Chapter 14

Was Einstein Right about Relativity?

In This Chapter

 Looking at early tests of relativity

 Testing general relativity with spacecraft

 Understanding the GPS connection to relativity

 Using NASA missions to check Einstein’s theories

 Realizing that theories can never be proved absolutely

S cientists have tested Einstein’s predictions from Mars, from the moon, and from orbit. NASA’s latest effort, Gravity Probe B, was launched in April of 2004 after 35 years of design and testing. The mission will measure, very precisely, how space and time are warped by the presence of the Earth and, more profoundly, how the Earth’s rotation drags spacetime around with it. These effects, though small for the Earth, have far-reaching implications for the nature of matter and the structure of the universe.

Gravity Probe B is among the most thoroughly researched programs ever undertaken by NASA, but it certainly isn’t the first attempt to test Einstein’s theory of relativity. In this chapter, I summarize other tests that have been conducted in the past century. I also show you how thousands of people test the theory of relativity every day by using GPS locating devices. Finally, I show you why we can’t be 100 percent certain that Einstein was right.

Conducting Early Tests of Relativity

Einstein’s scientific work is at the heart of today’s physics. For example,

 His special theory of relativity, which I discuss in Part III of this book, showed that Isaac Newton’s laws of motion (see Chapter 4) are only an approximation.

 His general theory of relativity, which I detail in Chapter 12, made radical changes to our notions of gravity and space and also showed that Newton’s mechanics is an approximation. According to Newton, gravity is a force that makes objects feel each other’s presence. But Einstein said that gravity is geometry, the result of the distortions of spacetime.

 His introduction of the light quantum idea, which I discuss in Part V, led to a whole new area of physics that didn’t exist before.

But how do we perceive these theories in the everyday world? Can we see what Einstein was talking about?

The discrepancy between special relativity and Newton’s mechanics, for example, is imperceptible in the motions that we experience in our everyday lives. Even for astronauts flying on the Space Shuttle, the discrepancy can’t be measured. The speed of the Shuttle in orbit is about 28,000 kilometers per hour (kph), which is 17,500 mph or 23 times the speed of sound (Mach 23). While the Shuttle is traveling a whole lot faster than your car can ever go, it’s only moving at about three-hundredths of a percent of the speed of light. At such slow speeds, the relativistic effects Einstein described aren’t noticed; space and time are not seen as being connected into spacetime. The speed measured on the Shuttle matches the speed that tracking stations measure on the ground. The distances that the Shuttle travels don’t change for the astronauts or for the technicians on the ground.

The subtle changes that general relativity brought to Newton’s theory of gravity also seem far removed from our everyday lives. The warping of spacetime can’t be measured without extremely precise instruments looking at distant galaxies and quasars.

Yet both theories, along with quantum physics, are closer to our lives than you may think. Digital cameras, cellular phones, computers, and GPS units all use one or several of Einstein’s theories.

In this sense, Einstein’s theories have been tested. The proof is in the pudding. However, scientists continue to perform sophisticated tests of the special and general theories of relativity. Do a Web search for “tests of relativity,” and you’ll find over 1,000 entries, many of them current research papers explaining delicate tests of the theories.

How do you test a theory? You test its predictions.

The special theory of relativity predicts that

 Clocks run more slowly in the moving frame.

 Lengths are shortened in the moving frame.

 Mass increases in the moving frame.

The general theory of relativity predicts that

 Gravity slows time. Time runs more slowly in the basement (where the gravitational field is stronger) than it does on the top floor of a structure.

 Signals are Doppler-shifted in gravitational fields.

 Spacetime is curved.

Testing special relativity: Life extension

One of the early tests of relativity involved the muon, a short-lived elementary particle that has its lifetime extended when traveling close to the speed of light. (I describe the effect in Chapter 10.) The muon travels at 0.998c in the atmosphere, and at that speed, its lifetime is extended 16 times. When scientists measure muons’ lifetimes in laboratories (where the muons are at rest), the lifetimes are extremely short — much too short for muons to travel from the Earth’s upper atmosphere, where they are born, to the ground. Yet they are detected on the ground because their lives are extended when they travel at such high speeds.

