True or false? Cooking with wine adds extra flavor to a dish because the alcohol dissolves flavor components that are not dissolvable in water.

This statement, or statements like it, have been repeated in several places. Chefs I've spoken with accept it as quite reasonable and indeed it does seem to make sense, because many substances do dissolve in alcohol, but not in water.

Nevertheless, the statement is false. The real reason we use wine in cooking is simply that a good wine contributes its good flavor to the dish. It has nothing to do with dissolving flavor components.

Here's the catch: In a mixture of alcohol and water such as wine, the alcohol doesn't act like pure alcohol and the water doesn't act like pure water. They act like a mixture of alcohol and water, and a mixture can have quite different properties from either of the pure liquids.

If we mix equal amounts of alcohol and water, the mixture will be more than 2 1/2 times as viscous ("thick") as either the pure alcohol or the pure water. The reason is that alcohol molecules and water molecules attract and stick to each other by forming so-called hydrogen bonds. They cannot flow as freely as the unhindered molecules can in pure alcohol or pure water. The properties of the mixture, including what it can and cannot dissolve, vary as the percentage of alcohol varies. If a given substance dissolves in pure alcohol or pure water, that doesn't mean it will dissolve in any given mixture of alcohol and water.

On Solvents, Solutes and Solvation The following two paragraphs are for technically inclined readers. Scan, skim or skip them as your disposition dictates.

For a liquid such as alcohol (a solvent) to dissolve a soluble substance (a solute), the solvent's molecules must surround (or solvate) each solute molecule like a swarm of hungry piranhas and drag it out into the liquid. But if the alcohol is mixed with water, the hydrogen bonds between them hamper the alcohol molecules' ability to solvate the molecules of the solute. Thus, a mixture of alcohol in water cannot effectively dissolve what pure alcohol might be able to dissolve.

Moreover, the less alcohol there is in the water, the more its solvent abilities are weakened. For example, when you add half a cup of wine containing 12 percent alcohol to a quart of braising liquid, the alcohol concentration is reduced to 1.5 percent. The alcohol molecules are outnumbered by water molecules by nearly 200 to 1, so there aren't enough of them to solvate the solute molecules.

The Experiment Is all of this mere theory? No. I did an experiment to test it.

Annatto seeds, also known as achiote (ah-chee-OH-tay), are the seeds of the tropical evergreen shrub Bixa orellana. They are coated with a paste-like oil containing an intense yellow-orange carotenoid pigment called bixin, which dissolves in oils and in alcohol but not in water. Annatto's bixin is an FDA-approved coloring for fatty foods such as butter, margarine and processed cheeses. In this experiment, I used the highly visible bixin to simulate an alcohol-soluble flavor component in a food.

I placed five annatto seeds into each of four small test tubes and added 15 milliliters (a tablespoon) of one of the following liquids to each tube: water, a chardonnay containing 13 percent alcohol, a vodka containing 40 percent alcohol (80 proof), and 95-percent-pure ethyl alcohol. I let the tubes stand at room temperature for several days, with occasional shaking.

Here are the results: Neither the water nor the wine showed any bixin color at all; the water remained colorless and the white wine remained, well, white-wine-colored. The vodka turned mildly yellow from a small amount of dissolved bixin, while the 95-percent-pure alcohol turned intensely yellow.

Conclusion: Wine -- even straight, undiluted wine -- doesn't dissolve or "release" any alcohol-soluble bixin from the seeds. The alcohol concentration has to be high, some 40 percent or higher, to extract any appreciable amount of bixin. But such high alcohol concentrations never occur in cooking. Adding half a cup of vodka to a quart of sauce would produce a solution of only about 5 percent alcohol, even lower than the completely ineffective alcohol concentration in undiluted wine.

Now We're Cooking But that was at room temperature. What happens in the heat of cooking?

Although most substances are more soluble at higher temperatures, the facts of life regarding hydrogen bonding are still in effect. So while hot, pure alcohol will extract more alcohol-soluble components at higher temperatures, hot wine still won't.

Nevertheless, the alcohol in wine can contribute to flavor beyond the flavors inherent in the wine itself. During cooking, the alcohol can react chemically with acids in the food to form fragrant, fruity compounds called esters. You can demonstrate this by vigorously shaking some denatured alcohol with vinegar (acetic acid) in a tightly sealed bottle. After shaking for a few minutes, open the bottle carefully and sniff; in addition to the odors of the alcohol and vinegar, you will detect a fruity note of ethyl acetate, one of the esters in the aroma of pineapple.

In the cooking pot, alcohol also can react with any oxidizing substance to form aldehydes -- compounds responsible for flavors such as almond, cinnamon and vanilla. Both the esters and aldehydes are new flavors that were not present in the original ingredients. And, contrary to widespread belief, the alcohol never "boils off" completely. It has plenty of time to take part in these chemical reactions during cooking. That's another virtue of cooking with wine.

So enjoy your coq au vin and boeuf Bourguignonne. The wine will add flavor in several ways but don't expect it to "extract" or "release" any alcohol-soluble flavors from your food.

Now that I think of it, why must we extract flavor compounds from our food, anyway? If they're in there, they're in there, and we'll taste them when we chew, whether they inhabit the solids or the sauces.

Robert L. Wolke (www.professor science.com) is professor emeritus of chemistry at the University of Pittsburgh and the author, most recently, of "What Einstein Told His Cook: Kitchen Science Explained" (W.W. Norton). He can be reached at wolke@pitt.edu.