12 SEPTEMBER 1908, Page 15


Sr,—In the interesting article on colour photography in the Spectator of last week the writer seems to have left out of account one fact which, I think, destroys his argument that true colour photography has arrived in the Lumiere process. He says that "the fact must be faced that colour photography is mechanical and its processes automatic." After describing the potato-starch method, he states that "colour reproduced in this way must preserve exactly the qualities and intensities of the local colours in the object photographed. There would be no trusting it if it did not." This assumes that a result produced by the new procedure is an accurate reproduction of the colour of Nature, like the reflection in a looking-glass. But we are told that the whole process depends on the grains of the starch- powder being "stained respectively green, violet, and orange." So here we have the old fallacy of the three-colour process that coloured inks and stains are the same thing as coloured light. There is no such thing as absolute colour in a dye. The pigment used in colouring the starch must always be liable to variation, and it remains an artificially applied paint, and not naturally coloured light reflected from the object photographed. It is all one whether the dye is distributed over the plate in millions of grains by the Lumiere Brothers before the photograph is taken, or whether it is applied in three different printings from photographically produced blocks. In either case, arbitrarily coloured substances—in fact, paints —are applied to the image, and it does not matter whether this is done before or after the sensitive plate is exposed. In both ways the results seem equally hideous. Plus fa change,

plus c'est la ?nem chose.—I am, Sir, &c., H. S.