Two Americans and a Frenchman won the Nobel Prize in chemistry yesterday for figuring out how to make relatively simple molecules that mimic the functions of far more complex molecules produced by living cells.

Practical applications of the work -- a form of molecular engineering in which molecules are designed with a rigid framework holding a flexible region that can do a specific job -- are few to date but experts say the future could see a wide range of uses in medicine and energy production.

The award is to be shared by Charles J. Pedersen, 83, who pioneered the field in the 1960s before his retirement from the Du Pont Co. in Wilmington, Del., and two others who pushed Pedersen's work in new directions. They are Donald J. Cram, 68, of the University of California at Los Angeles, and Jean-Marie Lehn, 48, who has joint appointments at the College de France in Paris and the Universite Louis Pasteur in Strasbourg.

Pedersen, a soft-spoken man who has lived in quiet retirement in Salem, N.J., for 20 years, said his award came as a great surprise. He learned of his selection yesterday morning in a telephone call from the Royal Swedish Academy of Sciences, which administers the awards established by Alfred Nobel, the inventor of dynamite. The phone did not stop ringing all day, Pedersen said.

"I'm not used to this sort of thing. I'm going to get off and take care of some personal matters," Pedersen told a reporter. "I've never had an experience like this. It is a great honor."

Cram, by contrast, said he had been dreaming of winning the Nobel for many years. He said he did not seriously think he was a contender until he turned 55. "Then I thought, 'Wouldn't it be fun to get the Nobel Prize?' " That was 13 years ago.

The research that earned the prize began in the 1960s with Pedersen's work on a previously known class of organic molecules called crown ethers. An ether is a particular link-up of carbon, hydrogen and oxygen atoms. Pedersen discovered that certain ethers made as circles with a repeating oxygen-carbon-carbon sequence, which forms a zig-zag ring reminiscent of a crown, had the ability to pull individual metal atoms into the center of the ring and hold them.

Pedersen recognized that by maintaining a rigid framework but changing the atoms facing the center of the crown, he could make molecules with specific affinities for taking in different metal atoms.

Lehn and Cram extended the concept to make somewhat different molecules that, instead of a ring, had a cup. Where Pedersen's molecules were compared in shape to crowns or doughnuts, Lehn's and Cram's were thought of as bowls, baskets, cradles and even vases on pedestals. Such molecules are called cryptands. By changing the shape of the crypt and the combination of atoms lining it, chemists can design cryptands that recognize and take on only one specific kind of molecule.

Some call the field host-guest chemistry, the host being the recipient molecule and the guest the one taken in. The importance of the host-guest interaction is its specificity.

The field excites chemists because host-guest interactions are much like those of protein molecules in living cells. Both kinds of interactions are based on a three-dimensional fit something like that of lock and key. In cells, these interactions include proteins that fit onto specific genes to control their operation, enzymes that act as catalysts, binding other molecules to speed up chemical reactions, and even antibodies that the immune system makes with specific shapes to bind to foreign proteins.

Cram said it was impossible to predict how this field might someday affect medical research or treatment.

"There's still 40, 50, 100 years of research needed to find out all we don't know," Cram said. "We're really just opening the door."

Earlier applications, others say, could come in making fuel cells more efficient. These are devices that convert chemical energy, such as that in petroleum, directly into electrical energy. Membranes bearing crown ethers or cryptands could catalyze the chemical reactions that result in electricity.