A short history of brief time

Split seconds seem to matter a lot these days. To Linford Christie - ejected from the Olympic 100m final for starting 14 thousandths of a second too soon - waiting for a few more would have been very useful.

But what use are split seconds to the rest of us? You might think there's not much real use for them, even though we worked out how to split the second long before we could do the same to the atom, and only a little after grammarians managed the same with the infinitive. The first watch was built in 1502, but only had one hand; timepieces able to accurately measure seconds only appeared after the Industrial Revolution. Nowadays, though, you only have to look around a bit to find that our daily lives are strewn with the twitching remnants of time's basic unit.

For example, in a second, any TV screen is refreshed 25 times, a film shows 24 frames, ABS brakes apply and release force hundreds of times, and a Tory backbencher (or a Labour frontbencher) can weigh the arguments for and against awarding himself a 26 per cent pay rise while condemning the Tube drivers for striking.

But you really want to get small? People boasting of their computer's prowess will tell you that its central chip runs at so many megahertz. Each of those is a millionth of a second: so a computer chip running at 100 MHz does something (probably not a very useful something, but there you are) every 10 billionths of a second. Watches keep time to precisions of a second or so in years; digital versions often offer stopwatches that count hundredths of a second - so useful for proud parents on school sports day.

Yet while athletics has gradually been adding new levels of refinement to its timing - replacing men in macs braced by the trackside with stopwatches with sophisticated photo-finish systems capable of splitting thousandths of a second, scientists have been delving even deeper into the second. In 1967 this was defined as the time corresponding to 1,192,631,770 vibrations of a caesium atom; Linford should have hung around another 16,696,845 vibrations or so. Presently scientists are exploring events that take "femtoseconds" (million millionths of a second), such as various atomic reactions.

So, as the rest of the world follows the precision that science introduces, which will be the first Olympics to introduce timing accurate to millionths of a second? Surely our successors will curse us for not having bothered to measure those vital little bits of time - the ones that, to them, will mean all the difference between a gold medal and fourth place. And as for false starts: if Linford Christie felt hard done by, imagine how an Olympic competitor in 2056 will feel on being disqualified for starting .000001 of a second too soon.