Internet users already can feast their eyes on colorful images and enticing video, and they can indulge their ears with music and other sounds. Someday--if Hamid Arastoopour, Firooz Rasouli and Ali Oskouie of the Illinois Institute of Technology (IIT) have their way--Web sites may also tickle the nostrils.

In August, the three scientists patented the design for a Tele-Aroma Drive (TAD)--a device that would allow Internet users to click on the image of a coffee cup or a flower and be showered with their aromas. Such a device would make the growing world of electronic commerce ever more realistic for consumers.

"The Internet brings things on the other side of the world to your living room," says Arastoopour, chairman of IIT's chemical and environmental engineering department. "To give that another touch of reality, smell is the next step."

Arastoopour, Rasouli and Oskouie are not the only ones chasing Web-based odors. Last month, an Oakland, Calif.-based startup company called DigiScents Inc. went public with its design for a speaker-like device that releases small amounts of scented oils into the air based on a computer user's mouse clicks. The company, which hopes to have the device on retail shelves by spring, says the system is in the testing phase.

A DigiScents device will use heat to liberate as many as 200 scented oils from a replaceable scent cartridge. A source familiar with the system said that while DigiScents' technology is much closer to fruition than the IIT scientists' version, it also is less sophisticated. Because of technical obstacles, the IIT team acknowledges that its dream is still several years from realization.

The setup in the IIT laboratory looks nothing like the ultimate product, which will be shaped much like a computer drive for floppy disks or CD-ROMs. Instead, the current model was designed to illustrate the principle.

The trio has strung wires through sealed tubes of high-grade plastic--the kind usually used for medical purposes. Inside the tubes are liquids, such as colognes. When a tube's wire is heated, the liquid in the tube becomes excited and some of the molecules vaporize. These vaporized molecules are small enough to escape from the tube through holes that are invisible to the naked eye. Once outside the tube, the molecules create an aroma--which is in turn pushed toward a volunteer's nose by a small fan.

A commercial TAD would take that same principle one step further. A computer user would insert a replaceable disk--something that will look much like an ordinary CD-ROM--into a special drive in a personal computer. (The disk would need to be replenished periodically, just like a printer cartridge.)

In the first generation of TADs, aroma molecules would be encased on or absorbed into the disk below a layer of polymer, a plastic-like chemical. As soon as a user clicked on an aroma-linked icon, electronic signals would direct the TAD's internal laser toward a given aroma pocket. The laser would heat the polymer layer, allowing the aroma molecules to escape through open pores in the heated polymer. A fan would push them in the computer user's direction.

During this phase of development, TAD disks would probably be able to hold only a limited selection of pre-mixed scents. Vendors--from perfume companies to wineries--might mass-mail the CD-ROM-style disks to consumers or include them in fashion or gourmet magazines.

The IIT team has even grander long-term plans. The researchers want to create TAD disks that contain a basic "alphabet" of smells--that is, a collection of basic chemical building blocks from which scents are constructed. If the components of that alphabet could be identified--and if a method could be developed to enable those basic components to be assembled reliably into the proper scents--then a single, standard disk could create almost any scent desired.

Eventually, as computers and televisions converge into a single piece of equipment, TAD technology might make possible a modern incarnation of "Smell-O-Vision," a process tried with little success in movie theaters in the 1950s. Under the new system, viewers could watch a movie on their home TV and smell the same pine forest or the same cloud of gunpowder as the character on screen.

The inventors face many challenges before their idea becomes a reality. The most obvious are technical challenges, which will take "not a couple of months--more like a couple of years" to solve, according to Arastoopour.

Beyond the struggles faced by other new technologies, the development of the TAD depends on major improvements in scientists' knowledge of how the sense of smell works.

"Talking to scientists, we find that we know very little," says Rasouli, who is now a research engineer with Philip Morris USA. "Every day we figure out the answer to one question, we discover that we don't know the answer to 10 others."

Nathan Lewis, a California Institute of Technology chemist and the developer of an "electronic nose," notes that odors comprise many components in complicated combinations and varied concentrations. So while it's relatively easy to distill a fruit or plant into its essential odor, it's much harder to build such essences from scratch.

"In most mixtures, we don't really know what we're smelling," Lewis says. "The scent of a strawberry may have hundreds of components. If you miss a couple of them, the mix may not smell right at all." Scientists have even found that slight differences in odor composition--differences that are undetectable to the nose--inspire markedly different responses in brain wave patterns.

The technology's success also will depend on how much consumers think they can benefit from it. Research has consistently shown that scents play a significant role in improving consumers' assessments of the products they're thinking about buying. Many retailers have taken such research to heart, wafting small amounts of pleasant scents in their stores, says Ayn Crowley, professor of marketing at Drake University.

But Crowley adds that such scents work best when they're hovering just below a consumer's consciousness. "We process scent information in a different way than information from our other senses," she says. "The other senses are wired to the thinking part of our brain, whereas smell is wired to a more emotional part of the brain."

So while scents may well make computer users feel good about the products they see on screen, the process of smelling scents over the Internet--which begins with the need to purchase a TAD and continues with the user's choice of a particular scent--is a much more active process than walking into a scented store unawares. This, according to Crowley, means that the previous studies of the impact of scents may not do much to explain the potential impact of Internet scents.

Then there is a potentially nasty problem: Devious users of the technology might be able to send stink-bomb e-mails to enemies, or, as practical jokes, to friends. Short of that, the advent of Internet scents could add to the mix of artificial odors that already pervade modern life, says Christoph Streicher, a Fairfield, Iowa-based scented-oils merchant who also is vice president of the National Association of Holistic Aromatherapy.

"It's another form of scent pollution on top of the odors we already have to cope with," Streicher says.

The trio of inventors acknowledge such drawbacks. But they prefer to emphasize a more romantic notion--one epitomized by rose-scented e-mails traded between lovers.

"Our idea," Rasouli says, "is of love and good smells."