LET OTHERS SEEK THE EPITOME OF Olympic greatness in the Mercurian speed of Gwen Torrence, in the Odyssean lifts of German superheavyweight Ronny Weller, in the nymphlike freestyle strokes of China's Li Jingyi. There is no question that, from the first round of field hockey bright and early Saturday morning to the men's handball final that closes the competition, the athletes of the world will be flesh-and-blood paeans to the heights (both kinds) that the human body can achieve. But to glimpse the purest embodiment of the athletic ideal, tear yourself away from NBC's marathon broadcasts. Cozy up, instead, with a volume of sports stats. See how not a single world record established before 1980 -- in any sport -- still stands. See how winning times in the men's 100 meter, for instance, have plunged from Jackson Scholz's 10.6 seconds in 1920 to Jesse Owens's 10.3 in 1936, Jim Hines's 9.9 in 1968, Carl Lewis's 9.86 in 1991 and Leroy Burrell's 9.85 last year. Notice the long-jump records. Even Bob Beamon's 29-foot 2.5-inch wonder at the 1968 Mexico City Olympics, which many experts said would stay on the books forever, fell before the quadriceps of Mike Powell in Tokyo in 1991. And Powell actually had some air to push through. Baron Pierre de Coubertin must be smiling down from Mount Olympus: when the Frenchman revived the Games of ancient Greece, he chose as a motto ""Citius, altius, fortius'' -- ""Swifter, higher, stronger.''
OK, but . . . forever? Look more closely at the march of winning times and record distances, of gold-medal weights and precedent-setting heights. The law of diminishing returns has set in. The world-record time in the women's 400-meter freestyle, for example, dropped more than two minutes -- a full 33 percent -- from 1921 (6:16.6) to 1976 (4:11.69). In the 20 years since, it has fallen just eight seconds, to Janet Evans's 4:03.85 at the 1988 Seoul Olympics. If you were to plot world records on graph paper, you would get curves that seem to approach a limit asymptotically, coming tantalizingly closer but never quite reaching it. It is as if the curves were little south-pole magnets and the limit an imposing bar of north polarity. But what is that limit? For the men's 100, is it 9.7 seconds or 9.4? Is the limit for the mile (where the world record is now 3:44.39) 3:40 or 3:30? Will anyone ever clean and jerk more than 1,020 pounds?
The pull of gravity against the runners and jumpers and weight lifters will not lessen. Nor will the density and viscosity of water (773 and 55 times greater than air's, respectively) impeding the swimmers. Only two variables can make mortals swifter, higher, stronger. One is equipment. In the pole vault, for instance, the world record edged up just two inches between 1942 and 1960 and seemed stuck around 16 feet. But when rigid steel poles gave way to springy aluminum, fiberglass and graphite composites in 1963, the record catapulted two feet in three years and is now 20 feet 1.75 inches. Well into the 1950s, tracks were made of cinders; the cinders became sticky and slowed the runners' long-spiked shoes. The new synthetic tracks make a difference of about one second per quarter mile. Shorter studs cut another couple of seconds from the mile. If these innovations had been available earlier, Glenn Cunningham, who set a world record in the mile of 4:06.8 in 1934, might have beaten Britain's Roger Bannister past the four-minute barrier by 20 years. The latest technology promising to improve performance by a quantum leap is a new ultrathin bicycle with a solid, not spoked, rear wheel (page 30); it could shave several seconds off times.
The other variable is the human body. With the realization that athletic achievement is bumping against a ceiling, sports scientists are mapping the body's outer limits, trying to understand where improvement is still possible and what holds it back. Much of the work is preliminary. Physiologists have identified several limit- ing factors, but have yet to throw them all into a computer and come up with a numerical prediction that, for instance, no human will ever long-jump 30 feet. What the sports scientists have done is pinpoint the systems that hold back athletic achieve- ment and focus all of their training tricks and biomechanics analy- sis on those from which they might still coax significant gains.
