In this way, not only have the very small particles of colouring matter in glass been seen individually, but it is thought that the actual molecules of matter have been seen, or if not the molecules individually, little groups of molecules, dancing and capering about, just as scientific people for years have believed them to be doing, although they could not see them. So here we have an instance in which manufacture has aided science--an inversion of the usual order of things. CHAPTER XVI COLOUR PHOTOGRAPHY Photography has introduced many of the general public to a branch of practical science which otherwise they would never have cared much about. The action of light upon certain chemicals, the subsequent action upon the same of other chemicals, such as developers, toning solutions and so on, form a very well-known region of the domain of science. And this is, too, a branch of chemistry in which the practical inventor has been very busy. The efforts, therefore, which have been made to invent ways of producing photographic pictures which shall give to the objects their natural colours, will probably be of special interest in a book like this. Of these there are two very well-known systems, and to them we will mainly confine our attention. It should first be pointed out, however, that what we are discussing is quite different from the simple "orthochromatic" plates which are used by many photographers. These latter are coated somewhat differently from other plates, with a view to their giving a more realistic picture, but the result is still in one colour. They are, in fact, a little more sensitive to differences in colour than ordinary plates, so that colours which appear, when the latter are used, very much the same, appear, when orthochromatic plates are employed, a little different. But the difference in colour in the object photographed is only, even then, represented by a difference in shade in the picture. The object is, it may be, in many colours, in all the colours, very likely, but the picture is only in one. And the step from that to a coloured picture is a very long one. True, the solution of the problem is very simple in principle, yet the practical difficulties are so great that even now they have not been entirely overcome.