lice Chandler breeds racehorses the way her father did, the way generations of Kentucky hardboots have bred the sleek, swift-footed competitors in the sport of kings.

Dazzling thoroughbreds have entered the world under her watchful eyes. Sir Ivor, the winner of the 1968 Epsom Derby, was bred at Mill Ridge Farms, Mrs. Chandler's 1,050- acre property in Lexington. So was Keeper Hill, a bay filly who in 1998 won the Kentucky Oaks.

Point Given, the favorite at Churchill Downs last Saturday for the 127th running of the Kentucky Derby, was raised at Mill Ridge. (The colt finished fifth.)

In matching stallion to mare, Mrs. Chandler follows the oft-quoted maxim, "Breed the best to the best and hope for the best." She scours pedigrees and racing records — the detailed accountings of lineage and success in the sport that stretch back 300 years and 30 generations, to the founding sires of the breed. She assesses conformation and ability with experienced eyes. She hopes for speed, but also for soundness, endurance, courage and "heart," that ineffable quality that makes a horse want to surge forward, leaving its competition behind.

But what if, in arranging a promising union, Mrs. Chandler and other breeders could harness not only the traditional science of the breeding shed but also the contemporary science of the molecular genetics laboratory? What if they could detect vulnerability to the illnesses or injuries that often disable racehorses? Or if a simple DNA test, performed on a strand of hair plucked from a horse's mane, could help breeders produce budding Secretariats and Cigars, by pinpointing genetic markers for exceptional performance?

A few years ago, no one even dreamed of such ideas. But they are now possibilities held out by the Horse Genome Project, the equine version of the human genome effort.

The genome project, begun six years ago, was started by a group of geneticists concerned by the lagging state of genetic research on horses. Hampered by limited funds, the field fell far behind work on other species.

In 1990, when the Department of Agriculture invited veterinary scientists to begin mapping the genomes of domestic animals, equine researchers were forced to decline.

"The cattle, the sheep, the pigs and the chickens all said yes," recalled Dr. Ernest Bailey, a geneticist at the University of Kentucky's Gluck Equine Research Center and the coordinator of the genome effort, "but we didn't have the resources, the researchers or the funding to do that type of project."

But in 1995, Dr. Bailey and other scientists decided the solution was to collaborate. They established two "reference" families of horses that all could study and agreed to share DNA samples and pool their laboratory resources.

The horse project, which now involves 25 labs in 15 countries, has a more modest goal than the human effort. Instead of spelling out the nucleotides that make up the DNA on all human chromosomes, the horse researchers are mapping, identifying genes and markers — short segments of DNA that serve as signposts — on the chromosomes of the horse. While humans have 23 pairs of chromosomes, horses have 32.

"We have a broad outline of where the horse genes sit on the horse chromosomes, and people are beginning to fill in the gaps," said Dr. Doug Antczak, an equine geneticist and director of the James A. Baker Institute for Animal Health at the Cornell College of Veterinary Medicine. Dr. Antczak is one of the organizers of the genome project, which receives financing in part from the Agriculture Department and the Dorothy Russell Havemeyer Foundation.

But the researchers are aided in their work by a twist of evolution: Large chunks of the human genetic code appear with only minor changes on the horse genome, allowing the researchers to make generous use of the human blueprint in locating equine genes. All of human chromosome 8, for example, is contained on horse chromosome 9.

The knowledge is already having an effect. In 1997, scientists at the University of Texas identified the gene mutation responsible for severe combined immunodeficiency syndrome or SCID, a fatal disease that affects Arabian foals. The location of the gene for another equine killer, lethal white disease, was reported independently by three teams of researchers in 1998. Both illnesses can now be detected with a DNA test.

And drawing upon the project, investigators are studying the genetic underpinnings of everything from coat color and muscle physiology to developmental bone diseases and "tying up," the severe muscle cramping racehorses can develop after a "breeze," or full-out gallop.

Geneticists say it is only a matter of time before progress in the laboratory translates into new treatments and predictive tools that can be used by breeders, including those who have their eyes on the Triple Crown.

Almost certainly, thoroughbred breeders will gain ways to screen stallions and mares for the genetic determinants of weakness in bone and muscle, allergies and other difficulties that can keep a horse off the track. A genetic basis may also eventually be found for problems like exercise-induced pulmonary hemorrhage, known in the racing world simply as "bleeding," a condition that affects up to 80 percent of racehorses.

"We will be able to give breeders more precise information and make them more effective in what they've been doing all along," Dr. Bailey said.

For Dr. Bailey and other researchers, the goal is to breed sounder and healthier horses.

No one in the racing industry would disagree.

"The more we know, the more we can do what's best for the horse," said Nick Nicholson, president and chief executive of Keeneland, the racetrack and horse sales company.

But what is revealed by genetic science may not always be so welcome. In 1992, for example, researchers discovered a gene mutation responsible for a peculiar muscle ailment in pedigreed quarter horses — hyperkalemic periodic paralysis or HYPP. Horses that had the genetic defect were strikingly beautiful, and many became champions. But they were also subject to muscle spasms so serious that they could topple over and die.

