Of all the founders of the theory of light undoubtedly the greatest were Newton and Huygens. Newton discovered the theory of the spectrum. By passing a fine beam of sunlight through a glass prism he resolved it into its component colours :—red, orange, yellow, green, blue, indigo, violet. It had already been known that white light was resolved into colours on passing through glass or water (for example the rainbow had been ex plained), but it was supposed that the glass had produced a definite alteration in the light. Newton showed that, if the light passed through a second prism reversed, the coloured lights would recombine into white, but that if a single colour were selected from the spectrum, no subsequent treatment could change it in any way. He was thus led to the correct explanation of white light as a compound of all the colours, but the further details of this question would lead us into physiological optics (see VISION).
Newton also investigated the colours of thin plates such as soap bubbles. He placed a slightly convex lens on a flat piece of glass and observed the light reflected. At the centre the surfaces are in contact and nothing is reflected, but round this point, where they are very close, though not in contact, there appears a suc cession of brilliantly coloured rings, the succession of colours being black, faint blue, strong white, orange, red, dark purple, violet, blue, faint green, vivid yellow, etc. The rings become narrower and fainter as we go outwards and soon become invisible. These are not the colours of the spectrum, and their origin is better appreciated by considering the case where the illumination is by monochromatic instead of white light. There is then a dark centre surrounded by a large number of rings of the colour used. (See Plate, fig. I.) The sizes of the rings depend on the colour of the light; they are closer together for blue than for red. If then we illuminate simultaneously with red and blue light there will be places where the rings fall together, so that we get alter nately purple and dark, and other places where they fall apart and we get alternate rings of red and blue. Plate. fig. 2 is a photograph of this type ; as the camera is unequally affected by the two colours, in some regions the rings look blurred, in others sharp. The rings in white light can be explained in the same way as due to a superposition of all the colours, and at no great distance from the centre so many of the coloured rings overlap that they become invisible.
Newton was very cautious in making theories and he did not really succeed in explaining his rings, but he attributed them to certain "fits" of reflection and transmission which later were seen to be very like the phases of the wave theory. As to the general theory of light he was also very non-committal, but he criticized the wave theory as being unable to explain rectilinear propagation and his followers, interpreting his views in a much narrower way than he intended, adopted a complete corpuscular theory of light which held the field for more than a century. This theory supposed that light consists of minute particles, or cor puscles, shot out from the luminous body, and attempted to explain all the phenomena by suitably modifying the properties of these corpuscles. The theory implies that the velocity of light
must be proportional to the refractive index, not inversely pro portional as in Fermat's principle and the wave theory, and this was later to provide a critical test which condemned it.
Huygens is the real founder of the wave theory of light. He based his belief in it primarily on the fact that, if a beam of light were like a flight of arrows, then when two beams cross these should be collisions be tween the arrows. He succeeded in ex plaining reflection and refraction, and we may consider his construction, as it lies at the foundation of modern methods. The general idea is that light is a disturbance in a medium, but it need not be specified what is the character of the disturbance ; for purposes of rough visualisation we may think of the medium as a jelly which is distorted so that its particles move out of their usual places. Any disturbance then acts as a centre causing the propagation of a wave of disturb ance to go out at a constant speed, so that at any subsequent time the effects of the initial disturbance will be found on a sphere surrounding it. When the initial dis turbance is not confined to a single point, each point of it is to be regarded as a source, and the subsequent disturbance is the geo metrical envelope of the spheres surrounding all these sources. Refraction is explained by supposing that the velocity of light is different in different media. Consider light obliquely incident on a flat surface, say of water (see fig. 1). The velocity of propa gation outside is C the velocity of light; in the water it is v a slower velocity. The advancing disturbance is at one moment spread as a pulse over the surface AB. Each point of AB gives out a spherical pulse and, to reconstruct the wave later, we draw spheres of equal radii round all the points of AB. Obviously one such set of spheres will give the line CD as their envelope, and this shows that the light goes in the direction AC outside the water. But if we repeat the construction starting at CD we have to allow for the fact that the velocity is less in water than in free space. Thus, corresponding to the sphere of radius DF about D, we draw round C a sphere of radius CE=v/cDF, and it is evident that EF will be bent round more nearly parallel to the face than was CD. After this both spheres are in the water and the propa gation goes straight again on to GH. This construction immediately gives Snell's law of refraction, for sin A CM/sin GCN = DF/CE = c/v, and the refractive index is simply the ratio of velocities. The con struction fails in the case of very oblique incidence if v is greater than c, for then the circle round C may have radius actu ally greater than CF itself. Refraction is then impossible and all the light is re flected. This is the phenomenon of total internal reflection which we shall discuss later. A simpler construction than that we have given applies for reflection, and the same principle also explains diffraction, but Huygens did not find this out.