""The day that a voluntary contraction of a muscle equals the maximum evoked contraction will be the day that we have reached our physical potential.'' So says Jaci VanHeest,director of exercise physiology for swimming at the U.S. Olympic Training Center in Colorado Springs, Colo. What she means is that an electric shock applied to a muscle (by the humane, lab version of a cattle prod, basically) evokes a twitch. The forcee of that twitch is the ""maximum evoked potential,'' the greatest force that muscle can exert. If you add up the maximum force of every muscle involved in a particular athletic move -- all the leg, back, shoulder and arm muscles that lift a weight, for instance -- you would get a rough idea of the total force available. At least in theory. In reality, humans do not get all this force out of their muscles. Doing so could break a bone. Still, this analysis provides an upper ceiling on strength and speed.
More realistic limits on strength are determined at the cellular level. Skeletal muscle is composed of fibers whose contraction speed ranges from fast to slow. Fast fibers derive the energy required to contract from biochemical reactions that occur in the absence of oxygen. Such an ""anaerobic'' reaction can run for only a few minutes, so it powers sprints, on both water and land. Sprinters tend to have a higher percentage of fast, anaerobic muscle fibers than do other athletes, which suggests that the body's maximum percentage of these fibers (which no one knows) limits sprint speeds. As Sir Roger Bannister, who became a neurologist after hanging up his running shoes, says, ""The 100-meter dash is just an extravagant fling of energy.'' And one that does not rely on aerobic (lung) capacity: since fast fibers do not require oxygen, some sprinters run the entire race without drawing a single breath.
In longer events, powered by slow fibers dependent on oxygen, the key is ""getting oxygen to the muscles,'' says physiologist Roger Woledge of University College, London. Oxygen is carried by blood. The amount of blood pumped out of the heart with each beat can be increased with training -- that's what aerobic fitness is -- but only so far. You can't pump out any more blood than the heart's chambers can hold; barring evolutionary changes that increase the size of human hearts, that volume is pretty much fixed. The heart muscle itself could pump faster, but the rest period between beats has to be long enough for the chamber to refill. (In tachycardia, heart rate soars but the volume of ejected blood drops.) The heart-rate maximum is roughly 220 beats per minute minus age.
Muscles' lack of efficiency also limits performance. Muscles contract when tiny levers on myosin, a muscle protein, fit into grooves on actin, another protein, and push it forward exactly like a ratchet wrench. But myosin can latch onto actin in any of several positions, not all of them ideal. ""Only when the myosin heads are in the right register can the muscle have the optimal tension,'' says VanHeest. But optimizing every actin-myosin pairing is less an achievable goal than a Platonic ideal.
Muscles move only when so ordered -- by a motor neuron. The thickness of the myelin sheath on motor neurons determines how quickly electrical impulses travel along them. Although faster impulses probably do not make a muscle contract with greater force or speed, they might help synchronize motor neurons. ""Why does a sprinter run a 10.2-second 100 one day and a 10.0 the next?'' asks physiologist Reggie Edgerton of the University of California, Los Angeles. Leaving aside factors such as track conditions and state of mind, it's because of how well, or poorly, the motor neurons are synchronized, he argues. ""The ideal situation would be if you could activate all of the neurons with perfect timing,'' so they moved the muscles with the efficiency of an engine firing on, oh, a few thousand cylinders. ""But that's impossible,'' says Edger- ton. It's another limit -- but one that sports psychologists think they can teach athletes to at least partially overcome (page 35).
The only way to surmount these limits in a big way is through evolution. In a few thousand generations, humans might develop larger hands, the better to pull themselves through water, or the oversize legs and flexible back of the cheetah (whose 400-meter time of 16 seconds makes the human record of 43.29 seconds look like a tortoise's). In fact, a little of that goes on already. Natural selection for the body type best suited to a particular environment has already, over 100,000 years or so, sculpted human physiques as different as gymnast Dominique Moceanu's and wrestler Bruce Baumgartner's. Then another kind of selection sets in. To become a world-class star, athletes need will and the right body type. Yes, training can alter the body's shape -- Russian swimmer Aleksandr Popov wasn't born with those shoulders -- but only within the parameters that nature sets. Female gymnasts are small because it reduces the force needed for flips. Soccer players are of average height (being too tall makes it hard to execute fancy moves like the somersaulting bicycle kick) and have huge quadriceps and hamstrings, providing the power for kicks and flips. Sprinters and high jumpers like Jackie Joyner-Kersee have almost no subcutaneous fat, since in these sports ""the key is the power-to-weight ratio,'' says Bannister. Fat provides weight but no power.