In 1976, physicists at the European Particle Physics Laboratory (CERN) in Geneva, Switzerland, accelerated muons that had been generated in the lab to 0.9994c in a particle accelerator (see Figure 14-1). The muon lifetimes increased by a factor of 30. At rest in the lab, the particles have a lifetime of 2.2 microseconds. Taking a leisurely trip around a 14-meter (46-foot) diameter track, a typical muon would complete 14 circles before its life was over. But at 99.94 percent of the speed of light, the muon would manage more than 400 turns around the track before it died. These measurements agree with the prediction of special relativity to within 2 parts in 1,000.

Since this experiment was conducted, similar experiments have increased the accuracy even more. Today, measuring the relativistic increase in the lifetime of muons is a lab exercise for graduate students at CERN and at other high energy laboratories around the world.

Figure 14-1: Muon storage ring at CERN where the lifetimes of relativistic muons were measured in 1976.

CERN Photo

Getting younger by flying east: Relativistic time

In 1971, J.C. Hafele of Washington University in St. Louis and Richard Keating of the U.S. Naval Observatory took two atomic clocks on commercial flights around the world to test the time dilation effect predicted by special relativity. One of the flights was eastbound and the other westbound. The clocks were compared with the reference atomic clock at the U.S. Naval Observatory in Washington, D.C.

Hafele and Keating did their experiment on a shoestring budget and bought tickets on commercial flights rather than chartering planes. Their clocks were connected to the airplanes’ power and placed against the front walls of the coach class cabins. (As a government employee, Keating wasn’t allowed to fly first class!)

According to special relativity, the clock on the eastbound flight should tick more slowly, because it is moving relative to the clock in the Naval Observatory. (As I explain in a moment, the effect on the westbound clock is different, because the plane is moving opposite the Earth’s rotation.) But general relativity says that both flying clocks should run faster, because gravity is slightly weaker in the air than on the ground. The two effects tend to offset each other, and the final result depends on how fast the planes fly and on the altitude.

For Hafele and Keating’s experiment, special relativity predicted that the flying clock on the eastbound plane should run 184 nanoseconds behind the ground clock. And general relativity predicted that both flying clocks should run 144 nanoseconds faster. The net effect predicted by relativity was that the flying clock on the eastbound plane would lose 40 nanoseconds on the trip around the world. The actual observed loss was 59 nanoseconds.

Relativity gives you a way to stay young: Fly fast and low. But you have to fly eastbound always. If you fly westbound, you’ll get older because you’ll be flying opposite to the Earth’s rotation. Looking from space, a clock on the ground is actually moving faster than your westbound plane (see Figure 14-2). The result is that your clock will run faster when flying westbound than a clock on the ground, and you’ll age faster — but only by a few nanoseconds. (You don’t have to worry about getting any more gray hairs during your next flight west than you’d get on the ground.)

Figure 14-2: When you travel westbound, a clock on the ground moves faster than your plane as seen from space.

Hafele and Keating’s results with their flying atomic clocks agreed very well with the time predictions of both the special and the general theories of relativity. Other more recent and more sophisticated experiments have also been in agreement with the time-dilation prediction of special relativity and with the time-lengthening prediction of general relativity.

Probing gravity: NASA’s first test of relativity

Hafele and Keating’s experiment was a nice first attempt at a direct test of relativity. But flying atomic clocks on commercial planes isn’t an ideal way to conduct a precision experiment. A better method is to fly the clocks on a satellite.

That’s what NASA proposed to do in 1970 with Robert Vessot and Martin Levine of Harvard as principal investigators. The idea was to put an atomic clock in orbit with a Titan rocket. But budget constraints forced NASA to switch to a suborbital flight on a Scout D rocket — the Gravity Probe A mission that flew in 1976.

