DIFFRACTION SPECTRA.
No solid advance was made in accurate measurement of the wave-lengths of light as long as prisms alone had to be used for producing spectra. The fact that the light must penetrate the glass makes the results dependent on the material. No tice the simplicity of the laws of reflection compared with those of refraction. For instance, the principal focal length of a mirror can be determined beforehand by simply having a knowledge of its geo metrical dimensions. There is no such absolute certainty about preparing a lens of given focal length, or determining the deviation given by a prism whose dimen sions are known. The first step towards the present methods employed in the measurement of wave-lengths was made when Newton noticed that light in passing by the edge of opaque bodies gave irises of colour, as he termed them, inside the edge of the shadows. These results can only be explained on the theory that light has something of the nature of a wave motion. As a matter of fact, light, heat
radiation, and the electro-magnetic waves of wireless telegraphy obey the same laws throughout, the only difference being in their wave-lengths. To get a concrete idea of what happens when light passes the edge of an opaque body, imagine a series of water waves impinging at right angles against a breakwater. Behind the breakwater, of course, no waves would be seen, but at the end of the wall the disturbed water would not be separated from the still portion by a straight line drawn in the direction of motion of the advancing waves. Secondary wavelets would bend inwards from the end of the wall, still travelling in straight lines, and the smaller the length of the waves the more bending there would he. This secondary effect is called diffraction, and must not be confused with the refraction of the whole of a ray when passing through a prism.