As athletes brush the upper limits of human performance, trainers have to throw the entire armamentarium of physiology, aerodynamics and biomechanics into the fray. The U.S. Olympic Training Center is, therefore, a monument to sports science. Working on the assumption that information is power, the weight lifters teeter atop platforms rigged with sensors that measure the power generated by each leg. Cyclists are hooked up to machines that measure precisely how much force they apply at each point in the pedal stroke. The swimmers get towed through the pool so they can feel where on their body the drag is greatest (it's like sticking a hand out the window of a speeding car and feeling the wind resistance). Then they can modify their angle of attack to reduce drag. For sheer complexity, though, no training gadget exceeds The Flume. It's basically a water treadmill, pumping 60,000 gallons of water through a pool with a transparent bottom and sides, and sits in a hyperbaric chamber so swimmers can train at any ""altitude.'' Cameras record every stroke and then feed the data to a computer that calculates fluid force equations. The equations incorporate the swimmer's every move. The result is a stick figure of the swimmer with vectors representing lift and drag.
Out of all this has come the hope that, though any world records set in Atlanta will be very hard won, they are not out of the question. In swimming, for instance, body shaving has already lopped a full second off the fastest times in the men's 100-meter freestyle. In Atlanta the U.S. team will don swimsuits woven of synthetics that promise a lower drag even than smooth skin. The suits also have tiny striations placed to reduce the turbulence produced during forward movement, much as winglets on an airplane turn air turbulence into laminar (smooth) flow. Even the biomechanics of the freestyle stroke, it turns out, have room for improvement. Bob Prichard, director of the Somax sports clinic in Corte Madera, Calif., has pioneered the theory that hydropower comes not from arm and shoulder strength but from hip dynamics. ""The rotation of the hips accelerates the hand through the water,'' says Prichard. ""Most swimmers pull with their arm first and then at the last second rotate their hip to get it out of the way. But if you rotate your hips as soon as your hand enters the water, you can decrease your strokes and improve your time.'' When Amy Van Dyken adopted the technique for the Olympic trials in March, she won the 50- and 100-meter freestyle races. ""I concentrated on rotating my hips,'' she said.
Britain's Jonathan Edwards has a shot at a goal as mythical as the four-minute mile once was: a 19-meter triple jump. Last year Edwards became the first human to triple-jump 18 meters, shattering the world record by .32 meter. As he leaped, sports scientists shot high-speed video footage and fed it into a computer that analyzed Edwards's winning technique. His first advantage was speed, found Erik Simonsen of the University of Copenhagen and colleagues. In his initial run-up, Edwards reached 11.9 meters per second (Carl Lewis, in setting a world record in the 100 in 1991, topped out at 11.8). Landing after the first jump, other athletes place their leading foot ahead of their center of gravity to avoid falling face first into the sand. But that foot placement acts as a brake. Edwards lands with his foot directly under his center of gravity -- and, somehow, remains upright. That lets him launch into the final jump faster than anyone else. By raising his takeoff angle slightly, calculates Simonsen, Edwards could clear 18.9 meters. With a legal tailwind, he has a shot at 19.
Athletes who believe in ever-higher peaks of athletic achievement could learn a lesson from racehorses. Thoroughbreds ""do not go any faster today than they did 100 years ago,'' says Woledge. (The record speed for a racehorse is 43.26 miles per hour, set by Big Racket in 1945.) ""I interpret that as meaning there is a physiological speed limit.'' Something will be lost from the Olympics when there are no more world records to set. But even when the ghosts of athletes past no longer haunt the competition, the men and women of the Games will still reach deep inside themselves, using all the tools of science to become, like every athlete since Olympia, ever swifter, higher, stronger.