slow- and fast-moving compounds could provide a key to understanding how olfactory processes are organized in the brain.

Brain mapping is only one aspect of the study of smell that lags behind studies of other senses, grumbles James M. Bower at California Institute of Technology in Pasadena. "There's a national eye institute, but there's not a national nose institute," he points out. He welcomes the Sobel study as a useful step. "People have known that they have two nostrils, but no one has known that it mattered," he says.