The scientists traced the gene back to one stallion, Impressive, whose popularity as a stud lay behind the spread of the condition. And quarter horse breeders were forced to decide which was more important: continuing the champion line or breeding horses free of the defect.

The American Quarter Horse Association now requires testing for HYPP in horses bred from the Impressive line, if parentage verification is required for registration. But it is up to the individual breeder to decide whether to breed the horse or not.

That thoroughbred breeders may at some point confront a similar dilemma is not inconceivable.

Still, what concerns breeders like Mrs. Chandler, chairwoman of the board for the University of Kentucky Equine Research Foundation, is something potentially even more controversial: whether the advances in gene research will yield the ability to tell if a horse has what it takes to win. Such knowledge, Mrs. Chandler and other breeders fear, could be exploited by wealthy owners in a business where the stud fees of elite stallions run into six figures and untrained, untested yearlings sell for $6 million at a Keeneland sale.

"I'm worried that it's going to be a way for somebody to buy the game," Mrs. Chandler said.

At least a few racing investors are betting that the new technology will pay off.

Dr. Herman Raadsma, for example, director of the Center for Advanced Technologies in Animal Genetics and Reproduction at the University of Sydney, said he had already been approached by a breeder interested in finding genetic markers of exceptional performance in certain broodmare lines.

Dr. Raadsma would not describe the plan in detail, saying that the breeder, who is working with a group of owners, asked him to keep silent.

But he noted, "If you have an information edge, it only takes a nose- length difference to be a winner."

Still, most scientists agree that, for a variety of reasons, such research is unlikely to give owners with unlimited resources an unstoppable advantage.

Heredity clearly plays a role in success on the track, though the extent that genes contribute to racing ability has been difficult for researchers to tease out, in part because the racing industry itself has no universally agreed on measure of performance.

Earnings, racing times and number of victories are used variously to describe a thoroughbred's prowess. Complicating matters further, races differ in length. Track conditions change daily. And only the winning time in a race is recorded, making comparison difficult.

But using indices including racing times, handicapping (the adding of weight based on past performance to equalize the field) and earnings on the track, researchers have come up with various estimates of how much of the variation in racing performance can be attributed to genes, from a low of 9 percent to a high of 49 percent.

Heredity, researchers have found, appears to exert a stronger influence in determining performance in sprints, like those run by standardbred trotters, than over the longer contests typical of thoroughbred races.

The genetic legacy passed on by sire and dam, however, does not translate into a single trait, but a complex mix of abilities and characteristics — speed, stamina, drive, efficiency of movement, the composition of muscle, heart, lung and limb.

And there is no one biological recipe for success. Champion thoroughbreds come in all shapes and sizes. They are tall and short, stocky and slim. Some are exemplars of textbook equine conformation; others exhibit flaws that would make a veterinarian shudder. Secretariat appeared as a resplendent, muscle- bound giant. Seabiscuit, the legendary racehorse of the 1930's and the protagonist of a recent best-seller by Laura Hillenbrand, looked more like a cow pony.

Few if any of the traits that combine to produce exceptional performance, scientists say, are likely to be governed by one powerful gene acting alone. Instead, they are probably influenced by many genes of smaller effect acting together.

Nor are genes the whole story. Training, nutrition, expert veterinary care and a variety of other environmental influences go into transforming raw talent into victory. Thus, the chances that scientists will ever find a "speed" gene, or that individual owners will be able to use genetic screens to "buy the game," as Mrs. Chandler put it, are remote.

"It's possible that new genetic technology may enable us to more precisely identify horses with exceptionally high or low potential," said Dr. Patrick Cunningham, a professor of animal genetics at Trinity College in Dublin. "But to say that it will provide a magic bullet that gives one horse an advantage over another in the short term is unlikely."

Still, Dr. Cunningham and other researchers said there was no question that some genes would be found that do have an effect on performance.

One candidate, called the ACE gene, is under study in Dr. Raadsma's laboratory in Australia. The gene, involved in blood pressure regulation, has been associated in humans with physical stamina. Studies have found that elite mountaineers, runners and rowers are more likely to have one variant of the gene, a form that occurs less frequently in the general population.

Dr. Raadsma and his colleagues, Dr. Imke Tammen and Tash Ellis, are hoping to find a similar gene variant in horses, a marker that might be used to identify exceptional talent. Making use of the human genome, they have located the ACE gene in the horse and have sequenced part of it, but they have yet to find a second form of the gene or to link it to performance.

Yet as tantalizing as such efforts appear, they are in the end likely to provide only one more potato for the pot, an additional source of information not unlike the ultrasounds and treadmill tests already used by some owners and breeders to evaluate horses' heart function or gait. And genetic techniques, scientists say, will never supplant the breeder's experienced eye or the expertise of the bloodstock agent.

"You'd have to give it consideration," said Dermot Ryan, the general manager of Ashford Stud in Lexington, where last year's Derby winner, Fusaichi Pegasus, is booked this season for closely chaperoned trysts with 140 mares, whose owners are paying $150,000 each. "But it will be a long time before science will take over this game."