Arose by any other name will smell as sweet, but a drug by another name might not sell as well.

Giving names to prescription drugs and deciding what to color and shape them is less of a "fly by the seat of your pants" endeavor than it used to be, but flights of fancy are still more common than standardized industry-wide guidelines.

Take Thorazine, for example, the first of a class of tranquilizers called phenothiazines, introduced in the 1950s, that dramatically changed the treatment of mental illness. The suffix 'zine' was easy; that came from the name of the chemical type. But, says Jeremy Heymsfeld, director of corporate information at SmithKline & French Laboratories, the prefix 'Thor' for the Norse god of thunder was chosen by the company's president "to reflect the reverberations that would roll across the medical world as a result of this revolutionary new drug."

Most drug names, though, don't have such thundering romance in their backgrounds. In fact, they're pretty prosaic and reflect either the generic name or the pharmacological action of the drug.

For instance, SmithKline's ulcer drug Tagamet was also the first of a new class of drugs, an antagonist to stomach acids called cimetidine.

Indocin, Merck Sharp & Dohme's popular arthritis drug, derives from its generic name, indomethacin, and Benemid, used to treat gout, is from probenecid. Decadron, a steroid, is named for dexamethasone, and Diuril is a drug that promotes diuresis, the elimination of excess fluid from the tissues.

Sandoz Pharmaceuticals developed cyclosporine, which suppresses the immune system, to prevent rejection of newly transplanted organs. They named the drug Sandimmune. Logical, right?

One of Sandoz's several migraine remedies, a combination of caffeine and ergotamine, is named Cafergot. Logical, too.

Then what about Fiorinal, another headache remedy composed of aspirin, caffeine and a barbiturate? "That was named for Montefiore Hospital in New York, where Dr. Arnold Friedman developed it and did the research," says Wayne Pisano, Sandoz's headache-line product manager. Also logical -- in a different way.

When Sandoz decided to manufacture Fiorinal with an aspirin substitute, they named it Fioricet for acetaminophen. Again, it made sense.

So too with Mellaril, a tranquilizer to make you feel mellow; Elavil, an antidepressant to elevate your mood; Ornade, an oral, nasal decongestant; Librium, another tranquilizer to set you free of anxiety; Larodopa, Hoffman-LaRoche's brand of levodopa (LaRoche and levodopa, get it?), a drug to treat Parkinson's disease; Ethril, Squibb's brand of the antibiotic erythromycin, and Mevacor, Merck's new drug, lovastatin, to lower blood cholesterol. Wait a minute -- Mevacor doesn't sound like anything! "That's just the point," says George DiDomizio, director of creative services at Merck. "Mevacor is the first drug that we've named using the 'blank canvas' approach." In other words, give the drug a name that's easy to spell, pronounce and remember but that means absolutely nothing and bears no relation to the generic name or the function of the drug.

Merck and other companies are moving to the blank canvas approach for two reasons: Most major pharmaceutical companies sell on an international market and therefore need names that mean nothing in any language. Moreover, the trademark (the brand name of the drug) is usually registered long before the drug is ready to be approved by the FDA and marketed.

In fact, some drugs are given trademarks even before clinical testing (research on humans) is complete, which means that the brand name cannot reflect the generic name. The U.S. Adopted Names Council, the Chicago-based organization that assigns generic names, (derived from the chemical compound that forms the basis of the drug) usually does not assign a name until clinical testing is well-established.

So here's Merck testing a new drug that's a sure bet for FDA approval, and although the company is prevented by law from advertising and marketing it before approval, they want to get the name Mevacor associated with cholesterol lowering, especially in physicians' minds. They need to do a little subliminal advertising, and they can't do it without a name. Because in prescription drug marketing, the name of the drug is the name of the game, and the sooner a company begins the name recognition process, the higher the sales.

So at Merck and other companies that use this approach, a bunch of people in marketing, sales and publicity sit down and string prefixes, middle syllables and suffixes together (any more than three syllables makes for too complicated a name) to see what they like. There are also computer programs such as one called Genix that will "think" up medical-sounding brand names for drugs. The goal is to come up with a name that will be instantly associated with the product -- like Exxon, Kodak, Xerox, Kleenex or Coke -- but that will have no negative connotations in any language.

Thinking up names is the easy part, says DiDomizio. It's much trickier to satisfy legal and linguistic requirements. Once the creative types find a hundred or so names that they like, they compare the list with the data base of the U.S. Patent and Trademark Office, which has about 25,000 registered names in pharmaceuticals alone. Many name candidates fall by the wayside here; the surviving names are then sent to a data base in Frankfort, Germany, which does an international search of its over 500,000 medical trademarks and terms in use in 40 countries.

James Meyer, trademark counsel for SmithKline & French, says that after his company has made certain that the name of the new drug is cleared of all legal restrictions, it is sent to marketing managers in foreign countries to make sure it doesn't mean anything unfortunate in another language. No more than a few names survive this elimination process, and a final decision is made -- usually by the company's president, and the trademark is registered in the United States and all countries in which the drug will be sold.

Now the drug company has a chemical compound, usually a white or off-white substance, with a name and some idea of what it will be used for. More decisions: Should it be a tablet or capsule? Should it be red, pink, blue or a brightly hued combination? What shape should it be? What size?

Molly O'Neill of Burson-Marsteller, a public relations firm that handles a number of pharmaceutical accounts, says that when a company markets a new drug, it looks " "at all the other drugs in that category and then goes for a different size, shape and color."

Studies have shown that white pills represent potency, yellow and orange are seen as stimulating, light blue appears soothing, green is associated with narcotics and black means death.

Most pharmaceutical companies choose a color to make the drug as attractive as possible -- to give it "pharmaceutical elegance," says Heymsfeld. But certain colors are difficult to produce, and some dyes cannot receive FDA approval.

A pill's size and shape are also important. "The smaller the better," says O'Neill. "And the more distinctive it looks, the more likely people are to identify it and to take it correctly." O'Neill says that the color of a drug affects compliance -- taking drugs the way they're prescribed, and the better it looks, the more successful it will be. This is especially true with psychotropic drugs -- mood-altering -- and those used by older people, who take almost twice as many drugs as do younger ones.

DiDomizio says that choosing a color used to be based almost totally on the whim of the product manager, but the company is much more careful now about making the decision. Merck will consider consumer and physician preference -- they'll actually take polls, and they use computer design systems.

This sounds like a lot of trouble and expense for a product that's going to disappear down your gullet, but in the highly competitive world of pharmaceutical manufacturing, where it costs about $125 million to develop a new drug and where dozens of different companies manufacture similar products, the total experience of looking at, tasting, smelling and swallowing a drug can contribute to the difference between success and failure.

Margot Joan Fromer is a free-lance writer in Silver Spring.