Re: The speed of light

Ron Kean (
Mon, 19 Jul 1999 19:02:02 -0400

On Mon, 19 Jul 1999 16:58:18 -0400 "John Clark" <> writes:

> Ron Kean <> Wrote:
> >According to the CRC Handbook of Chemistry and Physics, the
>second is the> >duration of 9,192,631,770 periods of the radiation
corresponding >to the> >transition between the two hyperfine levels of the ground state >of the> >atom of Cesium 133.

>Ok, but how long is a meter? If you define it as 1/299782458

that should be 1/299792458

of the >distance>light moves in one second that means that as your experimental
>procedures>improve and you know the exact speed of light better the
length of a
>meter changes.

Yes, that sounds correct. However I did not make the rules. The definition does not originate with me. The definition comes from the CRC Handbook, and ultimately by international scientific agreement, probably under the BIPM. It seems equally true that as measurement technology improves, the length of the meter, as measured, would also change under any of the previous definitions of the meter.

>You can't define everything as you wish and remain consistent.

Please don't blame me; I did not make the definition. My guess is that by the mid 1980s the speed of light had been measured to an accuracy as fine as the precision of the Krypton 86 definition of the meter, so it was decided to make the second and the speed of light the primary standards for defining the meter, rather than using the second and the meter as primary standards, as under the older system. I don't see any inconsistency under either system. Under the old system the speed of light 'changes' as measurement become more refined. Under the new system the meter 'changes' as measurement become more refined. I suspect the sentiment was that it is better to have something fundamental such as the speed of light remain constant by definition, rather than have something artificial (the meter) be arbitrarily held constant which results in a fundamental constant (the speed of light) changing. And since under the old system the meter was defined in a way that required a physical measurement to reproduce, the meter would change anyway as the measurement technology changed. So the new system would seem to be an improvement.

In the >last century>the Indiana state legislature tried to pass a law decreeing that the >exact value of PI>was 4. It missed becoming law by one vote, but probably wouldn't have >worked anyway.
> John K Clark

The version of the story I heard was that Kentucky or Tennessee had considered defining pi as 3. Legal definitions of units are usually made to provide a basis for settling commercial disputes, and it does not really matter to physics that those legal definitions are somewhat inaccurate, nor does it matter much to commerce, since for commerce it is more important to have an agreed and easily implemented definition than it is for the definition to be highly accurate.

Back in the 1790s the US by law defined the gallon as 231 cubic inches. Since then, the length of the inch has changed several times. So every time that happened the gallon changed. But it caused no commercial problem because the changes were minuscule. The Imperial (British, Canadian) gallon was defined in a very different way. That gallon was the volume taken by 10 pounds of water at 62 degrees F.

Likewise, defining pi as 3 may have been proposed for some commercial uses, not for scientific use. 22/7 is a good approximation of pi, but at the time lawmakers may have thought that fractions are too complicated for many people (including, perhaps, them) and that a 'simple' definition is better. If a commodity such as rope is being sold in a coil with a known number of turns, its length can be estimated by counting the turns, measuring the diameter of the coil, and multiplying by pi.

Ron Kean



Get the Internet just the way you want it. Free software, free e-mail, and free Internet access for a month! Try Juno Web: