Back to Contents
In the essay on Events in Inertial Frames I introduced you to an imaginary clock that comprises a laser and a mirror at opposite ends of a glass tube. Now I want to show you how physicists use such a clock in modern physics. Let's assume that only one company makes these clocks and that they make only one model, so everyone uses identical clocks. If the manufacturer makes the distance between the laser and the mirror in each clock seven feet four inches, then one tick of the clock represents one tenth of a second in our little fantasy world.
For the purpose of conducting the appropriate imaginary experiment let's imagine that we have gone to Goshen Junction, next to the Southern Pacific tracks that run straight from Bakersfield to Fresno. Night has fallen and we can watch the pulse in our clock flittering up and down like a hyper-kinetic firefly. A train goes by and we notice that someone has mounted a clock on the outside of one of the cars. On impulse I pick up my camera and snap a picture of the train as it passes, trusting the camera's mechanism to set the exposure. When I develop the picture, we see that the camera did, indeed, set the exposure in accordance with the light available and made it too long for a good picture: the photo shows only a horizontal blur with a red sawtooth pattern traced over it. That sawtooth pattern, resembling a series of red vees strung together in the manner of paper dolls, came from the light emanating from the clock mounted on the train and it's the only feature in the picture that makes sense, the more so because each vee represents the path that one of the clock's pulses followed in our inertial frame as it counted time in its clock.
Before we trash the photo (it's not the kind of thing that we want to stick into an album, after all), let's see if we can use it to learn something about how moving clocks tell time. The picture is large enough that the vees are several inches high, so we can easily draw two lines on one of them. We draw one straight line vertically through the bottom point of the vee, thereby bisecting the vee, and we draw one straight line horizontally in such a way that it just touches the upper points of the vee. Those two lines convert the vee into a pair of right triangles. Let's choose one of those triangles and list the things that we know immediately about the three lines that comprise it:
Line A; the length of the vertical line, from the bottom of the vee to the horizontal line is simply the length of the train clock's optical axis, the distance between its laser and mirror. By Lorentz Rule 1 it is also equal to the length of our clock's optical axis, because the train clock's optical axis is oriented perpendicular to the direction in which the train is moving;
Line B; that part of the horizontal line that extends from the vertical line to one of the vee's upper points is simply the distance that the train moves in the time it takes the train clock's pulse to go from one end of the clock to the other; that is, the distance that the train moves in one half of its clock's cycle. To calculate that length we must multiply the speed of the train by the time elapsed on our clock during that half cycle of the train's clock;
Line C; the hypotenuse is the line traced by the train clock's pulse as it moves from one end of its clock to the other. By Einstein's second postulate, it must so move at the same speed of light (100 miles per hour in our fantasy world) in both our frame and the train's frame. Thus the length of the line is equal to the product of multiplying the speed of light by the time elapsed on our clock, the same time that we used to calculate the length of Line B.
We also know that for the passengers on the train the pulse in their clock follows a path whose length is equal to the length of Line A. After all, in their frame the clock is not moving, so the pulse must follow a perfectly vertical straight line. They can calculate the length of that line by multiplying the speed of light by the amount of time that, in theory, elapses on their clock during that half cycle, one twentieth of a second in this case, and we can accept that calculation because we make the same calculation in regard to our own clock. But then we notice that Line C is longer than Line A, which means that a pulse in our clock will complete its traverse before a pulse in the train's clock completes its own traverse. That, in turn, means that if we pick two events, such as the train's clock passing the signal at a grade crossing and its crossing the points of a nearby switch, the train's clock will count fewer seconds between those events than we would calculate from the readings of our own clocks set up by the signal and at the switch. The train clock appears to us to be counting seconds that are dilated relative to the seconds that our clocks are counting.
It's not only the laser clock that slows down when it moves. The Principle of Relativity tells us that there is no observation or experiment that the passengers on the train can make that would tell them that they're moving and we're not. Thus everything on the train that could conceivably be used to count time appears to us to slow down in the same proportion in which the laser clock has slowed: the conductor's spring-wound pocket watch, the businessman's wristwatch, which counts the vibrations of an electrically stimulated quartz crystal, the rubber ball that a boy is bouncing off the floor of the coach, even the rates at which the passengers' hearts beat. Indeed, if we could measure the temperature of the train without touching it, we would discover that the coach and everything in it is astonishingly cold: even the random atomic motions that comprise heat have slowed down.
We might be tempted to think that all of the objects mentioned above have somehow been so altered by motion that they have slowed down. That's essentially what Hendrick Lorentz assumed when he worked out time dilation himself. He assumed that the aether blowing through a clock so altered the forces exerted among the clock's atoms that the clock's parts moved more slowly. That interpretation is wrong, as we can see by the fact that if it were true to Reality, then the passengers on our train would have to see our clocks ticking faster than theirs do. The Principle of Relativity flatly contradicts that idea. The passengers on the train will see the pulses in our clocks tracing the same sawtooth pattern that we saw their clock tracing in our frame and they will infer that our clocks are counting dilated time relative to their clocks. From that proposition we must infer that time dilation exists not in the objects but in the structure of the inertial frames in which the objects are embedded.
Having thus deduced the existence of time dilation, we now want to determine the amount of dilation that we can expect to observe on a given moving clock. We can fill that want by using our descriptions of Lines A, B, and C and the fact that they are the sides of a right triangle. Quite a few centuries ago a fellow named Pythagoras figured out a rule about any right triangle: the square of the length of the hypotenuse is equal to the sum of the squares of the lengths of the sides. Let's use that rule in a slightly different form and say that the square of the length of Line C minus the square of the length of Line B is equal to the square of the length of Line A. The lengths of Line C and Line B are calculated from a time measured on our clocks and the length of Line A is calculated from the corresponding time measured on the train's clock, so we now have a mathematical relation between those times. When we apply some straightforward algebraic manipulation to that relationship (and it's not as difficult as you might think, as I demonstrate in Appendix III), we obtain
LORENTZ RULE 2: The time that an observer measures elapsed between two events that are touched by a moving clock is equal to the time interval measured by the moving clock multiplied by the Lorentz factor between the inertial frames occupied by the moving clock and the observer.
The recipe for calculating the Lorentz factor is as follows: take the velocity between the two inertial frames, square it, divide the square by the square of the speed of light, subtract the resulting fraction from the number one, extract the square root of the result, and divide that square root into the number one. The first ten Lorentz factors (L.f.) and the relative speeds that correspond to them are:
|L.f.||Speed as % of lightspeed||Kilometers per second||Miles per second|
The Lorentz factor for something traveling at the speed of light would be infinite.
So far all that I have shown you is deduction and interpretation. But Science, properly understood, does not stand on logic alone. Yes, the reasoning that I have shown you so far is appealing and may seem unassailable, but how can we tell if it went wrong somewhere in some subtle way that eluded our notice? How can we know with any certainty that what we have deduced is, in fact, true to Reality? We must refer back to the empirical Scientific Method. Logic notwithstanding, this is the guild-mark of Natural Philosophy, which sets Science apart from all other belief systems ever conceived by human imagination, that the hypotheses of any theory must be put to the proof against Reality itself; that is, they must be subjected to experimental or observational tests that will either verify or falsify them.
You might think that with our current technology we might have some difficulty testing an effect that only shows itself in objects moving close to the speed of light. But in fact, David H. Frisch and James H. Smith carried out an experimental demonstration of time dilation in 1962. Because there was simply no way in which Frisch and Smith could have accelerated anything like a clock to any speed close to the speed of light, they chose to observe a naturally occurring phenomenon that can show an elapse of time if read correctly. They chose to observe the decay rates of cosmic-ray fragments.
Frisch and Smith chose to exploit the knowledge that had been gained concerning the flux of particles carrying extremely high energies into Earth's atmosphere from somewhere beyond the solar system. When one such particle enters the atmosphere it inevitably strikes an atom and shatters its nucleus, producing a jet of fragments, which goes on the shatter more atoms, producing yet another shower of fragments until the energy of the original particle has been used up. Because the original particle travels exceptionally close to the speed of light (astrophysicists have inferred Lorentz factors from the millions to the hundreds of billions from observations), the jet of fragments comes out of the collision in a tight beam. The fragments, produced by the shattering of the protons and neutrons that comprise the atomic nucleus, are weird little whizbangers called pi mesons (or, more compactly, pions). After flying some distance, they pop apart into mu antineutrinos and mu mesons (or, more compactly, muons). Each of those muish whizbangers then pops apart into a mu neutrino, an electron neutrino, and a positron (or, rarely, a mu neutrino, an electron antineutrino, and an electron).