In June of 1976, Vessot went to NASA’s Wallops Island rocket launch facility in Virginia to man the rocket clock while Levine traveled to the Kennedy Space Center in Florida to take charge of the ground clock. At 6:41 a.m. on June 18, the Gravity Probe Amission was underway. It took an atomic clock to an altitude of 10,000 km (6,200 mi) for a two-hour suborbital flight.

The atomic clocks transmitted their signals continuously. The electronic instrumentation hooked up to the atomic clock aboard the rocket was designed to compensate for the Doppler shifts of the rocket signal during ascent and descent. (This Doppler shift is the stretching of the wavelength of the signal emitted by the rocket as it moves away from the detector on the ground, and the subsequent shortening when the rocket approaches the ground during descent. I explain this effect in Chapter 12.)

With the standard Doppler shift out of the way, things would be simpler; the signal received from the rocket clock should reflect only changes due to relativity. With the rocket in motion, the clock on board should run more slowly than the one on the ground. And as the rocket gained altitude, the weaker gravitational field should speed up the clock on board.

Initially, the high speed of the rocket caused the clock to run slow. The rocket hadn’t yet gained too much altitude, so gravity hadn’t changed enough to speed up the clock. Therefore, the rocket clock was initially ticking at a slower rate than the one on the ground. Some three minutes later, when the rocket had slowed down some and had also gained elevation, the two clocks were ticking in step. Later, the rocket clock sped up because of the increased altitude (weaker gravity) and slower speed. Figure 14-3 shows the various stages of the rocket’s movement; the dots in the figure represent the clocks’ ticking rates.

Figure 14-3: When the rocket speed is high, the clock on the rocket slows down. When the altitude is high, the clock speeds up.

The Scout D rocket reached maximum altitude at 7:40 a.m. On the way down, as the speed of the rocket increased and its altitude decreased, the two effects reversed, canceling each other out sometime around 8:31 a.m. After that, the special relativity time dilation took over, and the rocket clock slowed down.

The rocket splashed into the Atlantic Ocean, about 900 miles east of Bermuda. It took the science team two years to analyze all the data. The final result was that the special relativity time-dilation and the general relativity time-contraction effects agreed with the theory to a precision of 70 parts in 1 million.

With this delicate experiment, Vessot and Levine’s team showed with great precision that time does slow down in the moving frame and in strong gravitational fields. Time really runs more slowly when you are moving and when you go down to the basement.

Confirming Gravity’s Effects on Light

The special theory of relativity is based on the constancy of the speed of light. Many tests have shown that Einstein’s assumption is correct: You always measure light traveling at the same speed c regardless of how fast you are moving. Light travels at exactly 299,792 kilometers per second (kps) in a vacuum. In air, light slows down a very tiny amount. When it goes through glass, it slows down just a bit more. When it comes out, it resumes its slightly faster speed. This change in the speed of light doesn’t contradict relativity. The speed of light in air is the same, regardless of how fast you’re moving through the air when you measure the speed.

It turns out that gravitational fields, in addition to bending light, also appear to slow it down. But you need a very strong gravitational field if you’re thinking of measuring this effect, which is the combined result of the strong gravity warping space and the gravitational time delay. Einstein actually came up with a gravitational model with this light time-delay effect built in. However, he didn’t pursue the model fully.

Calculating the sun’s impact

Light coming to Earth from a distant object travels along the distorted space and appears to slow down due to the curvature of space around the sun. The closer the path is to the sun, the more it appears to slow down. For example, a light signal coming from Mars takes longer when it grazes the sun because of the dip in space (see Figure 14-4). Away from the sun, the curvature is not as pronounced, and the path is more like in the flat Newtonian space.

Figure 14-4: The strong gravity around the sun warps space.

In the 1960s, MIT physicist Irwin Shapiro did some calculations to see how much longer it would take a signal from a distant object to arrive at the Earth if its path grazed the sun. As in Einstein’s earlier model, Shapiro’s calculations showed that the signal would be delayed due to the sun’s gravity. In the case of a radar signal bouncing off Venus, Shapiro calculated that the signal would take an additional 10 microseconds to arrive when Venus was on the other side of the sun.

