When you watch a thunder storm it is obvious that sound travels more slowly than light, since the rumble of thunder arrives later than the lightning flash.

There are lots of ways you can measure the speed of sound. A few are rather bizarre. One investigator inParis used a kilometre of empty water pipe to send sounds and measured the time taken. He also fired a pistol at one end of the pipe and found that it extinguished a candle at the other end. And an attempt was made on a Swiss lake to measure the speed of sound in water. A large bell was suspended below a boat and the ringing sound was detected on the other side of the lake.

A less ambitious method of measuring the speed of sound involves using the time taken for an echo to return from a wall. Try using a sports whistle, metal dustbin lid and hammer, or even a starting pistol to produce a loud sound.

Assemble the children about 200 metres from a large wall. Distribute stopwatches and time the sound as it goes to the wall until the echo returns. Check that the children realise that the sound has travelled 400 metres (out and then back from the wall). Calculate the speed of sound in air from the data. Any errors in timing will have a significant impact on the reliability of the result, since we are dealing with a time interval of just a second or two. The distance to the wall can be measured quite reliably using a tape measure.

The accepted value for the speed of sound in air is 330 metressecond. In water it is much faster, about 1500ms.

You can reproduce the effect of the pistol shot on the candle in a less hazardous way. Use a piece of plastic drainpipe about two metres long, supported horizontally at the height of a candle flame. Attach a card cone to one end so that it narrows the pipe next to the flame itself. Leave a 2cm hole at the pointed end of the cone.

Try clapping your hands or banging two books together at the open end. The flame will respond to the sound and will probably go out.