illuminated by a red, the other by a green, and the third by a blue light, the images from these three transparencies, when visually combined, will show the colours of the object. This plan was scientifically and practically worked out by F. E. Ives of Philadelphia, though in France and elsewhere it had been formulated, especially by Hauron Du Cros.
The following description may be taken as that of Ives's process by the trichromatic theory of colour-vision every colour in nature can be accounted for by the mixture of two or three of the three-colour sensations, red, green and blue, to which the eye is supposed to respond. Thus a mixture of a red and green sensation produces the sensation of yellow; of a green and blue, that of a blue-green; of red and blue, that of purple; and of all three, that of white For the sensations we may substitute those colours which most nearly respond to the theoretical sensations without any material loss of purity in the resulting sensation. We must take the spectrum of white light as the only perfect scale of pure colours It has been proved that the red sensation in the eye is excited by a large part of the visible spectrum, but with varying intensities. If, then, we can on a photographic plate produce a developed image of the spectrum which exactly corresponds in opacity and position to the amount of red stimulation excited in those regions, we shall, on illuminating a transparent positive taken from such a negative with a pure red light, have a representation of the spectrum such as would be seen by an eye which was only endow ed with the sensation of red Similarly, if negatives could be taken to fulfil the like conditions for the green and for the blue sensations, we should obtain positives from them which, when illuminated by pure green and blue light respectively, would show the spectrum as seen by an eye which was only endowed with a green or a blue sensation. Evidently if by some artifice we can throw the coloured images of these three positives on a screen, superposing them one over the other in their proper relative positions, the spectrum will be reproduced, for the overlapping colours, by their variation in intensity, will form the colours intermediate between those used for the illumination of the positives For the purpose of producing the three suitable negatives of the spectrum, three light-filters, through which the image has to pass before reaching the photographic plate, have to be found. With all present plates these are compromises Roughly speaking, the screens used for taking the three negatives are an orange, a bluish-green and a blue These transmit those parts of the spectrum which answer to the three sensations. When these are obtained an image of a coloured object can be reproduced in its true colours
Abney devised sensitometers for determining the colours of the screens to be placed before the lens in order to secure the three colour negatives which should answer these requirements. Their production depends upon the same principles indicated as necessary for the correct rendering in monochrome of a coloured object. When the sensitometer takes the form of glasses through which hight is transmitted to the plate, the luminosities of the coloured lights transmitted are determined, and also their percentage composition in terms of the red, green, and blue lights, and thence are deduced the luminosities in terms of red, green and blue. For ascertaining what screen should be used to produce the red negative the luminosity transmitted through each glass is so adjusted that the luminosity of the red components in each is made equal by rotating a disk with correct apertures cut out close to the row of glasses. This gives a sensitometer of equal red values. A coloured screen has to be found which, when placed in front of the lens, will cause the opacities of the deposit on the plate, corresponding to each square of glass, to be the same throughout. This is done by trial, the colour being altered till the proper result is obtained. In a similar way the “green” and “blue” screens are determined. Coloured pigments rotating on a disk can also be employed, as indicated in the paragraph on the correct rendering of colour in monochrome.
As to the camera for the amateur, whose plates are not as a rule large, all of the three negatives should be obtained on one plate, since only in this way can they be developed and the densities increased together. (For commercial work the negatives often cannot be taken on the same plate, as it would make the plate too large to manipulate.) The camera may be of an ordinary type, with a repeating back, bringing successively three different portions of the plate opposite the lens It is convenient to have a slide, in front of which a holder containing the three screens can be fixed, which will then be close to the plate; such a one has been devised by E. Sanger-Shepherd. The light passes through them one by one as the plate is moved into the three positions. The three exposures are given separately, after which the plate is ready for development. The three separate exposures are, however, a source of trouble at times, particularly in the case of landscapes, for the lighting may vary and the sky may have moving clouds, in which case the pictures would show variations which should not exist Sanger-Shepherd has a “one-exposure” camera by which the three images are thrown side by side on the plate. Thus any movement in the picture affects all three negatives alike. Abney has also introduced a “one-exposure” camera which takes in a larger angle than that of Sanger-Shepherd. The next point is the exposures which should be given through each screen. This can be done by placing in front of the plate and extending its whole length a scale of gradation through which the light coming from a sun-illuminated white card passes, as well as through the screens. In the case of the three-exposure camera the times of exposure are varied till the densities of the image of the gradation appear the same in each of the three images In the case of the one-exposure camera, the light reaching the plate through the screens is altered by cutting off with a shutter more or less of the lens used. As the plates employed for the purpose of the three-colour negatives must be sensitive to every colour, the ordinary dark-room ight should be most cautiously used. If used at all, it should be very feeble and development must be carried out in a dish with a cover to it. The plate is manipulated in the usual way.
Joly's Process.—Professor J. Joly, of Dublin, in 1897 introduced a colour process by which an image in approximately natural colours could be thrown upon a screen by an optical lantern, only one transparency being employed, instead of three, as in the Ives process. A “taking” screen was ruled with alternating orange, blue-green and blue lines 1200 to 1300 in apart, touching one another and following one another in the above order. When such a screen was placed in front of a sensitive plate in the camera, and exposure made to the image of a coloured object, there were practically three negatives on the same plate, each being confined to the area occupied by lines of the same colour. The shades of colour and the depth of the colours used in ruling depended on the brand of plate. When a perfect triune negative was obtained, a transparency was made from it, and in contact with this was placed a screen ruled with lines the same distance apart, but of the colours corresponding to the three colour sensations, namely red, green and blue. The red lines were made to fall on the image taken through the orange lines, the green on that of the blue-green, and the blue or violet on that of the blue On the screen there are practically three differently coloured images shown by one transparency. The eye blends the different colours together and a picture is seen in approximately the correct colours of the original.
Autorhrome.—A very remarkable process, founded on J. Joly's process, was introduced in 1907 by A. Lumière et ses Fils of Lyons. Starch grains of very minute size, some of which were dyed with a red stain, a second portion with a green, and a third portion with a blue, are mixed together in such proportions that a fine layer of them appears grey when viewed by transmitted light. Under a magnifying glass the grains are coloured, but owing to the want of focus in the eye the colours blend one with the other. Such a layer is embedded on the surface of a glass plate in a waterproof vehicle, and a film of sensitive emulsion held in situ in some material, the composition of which has not been published, covers this layer. When such a plate is placed in the camera, with the back of the plate next the lens, the light passes through the coloured granules, and again we have three negatives on one plate, but instead of each negative being represented by lines as in the Toly process they are represented by dots of silver deposit Owing to the way in which the three coloured film is prepared, it is evident that a positive taken from such a negative could not be backed with granules of the right colour, as the granules are placed at random in the layer. Lumière, to overcome this difficulty, converted the negative into a positive in a very ingenious way. The plate was developed with pyrogallic and ammonia in the usual way, but instead of fixing it it was plunged into a solution of potassium permanganate and sulphuric acid. This dissolved all the silver that had been deposited during development and left a film of unaltered silver salt. On looking through the plate the colours of the coloured layer coming through the different dots where the silver was at