At about the same time, Duane Muhleman and Paul Reichley at NASA’s Jet Propulsion Laboratory (JPL) came up with a similar calculation while studying the effects of general relativity on a radar signal bouncing off Venus. Like Shapiro’s calculation, their results showed that the signal would take an additional 10 microseconds to travel from Venus, pass by the sun, and arrive at the detectors on Earth.

Getting radar echoes from the planets wasn’t easy in the 1960s. But the Lincoln Laboratories at MIT were up to the task. In 1967, Shapiro and his group verified the time delays for the first time. They measured several hundred radar signals bouncing off Venus and performed the detailed calculations. The results came within 20 percent of the predicted amount that light is delayed due to the general relativity effects.

Testing the time delay from Mars

Also during the 1960s, NASA’s JPL was busy exploring Mars. The Mariner 4 spacecraft arrived at Mars in 1965 and changed our view of the planet. It was the first spacecraft to make it to Mars safely, and it took the first close-up photographs of the surface: 21 astonishing pictures of another planet. In the wake of its success, NASA sent two more spacecraft on a flyby mission to Mars, Mariner 6 and Mariner 7. The two spacecraft sent back 58 close-up photographs of the Martian surface and important data on the composition of the southern polar cap.

After the flybys, the two spacecraft were locked in an orbit around the sun that was similar to the orbit of Mars. Muhleman and Reichley at JPL knew that the Mariners would be close to Mars later on when the planet was going to be located on the other side of the sun from Earth. Although their missions would be over by then, the two scientists asked NASA Headquarters to extend the missions long enough to time the signals and do the general relativity test. NASA approved. The results confirmed the predictions to within 5 percent.

NASA planned a large mission to Mars for the following decade. The Viking landers would be the first robotic craft to land on the surface of another planet. Their mission was to look for life or signs of it. Testing relativity wasn’t in the plans, but Muhleman and Reichley, teaming up with Shapiro, convinced NASA Headquarters to approve relativity measurements.

The twin Viking missions of 1976 and 1977 were extremely successful, electrifying not just the scientists but the general public with beautifully detailed color photographs of the surface and atmosphere of Mars. The missions were also a success for Einstein’s general theory of relativity. Although Einstein himself never proposed this particular test of the apparent time delay of light by gravity, it proved to be the most accurate test of relativity ever. According to the Viking landers’ measurements, general relativity’s time-delay effect is correct to within 1 part in 1,000.

Making GPS Accurate

A test of the time-delay effects of both special relativity and general relativity is done daily by millions of people around the world — the users of Global Positioning Satellite (GPS) receivers.

The GPS system, which started in 1978 with the launch of the U.S. Department of Defense Navstar satellite, is an array of 24 satellites that orbit the Earth at 14,000 kph (8,600 mph) at an altitude of 20,000 km (12,000 mi). Each satellite carries an atomic clock so accurate that it will lose or gain only 1 second in 3 million years.

The receiver has a less accurate quartz clock, like the one in your digital watch. Without a correction, receivers will get out of sync with the satellites. To synchronize the clocks, at least four satellites are always in line of sight of any receiver on the ground. The signals from three satellites provide enough information to triangulate the position, while the fourth satellite measurement provides a correction factor.

With a moderately priced GPS receiver that you can buy at the mall, you can locate your position to within a couple of meters and get the local time to 50 billionths of a second. The GPS receiver in a car can give you position, speed, and direction in a few seconds.

To get this kind of precision, the receivers must keep pace with the satellite clocks to within 20 to 30 nanoseconds. But even with the correction factor of the fourth satellite, the clocks will get out of sync and lose this accuracy because of relativistic effects:

 At 14,000 kph, the satellite clocks circle the Earth twice a day; obviously, they move much faster than the clocks on the ground. According to Einstein’s special theory of relativity, a clock on a satellite moving relative to one on the ground will run more slowly, by 7 microseconds every day (see the Speed Clock in Figure 14-5).

