Maxwell showed that all waves which come up to and are absorbed by a surface exert a pressure on every square centimetre of the surface equal to the amount of energy contained in one cubic centimetre of the beam. If the surface is a perfect reflector, the reflected waves produce an equal back pressure, and therefore the pressure is doubled. As the waves are reflected back along their original direction, the energy in the beam will also be doubled, and so {65} the pressure will still be equal to the energy per cubic centimetre of the beam. As the energy which is received in one second from the sun on any area can be measured by measuring the heat absorbed, and since the speed of light is known, we can calculate the energy contained in one cubic centimetre of full sunlight, and hence the pressure on one square centimetre of surface. For the energy received on one square centimetre of surface in one second must have been spread originally over a length of beam equal to the distance which the light has travelled in one second, _i.e._ over a length equal to the speed of light. If we divide that energy, therefore, by the speed of light, we shall get the energy in a one-centimetre length of the beam, and therefore in one cubic centimetre. This turns out to be an extremely small pressure indeed, being only a little more than the weight of half a milligram, on a square metre of surface. Maxwell suggested that a much greater energy of radiation might be obtained by means of the concentrated rays of an electric lamp. Such rays falling on a thin, metallic disc delicately suspended in a vacuum might perhaps produce an observable mechanical effect. Nearly thirty years after Maxwell's suggestion it was successfully carried out by Prof. Lebedew of Moscow, who used precisely the arrangement which Maxwell had suggested. +Measurement of the Pressure.+--A beam of light from an arc lamp was concentrated on to a disc suspended very delicately in an exhausted glass {66} globe about 8 inches across. Actually four discs were suspended, as in Fig. 24, and arrangements were made to concentrate the beam on to either side of any of the four discs. [Illustration: FIG. 24.] The suspension was a very fine quartz fibre _q_. The discs _d_, _d_, _d_, _d_, were half a centimetre in diameter and were fixed on two light arms, so that their centres were one centimetre from the glass rod, _g_, which carried them. A mirror, _m_, served to measure the angle through which the whole system was twisted owing to the pressure of the beam on one of the discs. In order to measure the angle a telescope viewed the reflection of a scale in _m_, and as _m_ turned different divisions of the scale came into view. The two discs on the left were polished and therefore the pressure on them should be about twice that on the blackened discs on the right.