Those pop-aparts, technically called decays, don't adhere to a strict schedule. Once a muon has been created, it is not doomed to decay at a particular time; rather, in any given interval after the muon's creation the muon has a certain probability of decaying. Such decays, as well as the decays of radioactive nuclei, are subject to the Law of the Half-Life, the law which requires that the proportion of a given substance that decays is the same in any given interval. From experiments performed with cyclotrons physicists had, by 1962, determined that the half-life of muons is 1.523 millionths of a second. By the Law of the Half-Life, then, if you suddenly create 1054 muons, then 1.523 microseconds later you will have 512 muons, give or take a few; after another 1.523 microseconds elapse you will have 256 muons, give or take a few; 1.523 microseconds later you will have 128 muons, give or take a few; and so on. You can also carry out that calculation backward: if you know how many muons were created and then count the number that you have at some later instant, you can then calculate how much time elapsed in the muons' inertial frame. That's what Frisch and Smith did.
A transparent block of doped polystyrene eleven inches thick was the window through which Frisch and Smith hoped to view relativistic muons. The dopant was so chosen that the block would emit flashes of light wherever and whenever a muon decayed. Thus, when a muon-laden spray from a cosmic ray passed through the block, the light stimulated by the spray triggered an oscilloscope, which then recorded the flashes from the muons that had stopped in the block and decayed.
For their first data-collecting runs, Frisch and Smith set up their apparatus on Mount Washington in New Hampshire. There, 6265 feet above sea level, they built a barrier of iron two-and-a-half feet thick over their plastic block and took their data over six intervals of one hour each. Because of collisions with the atoms in the iron, all muons flying slower than 99.50 percent of lightspeed would be stopped in the barrier. All muons flying faster than 99.54 percent of lightspeed would pass through both the iron and the polystyrene, so all of the muons with speeds between 99.50 and 99.54 percent of lightspeed would be brought to a stop within the plastic block, where they would decay, induce the emission of light, and be counted. With that setup Frisch and Smith detected an average of 563 decays per hour.
What that means is that of muons with speeds in that range, at 6265 feet above sea level, we can expect about 563 per hour to pass through any horizontal surface whose area is equal to the area of the top of Frisch and Smith's plastic block. For the purpose of applying the Law of the Half-Life we can interpret that number as the number of muons with the desired speeds being created at that altitude and sent downward. The flight to sea level takes 6.3918 microseconds, which is 4.197 muon half-lives. If time in the muons' frame is not dilated relative to time elapsed in our frame, then that flux of muons would be reduced from 563 per hour on Mount Washington (or at the same altitude above our sea level site) to about 31 per hour at sea level. On the other hand, if the muons' time is dilated, then the Lorentz factor (for a relative speed of 99.52 percent of lightspeed) is 10.22, which dilates the muon half-life to 15.565 microseconds: the flux of the selected muons would then be reduced at sea level to 423, give or take a few, per hour.
For their second set of six one-hour data-collecting runs, Frisch and Smith rebuilt their detector on the campus of the Massachusetts Institute of Technology, in Cambridge, which is ten feet above sea level. There they built their iron barrier with one foot less thickness to compensate the shielding effect of the extra 6255 feet of atmosphere between their plastic block and the source of the muon-laden sprays that they were observing. When they analyzed the data that they gathered, they found that they had been detecting muons at an average rate of 408 per hour.
Given the relative
crudity of the experiment, that's excellent agreement between the fruit of
observation and the expectation of theory. And the wide difference between the
two theoretical possibilities, both in actual numbers of muons decaying at sea
level and in those decay rates as a proportion of the original decay rates on
Mount Washington, leaves little room for doubt: muons generated in cosmic-ray
showers choreograph their decays in dilated time.
Back to Contents