 But at the 20,000 km altitude, the gravitational field is weaker than it is on the ground. Therefore, the general theory of relativity says that the satellite clock should tick faster than the clock on the ground by about 45 microseconds per day (see the Gravity Clock in Figure 14-5).

If you combine the effects, the net result is that clocks in the receivers will run 38 microseconds slow in just one day, as shown on the clock on the right side of Figure 14-5. At this rate, the clocks will lose the required 30-nanosecond accuracy (which ensures navigational accuracy to within 10 meters) in just two minutes!

Figure 14-5: The atomic clock in the GPS satellite and the one in the receiver on the ground will get out of sync in two minutes because of relativistic effects.

To prevent that problem, the GPS system was designed with a relativistic correction built in. The ticking of the satellite clocks was slowed down before launch to compensate for the gravitational stretching of time once in orbit. And the receiver computers have software that incorporates relativistic formulas. Relativistic errors cancel out in newer GPS receivers, which use the wide-area augmentation system. (See the sidebar “Enhanced GPS systems.”)

These relativistic corrections came out of Einstein’s theories. The 30 million GPS users getting their accurate locations are constantly testing the accuracy of Einstein’s predictions. The GPS system works.

Enhanced GPS systems

The wide-area augmentation system (WAAS), designed by the U.S. Federal Aviation Administration, has improved the accuracy of GPS signals to 1 to 2 meters horizontally and 2 to 3 meters vertically throughout most of the United States. The system uses a network of 25 ground stations that checks GPS readings with map coordinates and issues corrections. These corrections are uplinked to geostationary relay satellites, which broadcast to WAAS receivers. With this method, relativistic errors cancel out.

Measuring the Curvature of Spacetime

In 1959, Leonard Schiff, the chairman of the physics department at Stanford University, was at his regular lunch hour break, swimming 400 yards at the Stanford pool, when two colleagues joined him to talk about gyroscopes. (Gyroscopes are based on a principle of physics that says that an object rotating freely keeps its orientation for as long as it’s free of external influences.)

Schiff was a nuclear physicist and had done work in quantum mechanics but had recently become interested in relativity. He’d been thinking about using very precise gyroscopes to test the theory. One of his colleagues, William Fairbank, was a low-temperature expert and had also been thinking about gyroscopes, but he wanted to do experiments with them at low temperatures. The third professor, Robert Cannon, was an expert on gyroscopes and had worked with them in guidance systems for submarines, planes, and missiles before joining the Stanford faculty.

After their swim, Schiff told his two colleagues about an idea he’d had about using gyroscopes to observe the curvature of spacetime and the way the Earth drags spacetime with it as it rotates. According to Einstein, the gyroscope remains fixed in spacetime. If spacetime is curved by the Earth, the gyroscope should “see” that. If spacetime is dragged by the Earth, the gyroscope should be dragged along.

Spacetime drag: The universe in a bucket

Schiff’s idea was new, but the effects had been known for a long time. Just two years after Einstein developed his general theory of relativity, two German physicists, Josef Lense and Hans Thirring, used Einstein’s theory to show that a large spinning object would not only curve spacetime but also drag spacetime along with it. Physicists call this effect frame dragging.

One way to visualize frame dragging is to take a bucket of white paint and pour a small amount of a dark-colored paint in it without mixing it. Spin a small paint brush in the paint, and notice how the paint tends to turn behind it, creating a whirlpool (see Figure 14-6). The spinning brush swirls the paint around it.

The frame dragging effect explains why rotation is relative. Consider another bucket, this time filled with water and hanging by a rope. If you start spinning the bucket, what do you see? Initially, the water slowly begins to spin, trying to catch up to the bucket. Soon, when the water is spinning along with the bucket, the surface of the water becomes curved, concave (see Figure 14-7). That’s because the water molecules, once set in motion, want to keep that motion. That’s what Newton told us with his first law of motion (see Chapter 4). Some of the molecules end up bunching up against the sides of the bucket. Because there isn’t room for every molecule at the sides of the bucket, some bunch up right behind, and the rest are farther back, giving the surface of the water that concave look. The effect seems pretty straightforward nowadays.

Figure 14-6: The spinning brush drags the paint along with it.

Figure 14-7: The surface of the water in the spinning bucket takes a concave shape.

Back in 1689, Newton didn’t think it was straightforward. Newton wanted to get to the bottom of this phenomenon. With respect to what was the water spinning? Not the bucket, because the bucket is spinning along with the water. The rest of the universe? If that is the case, then imagine that you do the experiment in a completely empty universe: a universe that consists of only a bucket with water. What happens now? Does the water take the same concave shape or stay flat? With respect to what is the water spinning? In an empty universe, there’s nothing to refer the motion to.

Newton believed that even in an empty universe, you still have space as a reference. The water spins relative to space.

Newton proposed a variation of his thought experiment. Suppose that you go back to the initial experiment with the bucket hanging from the rope. Now, instead of spinning the bucket, you, the room you’re in, the entire building, and everything else in the universe rotates around the bucket. All the solar systems, galaxies, clusters . . . the entire universe spins around the bucket. What happens to the surface of the water then? Does it take a concave shape, or does it stay flat?

Newton couldn’t answer this question.

What did Einstein say about these two thought experiments? In relativity, there is no absolute space. That idea is at the core of special relativity (see Chapter 10). Newton’s view that in an empty universe the water spins relative to absolute space is wrong. But spacetime is not relative. For Einstein, space is relative and time is relative, but spacetime is absolute. According to Einstein,

The water in the bucket in an otherwise empty universe spins relative to spacetime.

What about the second thought experiment, the one Newton didn’t have an answer to? Did Einstein have an answer? In principle, he did. According to general relativity, it shouldn’t matter whether you say that the bucket spins and the universe doesn’t or that the universe spins and the bucket doesn’t. The water takes a concave shape in either case. Rotation is relative.

Imagine that you place your bucket inside a large and massive empty shell. Suppose that the shell is actually the size of the Earth and contains all of its mass. Inside this shell, there is no gravity. The shell attracts the bucket from all directions, and all these attractions cancel out. This is true even when you place the bucket away from the center of the shell (see Figure 14-8). The chunk of shell closest to the bucket would attract it with a greater force than the chunk on the other side, except that the bucket “sees” more of the shell that’s farther away. This effect is true in Newtonian mechanics and in general relativity.

Figure 14-8: A bucket inside a shell feels no gravitational force. All the pieces of the shell pull in all directions with an equal force and cancel out.

Now, imagine starting this hollowed Earth spinning. According to Newton, nothing should happen to the water because the shell doesn’t exert a force on the bucket. According to Einstein, the rotating shell drags spacetime along with it. The frame dragging forces cause the water to slightly move over to the sides and take a curved shape. If you enlarge the shell more and more while at the same time increasing its mass, the spacetime effect should increase. If the shell becomes the size of the universe and contains all the mass of the universe, the frame dragging increases in such a way that the curvature of the water matches the shape of the water when the bucket itself rotates.

Proving this idea mathematically wasn’t easy. In 1912, even before completing his general theory, Einstein started some calculations of this frame dragging effect in a shell. In 1965, two theoretical physicists were able to partially solve the problem. Finally, in 1985, Herbert Pfister and K. Braun in Germany were able to complete the calculation, showing that, in fact, Einstein was right in his initial insight. Space inside a spinning shell is dragged along.

Can this effect be checked experimentally? That’s what Schiff and his swimming buddies wanted to know.

Embarking on the mission

Schiff and his two colleagues pitched to NASA the idea of using a gyroscope to measure the curvature of spacetime and to detect frame dragging. As it turned out, NASA had started plans for an orbiting observatory and was interested in the research. Four decades of planning, development, and deployment started right then.

NASA called the mission Gravity Probe B. The Gravity Probe A mission in 1976 had flown an atomic clock in a rocket to measure the relativity of time predicted by Einstein. The new mission was going to measure two other predictions of Einstein: the curvature of spacetime and the frame dragging effect.

Gravity Probe B was launched atop a Delta 2 rocket on April 20, 2004, from Vandenberg Air Force Base in California. The satellite was placed into orbit 640 km (400 mi) above the Earth carrying some of the most advanced technology ever built. Unfortunately, Leonard Schiff wasn’t around to see the realization of his dream. He passed away in 1971 at the age of 55. The mission is now led by Francis Everitt of Stanford University.

The centerpieces of the mission are the four gyroscopes, each a 4-cm diameter sphere of fused quartz spinning at 10,000 rpm. They are the most perfectly spherical objects ever made, polished to within a few atomic layers (see Figure 14-9). The fused quartz used to make the spheres was refined into some of the purest materials in the world.

Figure 14-9: Fused quartz gyroscopes will measure the general relativity predictions.

Courtesy NASA

The gyroscopes needed to be this perfect because the mission will measure the two relativistic effects to an incredible accuracy. If the spheres were not almost perfectly round, the Earth would pull on each of them with a slightly different force, causing the gyroscope to spin and ruining the precision of the measurement. Each sphere also needed to be made up of exactly the same stuff throughout, so that the Earth pulls on all its parts with equal force.

Schiff had calculated that at a 640-km altitude, the curvature of spacetime would turn the gyroscope 6.6 seconds of arc per year and frame dragging would turn it 42 milliseconds of arc per year. A second of arc is a very small angle. If you look at a round clock, each minute mark is 6 degrees apart. One degree has 60 minutes, 3,600 seconds, or 3,600,000 milliseconds of arc. So the minute mark on your clock is separated by 216 million milliseconds of arc (see Figure 14-10). Gravity Probe B has to measure the 42 milliseconds of arc with half a millisecond of arc precision. That’s like looking at the edge of a sheet of paper from 100 miles away! You can appreciate the reason for the ultra precision in the manufacturing of the gyroscopes.

Figure 14-10: There are 216 million milliseconds of arc in the minute mark of a round clock.

If the gyroscopes are almost perfect spheres, how do you know which way they are moving? The whole purpose of the mission is to measure tiny deviations in the orientation of the spinning balls. To be able to do this, the spheres were coated with a layer of niobium, an element that becomes a superconductor when cooled to very low temperatures, close to absolute zero. In a superconductor, there is no resistance to the flow of current. As a result, it generates a magnetic field with the north and south poles lined up with the axis of rotation of the ball. Measuring the magnetic fields gives the scientists the direction in which the balls are spinning.

Because the niobium needed to become a superconductor by lowering the temperature, 650 gallons of superfluid helium went on the mission. That amount will keep the gyroscopes cooled for the 16-month estimated duration of the mission.

Many other technological hurdles had to be overcome to get to this point. About 100 doctoral dissertations were written on the new technologies during the nearly 40 years that it took to develop the mission. If the experiment is successful, it will be the most precise test of general relativity ever conducted.

So, Was He Right?

A scientific theory can never be proven correct. It can, however, be disproved. “There could be no fairer destiny for any . . . theory,” wrote Einstein, “than that it should point the way to a more comprehensive theory in which it lives on, as a limiting case.”

If scientific theories can’t be proven correct, how do scientists know if they are on the right track? The technique is simple and powerful. They use the theory to make predictions of the outcome of experiments. If the predictions turn out to be correct, the theory gains strength. The more confirmed predictions, the stronger the theory.

You can see why theories can never be proven correct. There is always the possibility that someone will make that one experiment or observation that will disprove one or more of the predictions of the theory. If Gravity Probe B determines that there is no frame drag, that space isn’t dragged along as the Earth rotates, then that particular prediction of general relativity is wrong. But it won’t be just that prediction. The entire theory will be shown to be incorrect. Or, most likely, incomplete.

On the other hand, if Gravity Probe B measures a frame drag with the values predicted by general relativity, all you can say is that the theory is very strong, that it has passed all the tests that have been designed for almost a century. But you can’t say that all future tests will be equally successful.